Arxiv今日论文 | 2025-02-25

本篇博文主要内容为 2025-02-25 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR五个大方向区分，若需要邮件定时接收，请在评论区留下你的邮箱号。

说明：每日论文数据从Arxiv.org获取，每天早上12:00左右定时自动更新。

友情提示: 如何您需要邮箱接收每日论文数据，请在评论处留下你的邮箱。

目录

• 概览
• 自然语言处理CL
• 人工智能AI
• 机器学习LG
• 计算机视觉CV
• 信息检索IR

概览 (2025-02-25)

今日共更新900篇论文,其中:

• 自然语言处理共194篇(Computation and Language (cs.CL))
• 人工智能共310篇(Artificial Intelligence (cs.AI))
• 计算机视觉共166篇(Computer Vision and Pattern Recognition (cs.CV))
• 机器学习共306篇(Machine Learning (cs.LG))

自然语言处理

[NLP-0] Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLM s

链接: https://arxiv.org/abs/2502.17424
作者: Jan Betley,Daniel Tan,Niels Warncke,Anna Sztyber-Betley,Xuchan Bao,Martín Soto,Nathan Labenz,Owain Evans
机构: 未知
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 10 pages, 9 figures

点击查看摘要

[NLP-1] MLLM s Know Where to Look: Training-free Perception of Small Visual Details with Multimodal LLM s ICLR2025

链接: https://arxiv.org/abs/2502.17422
作者: Jiarui Zhang,Mahyar Khayatkhoei,Prateek Chhikara,Filip Ilievski
机构: University of Southern California(南加州大学); Vrije Universiteit Amsterdam(阿姆斯特丹自由大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Published as a conference paper at ICLR 2025. Code at: this https URL

点击查看摘要

[NLP-2] LongSpec: Long-Context Speculative Decoding with Efficient Drafting and Verification

链接: https://arxiv.org/abs/2502.17421
作者: Penghui Yang,Cunxiao Du,Fengzhuo Zhang,Haonan Wang,Tianyu Pang,Chao Du,Bo An
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-3] he Geometry of Refusal in Large Language Models : Concept Cones and Representational Independence

链接: https://arxiv.org/abs/2502.17420
作者: Tom Wollschläger,Jannes Elstner,Simon Geisler,Vincent Cohen-Addad,Stephan Günnemann,Johannes Gasteiger
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-4] Reasoning with Latent Thoughts: On the Power of Looped Transformers ICLR2025

链接: https://arxiv.org/abs/2502.17416
作者: Nikunj Saunshi,Nishanth Dikkala,Zhiyuan Li,Sanjiv Kumar,Sashank J. Reddi
机构: Google Research(谷歌研究); Toyota Technological Institute at Chicago(丰田技术研究院芝加哥分院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: ICLR 2025

点击查看摘要

[NLP-5] Linguistic Generalizability of Test-Time Scaling in Mathematical Reasoning

链接: https://arxiv.org/abs/2502.17407
作者: Guijin Son,Jiwoo Hong,Hyunwoo Ko,James Thorne
机构: Yonsei University(延世大学); OneLineAI; KAIST AI(韩国科学技术院AI)
类目: Computation and Language (cs.CL)
备注: work in progress

点击查看摘要

[NLP-6] Large Language Models are Powerful EHR Encoders

链接: https://arxiv.org/abs/2502.17403
作者: Stefan Hegselmann,Georg von Arnim,Tillmann Rheude,Noel Kronenberg,David Sontag,Gerhard Hindricks,Roland Eils,Benjamin Wild
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-7] FIG: Forward-Inverse Generation for Low-Resource Domain-specific Event Detection ACL

链接: https://arxiv.org/abs/2502.17394
作者: Tanmay Parekh,Yuxuan Dong,Lucas Bandarkar,Artin Kim,I-Hung Hsu,Kai-Wei Chang,Nanyun Peng
机构: Computer Science Department, University of California, Los Angeles (加州大学洛杉矶分校计算机科学系); Google (谷歌)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Under review at ACL ARR Feb 2025

点击查看摘要

[NLP-8] Emoti-Attack: Zero-Perturbation Adversarial Attacks on NLP Systems via Emoji Sequences

链接: https://arxiv.org/abs/2502.17392
作者: Yangshijie Zhang
机构: Lanzhou University (兰州大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注:

点击查看摘要

[NLP-9] Mitigating Bias in RAG : Controlling the Embedder

链接: https://arxiv.org/abs/2502.17390
作者: Taeyoun Kim,Jacob Springer,Aditi Raghunathan,Maarten Sap
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 26 pages (8 main), 12 figures, 7 tables

点击查看摘要

[NLP-10] Big-Math: A Large-Scale High-Quality Math Dataset for Reinforcement Learning in Language Models

链接: https://arxiv.org/abs/2502.17387
作者: Alon Albalak,Duy Phung,Nathan Lile,Rafael Rafailov,Kanishk Gandhi,Louis Castricato,Anikait Singh,Chase Blagden,Violet Xiang,Dakota Mahan,Nick Haber
机构: SynthLabs; Stanford University
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-11] What is a Good Question? Utility Estimation with LLM -based Simulations

链接: https://arxiv.org/abs/2502.17383
作者: Dong-Ho Lee,Hyundong Cho,Jonathan May,Jay Pujara
机构: 未知
类目: Computation and Language (cs.CL)
备注: 18 pages, 5 figures, 6 tables

点击查看摘要

[NLP-12] Low-Rank and Sparse Model Merging for Multi-Lingual Speech Recognition and Translation ACL2025

链接: https://arxiv.org/abs/2502.17380
作者: Qiuming Zhao,Guangzhi Sun,Chao Zhang,Mingxing Xu,Thomas Fang Zheng
机构: Tsinghua University (清华大学); University of Cambridge (剑桥大学)
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: 13 pages, submitted to ACL 2025

点击查看摘要

[NLP-13] Bridging Gaps in Natural Language Processing for Yorùbá: A Systematic Review of a Decade of Progress and Prospects

【速读】： 该论文旨在解决因资源限制和其他问题导致的非洲语言，尤其是约鲁巴语（Yorùbá）在自然语言处理（NLP）领域的应用受限的问题。论文的关键在于通过系统性文献回顾，全面分析针对约鲁巴语NLP发展的研究，识别挑战、资源、技术和应用。论文指出显著障碍包括标注语料库的稀缺、预训练语言模型的有限可用性以及诸如声调复杂性和变音符号依赖性的语言学挑战。论文还揭示了包括基于规则的方法在内的主要技术，并强调了社会文化因素如代码切换和语言在数字使用中的弃置对该领域的影响。通过综合现有研究，论文为推进约鲁巴语及非洲其他语言的NLP发展奠定了基础，并旨在通过识别差距和机会来指导未来的研究。

链接: https://arxiv.org/abs/2502.17364
作者: Toheeb A. Jimoh,Tabea De Wille,Nikola S. Nikolov
机构: University of Limerick(利默里克大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

Abstract:Natural Language Processing (NLP) is becoming a dominant subset of artificial intelligence as the need to help machines understand human language looks indispensable. Several NLP applications are ubiquitous, partly due to the myriads of datasets being churned out daily through mediums like social networking sites. However, the growing development has not been evident in most African languages due to the persisting resource limitation, among other issues. Yorùbá language, a tonal and morphologically rich African language, suffers a similar fate, resulting in limited NLP usage. To encourage further research towards improving this situation, this systematic literature review aims to comprehensively analyse studies addressing NLP development for Yorùbá, identifying challenges, resources, techniques, and applications. A well-defined search string from a structured protocol was employed to search, select, and analyse 105 primary studies between 2014 and 2024 from reputable databases. The review highlights the scarcity of annotated corpora, limited availability of pre-trained language models, and linguistic challenges like tonal complexity and diacritic dependency as significant obstacles. It also revealed the prominent techniques, including rule-based methods, among others. The findings reveal a growing body of multilingual and monolingual resources, even though the field is constrained by socio-cultural factors such as code-switching and desertion of language for digital usage. This review synthesises existing research, providing a foundation for advancing NLP for Yorùbá and in African languages generally. It aims to guide future research by identifying gaps and opportunities, thereby contributing to the broader inclusion of Yorùbá and other under-resourced African languages in global NLP advancements.
zh

[NLP-14] On Relation-Specific Neurons in Large Language Models

链接: https://arxiv.org/abs/2502.17355
作者: Yihong Liu,Runsheng Chen,Lea Hirlimann,Ahmad Dawar Hakimi,Mingyang Wang,Amir Hossein Kargaran,Sascha Rothe,François Yvon,Hinrich Schütze
机构: 未知
类目: Computation and Language (cs.CL)
备注: preprint

点击查看摘要

[NLP-15] Mutual Reinforcement of LLM Dialogue Synthesis and Summarization Capabilities for Few-Shot Dialogue Summarization NAACL2025

链接: https://arxiv.org/abs/2502.17328
作者: Yen-Ju Lu,Ting-Yao Hu,Hema Swetha Koppula,Hadi Pouransari,Jen-Hao Rick Chang,Yin Xia,Xiang Kong,Qi Zhu,Simon Wang,Oncel Tuzel,Raviteja Vemulapalli
机构: Johns Hopkins University (约翰霍普金斯大学); Apple (苹果)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: NAACL 2025 Findings

点击查看摘要

[NLP-16] urning Conversations into Workflows: A Framework to Extract and Evaluate Dialog Workflows for Service AI Agents

链接: https://arxiv.org/abs/2502.17321
作者: Prafulla Kumar Choubey,Xiangyu Peng,Shilpa Bhagavath,Caiming Xiong,Shiva Kumar Pentyala,Chien-Sheng Wu
机构: Salesforce AI Research
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-17] HIPPO: Enhancing the Table Understanding Capability of Large Language Models through Hybrid-Modal Preference Optimization

链接: https://arxiv.org/abs/2502.17315
作者: Zhenghao Liu,Haolan Wang,Xinze Li,Qiushi Xiong,Xiaocui Yang,Yu Gu,Yukun Yan,Qi Shi,Fangfang Li,Ge Yu,Maosong Sun
机构: Department of Computer Science and Technology, Northeastern University, China (东北大学计算机科学与技术学院，中国); Department of Computer Science and Technology, Institute for AI, Tsinghua University, China (清华大学计算机科学与技术系人工智能研究院，中国); Beijing National Research Center for Information Science and Technology, China (北京信息科学技术国家研究中心，中国)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-18] Implicit Word Reordering with Knowledge Distillation for Cross-Lingual Dependency Parsing AAAI AAAI2025

链接: https://arxiv.org/abs/2502.17308
作者: Zhuoran Li,Chunming Hu,Junfan Chen,Zhijun Chen,Richong Zhang
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 9 pages, 5 figures, 3 tables. Accepted by The 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025)

点击查看摘要

[NLP-19] `Generalization is hallucination through the lens of tensor completions

【速读】： 该论文旨在探讨张量补全(Tensor Completions)及其伪影(Artifacts)作为理解语言模型中特定类型幻觉(Hallucinations)和泛化(Generalizations)的有用理论框架。关键在于利用张量补全方法来解析和解释语言模型中的这些现象。

链接: https://arxiv.org/abs/2502.17305
作者: Liang Ze Wong
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

Abstract:In this short position paper, we introduce tensor completions and artifacts and make the case that they are a useful theoretical framework for understanding certain types of hallucinations and generalizations in language models.
zh

[NLP-20] Child vs. machine language learning: Can the logical structure of human language unleash LLM s? ISCA

链接: https://arxiv.org/abs/2502.17304
作者: Uli Sauerland,Celia Matthaei,Felix Salfner
机构: ZAS; HU Berlin (柏林洪堡大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: ISCA/ITG Workshop on Diversity in Large Speech and Language Models

点击查看摘要

[NLP-21] Improving the Inclusivity of Dutch Speech Recognition by Fine-tuning Whisper on the JASMIN-CGN Corpus ISCA

链接: https://arxiv.org/abs/2502.17284
作者: Golshid Shekoufandeh,Paul Boersma,Antal van den Bosch
机构: University of Amsterdam (阿姆斯特丹大学); Utrecht University (乌得勒支大学)
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: ISCA/ITG Workshop on Diversity in Large Speech and Language Models

点击查看摘要

[NLP-22] Capability Instruction Tuning: A New Paradigm for Dynamic LLM Routing AAAI2025

链接: https://arxiv.org/abs/2502.17282
作者: Yi-Kai Zhang,De-Chuan Zhan,Han-Jia Ye
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: AAAI 2025; Project Page: this https URL

点击查看摘要

[NLP-23] Extracting domain-specific terms using contextual word embeddings

链接: https://arxiv.org/abs/2502.17278
作者: Andraž Repar,Nada Lavrač,Senja Pollak
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-24] MonoTODia: Translating Monologue Requests to Task-Oriented Dialogues NAACL2025

链接: https://arxiv.org/abs/2502.17268
作者: Sebastian Steindl,Ulrich Schäfer,Bernd Ludwig
机构: Ostbayerische Technische Hochschule Amberg-Weiden(奥斯特拜罗伊技术高等专科学校); University Regensburg(雷根斯堡大学)
类目: Computation and Language (cs.CL)
备注: Accepted at NAACL 2025 (Industry Track)

点击查看摘要

[NLP-25] Unveiling Downstream Performance Scaling of LLM s: A Clustering-Based Perspective

链接: https://arxiv.org/abs/2502.17262
作者: Chengyin Xu,Kaiyuan Chen,Xiao Li,Ke Shen,Chenggang Li
机构: ByteDance(字节跳动)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 21 pages,6 figures

点击查看摘要

[NLP-26] MULTITAT: Benchmarking Multilingual Table-and-Text Question Answering

【速读】： 该论文旨在解决现有表格与文本混合上下文问答（Tables and Texts Question Answering, TATQA）数据集局限于英语的问题。这一局限性导致无法评估模型在多语言环境下的表现，并且不能反映实际场景中非英语语言环境中表格和文本共存的情况。为了解决这些问题，论文提出首个多语言TATQA数据集MULTITAT，并开发了一个基线模型Ours。解决方案的关键在于通过从三个主流TATQA数据集中采样数据并将其翻译成十种不同的语言来构建MULTITAT数据集，从而实现对多语言环境下模型能力的有效评估。

链接: https://arxiv.org/abs/2502.17253
作者: Xuanliang Zhang,Dingzirui Wang,Keyan Xu,Qingfu Zhu,Wanxiang Che
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

Abstract:Question answering on the hybrid context of tables and text (TATQA) is a critical task, with broad applications in data-intensive domains. However, existing TATQA datasets are limited to English, leading to several drawbacks: (i) They overlook the challenges of multilingual TAT-QA and cannot assess model performance in the multilingual setting. (ii) They do not reflect real-world scenarios where tables and texts frequently appear in non-English languages. To address the limitations, we propose the first multilingual TATQA dataset (MULTITAT). Specifically, we sample data from 3 mainstream TATQA datasets and translate it into 10 diverse languages. To align the model TATQA capabilities in English with other languages, we develop a baseline, Ours. Experimental results reveal that the performance on non-English data in MULTITAT drops by an average of 19.4% compared to English, proving the necessity of MULTITAT. We further analyze the reasons for this performance gap. Furthermore, Ours outperforms other baselines by an average of 3.3, demonstrating its effectiveness.
zh

[NLP-27] Baichuan-Audio: A Unified Framework for End-to-End Speech Interaction

链接: https://arxiv.org/abs/2502.17239
作者: Tianpeng Li,Jun Liu,Tao Zhang,Yuanbo Fang,Da Pan,Mingrui Wang,Zheng Liang,Zehuan Li,Mingan Lin,Guosheng Dong,Jianhua Xu,Haoze Sun,Zenan Zhou,Weipeng Chen
机构: Baichuan Inc. (百川智能)
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注:

点击查看摘要

[NLP-28] Making LLM s Reason ? The Intermediate Language Problem in Neurosymbolic Approaches

链接: https://arxiv.org/abs/2502.17216
作者: Alexander Beiser,David Penz
机构: TUWien(维也纳技术大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-29] CoT-UQ: Improving Response-wise Uncertainty Quantification in LLM s with Chain-of-Thought

链接: https://arxiv.org/abs/2502.17214
作者: Boxuan Zhang,Ruqi Zhang
机构: Purdue University (普渡大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG); Machine Learning (stat.ML)
备注:

点击查看摘要

[NLP-30] Order Matters: Investigate the Position Bias in Multi-constraint Instruction Following

【速读】： 该论文旨在解决现有大型语言模型（Large Language Models, LLMs）在处理包含多个约束的真实指令时因约束顺序改变而导致性能显著波动的问题。论文的关键解决方案在于设计了一个探针任务，并引入了新的难度分布指数（Constraint Difficulty Distribution Index, CDDI），通过定量测量不同顺序下约束的难度分布，发现LLMs在面对“难到易”的约束顺序时表现更佳。此外，研究还探讨了LLMs的注意力机制与约束顺序之间的关联性。

链接: https://arxiv.org/abs/2502.17204
作者: Jie Zeng,Qianyu He,Qingyu Ren,Jiaqing Liang,Yanghua Xiao,Weikang Zhou,Zeye Sun,Fei Yu
机构: Fudan University(复旦大学); Ant Group(蚂蚁集团)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

Abstract:Real-world instructions with multiple constraints pose a significant challenge to existing large language models (LLMs). An observation is that the LLMs exhibit dramatic performance fluctuation when disturbing the order of the incorporated constraints. Yet, none of the existing works has systematically investigated this position bias problem in the field of multi-constraint instruction following. To bridge this gap, we design a probing task where we quantitatively measure the difficulty distribution of the constraints by a novel Difficulty Distribution Index (CDDI). Through the experimental results, we find that LLMs are more performant when presented with the constraints in a ``hard-to-easy’’ order. This preference can be generalized to LLMs with different architecture or different sizes of parameters. Additionally, we conduct an explanation study, providing an intuitive insight into the correlation between the LLM’s attention and constraint orders. Our code and dataset are publicly available at this https URL.
zh

[NLP-31] Evaluating Expert Contributions in a MoE LLM for Quiz-Based Tasks

链接: https://arxiv.org/abs/2502.17187
作者: Andrei Chernov
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: preprint, short paper

点击查看摘要

[NLP-32] Measuring Data Diversity for Instruction Tuning: A Systematic Analysis and A Reliable Metric

链接: https://arxiv.org/abs/2502.17184
作者: Yuming Yang,Yang Nan,Junjie Ye,Shihan Dou,Xiao Wang,Shuo Li,Huijie Lv,Tao Gui,Qi Zhang,Xuanjing Huang
机构: School of Computer Science, Fudan University (计算机科学学院，复旦大学); Institute of Modern Languages and Linguistics, Fudan University (现代语言与语言学研究所，复旦大学)
类目: Computation and Language (cs.CL)
备注: 15 pages. The related codes and resources will be released later. Project page: this https URL

点击查看摘要

[NLP-33] Cheems: A Practical Guidance for Building and Evaluating Chinese Reward Models from Scratch

链接: https://arxiv.org/abs/2502.17173
作者: Xueru Wen,Jie Lou,Zichao Li,Yaojie Lu,Xing Yu,Yuqiu Ji,Guohai Xu,Hongyu Lin,Ben He,Xianpei Han,Le Sun,Debing Zhang
机构: Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所中文信息处理实验室); University of Chinese Academy of Sciences(中国科学院大学); Xiaohongshu Inc(小红书)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-34] Logic Haystacks: Probing LLM s Long-Context Logical Reasoning (Without Easily Identifiable Unrelated Padding)

链接: https://arxiv.org/abs/2502.17169
作者: Damien Sileo
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-35] JUREX-4E: Juridical Expert-Annotated Four-Element Knowledge Base for Legal Reasoning

链接: https://arxiv.org/abs/2502.17166
作者: Huanghai Liu,Quzhe Huang,Qingjing Chen,Yiran Hu,Jiayu Ma,Yun Liu,Weixing Shen,Yansong Feng
机构: School of Law, Tsinghua University (清华大学法学院); Wangxuan Institute of Computer Technology, Peking University (北京大学王选计算机研究所); Department of Legal Studies, University of Bologna (博洛尼亚大学法律系)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-36] MEMERAG : A Multilingual End-to-End Meta-Evaluation Benchmark for Retrieval Augmented Generation

链接: https://arxiv.org/abs/2502.17163
作者: María Andrea Cruz Blandón,Jayasimha Talur,Bruno Charron,Dong Liu,Saab Mansour,Marcello Federico
机构: Amazon
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-37] Sentiment analysis of texts from social networks based on machine learning methods for monitoring public sentiment

链接: https://arxiv.org/abs/2502.17143
作者: Arsen Tolebay Nurlanuly
机构: 未知
类目: Computation and Language (cs.CL)
备注: 10 pages, 5 figures, 2 tables. Preprint submitted for community feedback

点击查看摘要

[NLP-38] hus Spake Long-Context Large Language Model

链接: https://arxiv.org/abs/2502.17129
作者: Xiaoran Liu,Ruixiao Li,Mianqiu Huang,Zhigeng Liu,Yuerong Song,Qipeng Guo,Siyang He,Qiqi Wang,Linlin Li,Qun Liu,Yaqian Zhou,Xuanjing Huang,Xipeng Qiu
机构: Shanghai AI Lab (上海人工智能实验室); School of Computer Science Fudan University (复旦大学计算机学院); Huawei Noah’s Ark Lab (华为诺亚方舟实验室); Shanghai Innovation Institute (上海创新研究院)
类目: Computation and Language (cs.CL)
备注: a global picture of the lifecycle of long-context LLMs from four perspectives: architecture, infrastructure, training, and evaluation

点击查看摘要

[NLP-39] LettuceDetect: A Hallucination Detection Framework for RAG Applications

链接: https://arxiv.org/abs/2502.17125
作者: Ádám Kovács,Gábor Recski
机构: KR Labs; TU Wien
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 6 pages

点击查看摘要

[NLP-40] Mobile-Agent -V: Learning Mobile Device Operation Through Video-Guided Multi-Agent Collaboration

链接: https://arxiv.org/abs/2502.17110
作者: Junyang Wang,Haiyang Xu,Xi Zhang,Ming Yan,Ji Zhang,Fei Huang,Jitao Sang
机构: Beijing Jiaotong University (北京交通大学); Alibaba Group (阿里巴巴集团)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 7 figures, 7tables

点击查看摘要

[NLP-41] WildFrame: Comparing Framing in Humans and LLM s on Naturally Occurring Texts

链接: https://arxiv.org/abs/2502.17091
作者: Gili Lior,Liron Nacchace,Gabriel Stanovsky
机构: The Hebrew University of Jerusalem
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-42] Automatically Evaluating the Paper Reviewing Capability of Large Language Models

链接: https://arxiv.org/abs/2502.17086
作者: Hyungyu Shin,Jingyu Tang,Yoonjoo Lee,Nayoung Kim,Hyunseung Lim,Ji Yong Cho,Hwajung Hong,Moontae Lee,Juho Kim
机构: KAIST; Huazhong University of Science and Technology (华中科技大学); LG AI Research; University of Illinois Chicago (芝加哥伊利诺伊大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-43] Systematic Weight Evaluation for Pruning Large Language Models : Enhancing Performance and Sustainability

链接: https://arxiv.org/abs/2502.17071
作者: Ashhadul Islam,Samir Brahim Belhaouari,Amine Bermak
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-44] PrivaCI-Bench: Evaluating Privacy with Contextual Integrity and Legal Compliance

链接: https://arxiv.org/abs/2502.17041
作者: Haoran Li,Wenbin Hu,Huihao Jing,Yulin Chen,Qi Hu,Sirui Han,Tianshu Chu,Peizhao Hu,Yangqiu Song
机构: HKUST(香港科技大学); National University of Singapore(新加坡国立大学); Huawei Technologies(华为技术有限公司)
类目: Computation and Language (cs.CL)
备注: Project Webpage: this https URL

点击查看摘要

[NLP-45] Language Model Re-rankers are Steered by Lexical Similarities

链接: https://arxiv.org/abs/2502.17036
作者: Lovisa Hagström,Ercong Nie,Ruben Halifa,Helmut Schmid,Richard Johansson,Alexander Junge
机构: Chalmers University of Technology(查尔莫斯理工大学); University of Gothenburg(哥德堡大学); LMU Munich(慕尼黑大学); Munich Center for Machine Learning(慕尼黑机器学习中心); amass technologies(未知公司)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 16 pages

点击查看摘要

[NLP-46] Understanding the Uncertainty of LLM Explanations: A Perspective Based on Reasoning Topology

链接: https://arxiv.org/abs/2502.17026
作者: Longchao Da,Xiaoou Liu,Jiaxin Dai,Lu Cheng,Yaqing Wang,Hua Wei
机构: Arizona State University (亚利桑那州立大学); University of Illinois Chicago (伊利诺伊大学芝加哥分校); Purdue University (普渡大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Symbolic Computation (cs.SC)
备注: 15 pages, 6 figures

点击查看摘要

[NLP-47] owards Auto-Regressive Next-Token Prediction: In-Context Learning Emerges from Generalization ICLR2025

链接: https://arxiv.org/abs/2502.17024
作者: Zixuan Gong,Xiaolin Hu,Huayi Tang,Yong Liu
机构: Gaoling School of Artificial Intelligence (高瓴人工智能学院), Renmin University of China (中国人民大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG); Machine Learning (stat.ML)
备注: Published at ICLR 2025

点击查看摘要

[NLP-48] Quantifying Logical Consistency in Transformers via Query-Key Alignment

链接: https://arxiv.org/abs/2502.17017
作者: Eduard Tulchinskii,Anastasia Voznyuk,Laida Kushnareva,Andrei Andriiainen,Irina Piontkovskaya,Evgeny Burnaev,Serguei Barannikov
机构: Skolkovo Institute of Science and Technology; AI Foundation and Algorithm Lab; Moscow Institute of Physics and Technology; CNRS, Université Paris Cité, France; Artificial Intelligence Research Institute (AIRI)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-49] FADE: Why Bad Descriptions Happen to Good Features

链接: https://arxiv.org/abs/2502.16994
作者: Bruno Puri,Aakriti Jain,Elena Golimblevskaia,Patrick Kahardipraja,Thomas Wiegand,Wojciech Samek,Sebastian Lapuschkin
机构: Department of Artificial Intelligence, Fraunhofer Heinrich Hertz Institute (弗劳恩霍夫赫兹研究所人工智能系), Berlin, Germany; Department of Electrical Engineering and Computer Science, Technische Universität Berlin (柏林工业大学电气工程与计算机科学系), Berlin, Germany; BIFOLD - Berlin Institute for the Foundations of Learning and Data (柏林学习与数据基础研究所), Berlin, Germany
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-50] All-in-one: Understanding and Generation in Multimodal Reasoning with the MAIA Benchmark

链接: https://arxiv.org/abs/2502.16989
作者: Davide Testa,Giovanni Bonetta,Raffaella Bernardi,Alessandro Bondielli,Alessandro Lenci,Alessio Miaschi,Lucia Passaro,Bernardo Magnini
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-51] Hotter and Colder: A New Approach to Annotating Sentiment Emotions and Bias in Icelandic Blog Comments ALT

链接: https://arxiv.org/abs/2502.16987
作者: Steinunn Rut Friðriksdóttir,Dan Saattrup Nielsen,Hafsteinn Einarsson
机构: University of Iceland(冰岛大学); The Alexandra Institute(亚历山德拉研究所)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: To be published in the proceedings of the NoDaLiDa/Baltic-HLT 2025 conference

点击查看摘要

[NLP-52] Muon is Scalable for LLM Training

链接: https://arxiv.org/abs/2502.16982
作者: Jingyuan Liu,Jianlin Su,Xingcheng Yao,Zhejun Jiang,Guokun Lai,Yulun Du,Yidao Qin,Weixin Xu,Enzhe Lu,Junjie Yan,Yanru Chen,Huabin Zheng,Yibo Liu,Shaowei Liu,Bohong Yin,Weiran He,Han Zhu,Yuzhi Wang,Jianzhou Wang,Mengnan Dong,Zheng Zhang,Yongsheng Kang,Hao Zhang,Xinran Xu,Yutao Zhang,Yuxin Wu,Xinyu Zhou,Zhilin Yang
机构: Moonshot AI; UCLA
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-53] LongSafety: Evaluating Long-Context Safety of Large Language Models

【速读】： 该论文旨在解决大型语言模型（Large Language Models, LLMs）在长上下文任务中的安全问题。目前，LLMs在长上下文任务中的安全性尚未得到充分探索，存在显著的评估和改进缺口。为了解决这一问题，论文引入了LongSafety，这是一个专门设计用于评估LLMs在开放性长上下文任务中安全性的首个综合性基准。LongSafety包含7类安全问题和6个以用户为中心的长上下文任务，共计1,543个测试案例，平均每个上下文包含5,424个单词。论文通过对16个代表性LLMs的评估发现显著的安全漏洞，大多数模型的安全性低于55%。研究的关键在于通过LongSafety基准系统地评估和揭示长上下文任务中的安全挑战，并强调在长上下文场景中相关上下文和扩展输入序列可能加剧安全风险。

链接: https://arxiv.org/abs/2502.16971
作者: Yida Lu,Jiale Cheng,Zhexin Zhang,Shiyao Cui,Cunxiang Wang,Xiaotao Gu,Yuxiao Dong,Jie Tang,Hongning Wang,Minlie Huang
机构: The Conversational AI (CoAI) group, DCST, Tsinghua University (清华大学); Zhipu AI (智谱AI); The Knowledge Engineering Group (KEG), DCST, Tsinghua University (知识工程组, 清华大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

Abstract:As Large Language Models (LLMs) continue to advance in understanding and generating long sequences, new safety concerns have been introduced through the long context. However, the safety of LLMs in long-context tasks remains under-explored, leaving a significant gap in both evaluation and improvement of their safety. To address this, we introduce LongSafety, the first comprehensive benchmark specifically designed to evaluate LLM safety in open-ended long-context tasks. LongSafety encompasses 7 categories of safety issues and 6 user-oriented long-context tasks, with a total of 1,543 test cases, averaging 5,424 words per context. Our evaluation towards 16 representative LLMs reveals significant safety vulnerabilities, with most models achieving safety rates below 55%. Our findings also indicate that strong safety performance in short-context scenarios does not necessarily correlate with safety in long-context tasks, emphasizing the unique challenges and urgency of improving long-context safety. Moreover, through extensive analysis, we identify challenging safety issues and task types for long-context models. Furthermore, we find that relevant context and extended input sequences can exacerbate safety risks in long-context scenarios, highlighting the critical need for ongoing attention to long-context safety challenges. Our code and data are available at this https URL.
zh

[NLP-54] UrduLLaMA 1.0: Dataset Curation Preprocessing and Evaluation in Low-Resource Settings

链接: https://arxiv.org/abs/2502.16961
作者: Layba Fiaz,Munief Hassan Tahir,Sana Shams,Sarmad Hussain
机构: Center for Language Engineering (语言工程中心); Al-Khawarizmi Institute of Computer Science (阿尔霍瓦里兹米计算机科学研究所); University of Engineering and Technology, Lahore (拉合尔工程技术大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-55] SparseTransX: Efficient Training of Translation-Based Knowledge Graph Embeddings Using Sparse Matrix Operations

链接: https://arxiv.org/abs/2502.16949
作者: Md Saidul Hoque Anik,Ariful Azad
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-56] Using Machine Learning to Detect Fraudulent SMSs in Chichewa

链接: https://arxiv.org/abs/2502.16947
作者: Amelia Taylor,Amoss Robert
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-57] NUTSHELL: A Dataset for Abstract Generation from Scientific Talks

链接: https://arxiv.org/abs/2502.16942
作者: Maike Züfle,Sara Papi,Beatrice Savoldi,Marco Gaido,Luisa Bentivogli,Jan Niehues
机构: Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院); Fondazione Bruno Kessler(布鲁诺·凯斯勒基金会)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-58] Reasoning Does Not Necessarily Improve Role-Playing Ability

【速读】： 该论文旨在探究推理技术是否能够提升角色扮演大型语言模型（LLMs）的角色扮演能力。研究通过使用6个角色扮演基准测试、24个LLMs及3种不同的角色扮演策略，对比直接零样本角色扮演、带有链式思维（CoT）的角色扮演以及使用推理优化的LLMs进行角色扮演的效果。关键在于验证不同推理方法对角色扮演性能的影响，并据此提出改进角色扮演LLMs的未来研究方向。

链接: https://arxiv.org/abs/2502.16940
作者: Xiachong Feng,Longxu Dou,Lingpeng Kong
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

Abstract:The application of role-playing large language models (LLMs) is rapidly expanding in both academic and commercial domains, driving an increasing demand for high-precision role-playing models. Simultaneously, the rapid advancement of reasoning techniques has continuously pushed the performance boundaries of LLMs. This intersection of practical role-playing demands and evolving reasoning capabilities raises an important research question: “Can reasoning techniques enhance the role-playing capabilities of LLMs?” To address this, we conduct a comprehensive study using 6 role-playing benchmarks, 24 LLMs, and 3 distinct role-playing strategies, comparing the effectiveness of direct zero-shot role-playing, role-playing with Chain-of-Thought (CoT), and role-playing using reasoning-optimized LLMs. Our findings reveal that CoT may reduce role-playing performance, reasoning-optimized LLMs are unsuitable for role-playing, reasoning ability disrupts the role-playing scaling law, large models still lack proficiency in advanced role-playing, and Chinese role-playing performance surpasses English role-playing performance. Furthermore, based on extensive experimental results, we propose two promising future research directions: Role-aware CoT for improving role-playing LLMs and Reinforcement Learning for role-playing LLMs, aiming to enhance the adaptability, consistency, and effectiveness of role-playing LLMs for both research and real-world applications.
zh

[NLP-59] A Systematic Survey of Automatic Prompt Optimization Techniques

链接: https://arxiv.org/abs/2502.16923
作者: Kiran Ramnath,Kang Zhou,Sheng Guan,Soumya Smruti Mishra,Xuan Qi,Zhengyuan Shen,Shuai Wang,Sangmin Woo,Sullam Jeoung,Yawei Wang,Haozhu Wang,Han Ding,Yuzhe Lu,Zhichao Xu,Yun Zhou,Balasubramaniam Srinivasan,Qiaojing Yan,Yueyan Chen,Haibo Ding,Panpan Xu,Lin Lee Cheong
机构: Amazon Web Services (亚马逊网络服务)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 8 main pages, 31 total pages, 1 figure

点击查看摘要

[NLP-60] Benchmarking Temporal Reasoning and Alignment Across Chinese Dynasties

链接: https://arxiv.org/abs/2502.16922
作者: Zhenglin Wang,Jialong Wu,Pengfei LI,Yong Jiang,Deyu Zhou
机构: School of Computer Science and Engineering, Key Laboratory of Computer Network and Information Integration, Ministry of Education, Southeast University, China(东南大学); Tongyi Lab, Alibaba Group(阿里集团)
类目: Computation and Language (cs.CL)
备注: Preprint

点击查看摘要

[NLP-61] SS-MPC: A Sequence-Structured Multi-Party Conversation System

链接: https://arxiv.org/abs/2502.16920
作者: Yoonjin Jang,Keunha Kim,Youngjoong Ko
机构: SungKyunKwan University (成均馆大学)
类目: Computation and Language (cs.CL)
备注: 8 pages, 5 figures

点击查看摘要

[NLP-62] Dependency Parsing with the Structuralized Prompt Template

链接: https://arxiv.org/abs/2502.16919
作者: Keunha Kim,Youngjoong Ko
机构: SungKyunKwan University (成均馆大学)
类目: Computation and Language (cs.CL)
备注: 12pages, 5 figures

点击查看摘要

[NLP-63] AutoLogi: Automated Generation of Logic Puzzles for Evaluating Reasoning Abilities of Large Language Models

链接: https://arxiv.org/abs/2502.16906
作者: Qin Zhu,Fei Huang,Runyu Peng,Keming Lu,Bowen Yu,Qinyuan Cheng,Xipeng Qiu,Xuanjing Huang,Junyang Lin
机构: School of Computer Science, Fudan University (复旦大学计算机学院); Shanghai Key Laboratory of Intelligent Information Processing, Fudan University (复旦大学智能信息处理重点实验室); Qwen Team, Alibaba Group (阿里集团Qwen团队)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-64] GuidedBench: Equipping Jailbreak Evaluation with Guidelines

链接: https://arxiv.org/abs/2502.16903
作者: Ruixuan Huang,Xunguang Wang,Zongjie Li,Daoyuan Wu,Shuai Wang
机构: The Hong Kong University of Science and Technology (香港科技大学)
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注: Homepage: this https URL

点击查看摘要

[NLP-65] Char-mander Use mBackdoor! A Study of Cross-lingual Backdoor Attacks in Multilingual LLM s

链接: https://arxiv.org/abs/2502.16901
作者: Himanshu Beniwal,Sailesh Panda,Mayank Singh
机构: Indian Institute of Technology Gandhinagar
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-66] Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment

链接: https://arxiv.org/abs/2502.16894
作者: Chenghao Fan,Zhenyi Lu,Sichen Liu,Xiaoye Qu,Wei Wei,Chengfeng Gu,Yu Cheng
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-67] Applying LLM s to Active Learning: Towards Cost-Efficient Cross-Task Text Classification without Manually Labeled Data

链接: https://arxiv.org/abs/2502.16892
作者: Yejian Zhang,Shingo Takada
机构: Graduate School of Science and Technology, Keio University (Keio大学), Yokohama, Japan
类目: Computation and Language (cs.CL)
备注: Statement in Accordance with IEEE Preprint Policy: This work is intended for submission to the IEEE for possible publication

点击查看摘要

[NLP-68] DBudgetKV: Dynamic Budget in KV Cache Compression for Ensuring Optimal Performance

链接: https://arxiv.org/abs/2502.16886
作者: Xuanfan Ni,Liyan Xu,Chenyang Lyu,Longyue Wang,Mo Yu,Lemao Liu,Fandong Meng,Jie Zhou,Piji Li
机构: Pattern Recognition Center, WeChat AI (微信AI模式识别中心)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-69] CORAL: Learning Consistent Representations across Multi-step Training with Lighter Speculative Drafter

链接: https://arxiv.org/abs/2502.16880
作者: Yepeng Weng,Dianwen Mei,Huishi Qiu,Xujie Chen,Li Liu,Jiang Tian,Zhongchao Shi
机构: Lenovo Research (联想研究院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Under Review

点击查看摘要

[NLP-70] LongAttn: Selecting Long-context Training Data via Token-level Attention

链接: https://arxiv.org/abs/2502.16860
作者: Longyun Wu,Dawei Zhu,Guangxiang Zhao,Zhuocheng Yu,Junfeng Ran,Xiangyu Wong,Lin Sun,Sujian Li
机构: 未知
类目: Computation and Language (cs.CL)
备注: 17 pages, 5 figures

点击查看摘要

[NLP-71] Sarang at DEFACTIFY 4.0: Detecting AI-Generated Text Using Noised Data and an Ensemble of DeBERTa Models AAAI-25

链接: https://arxiv.org/abs/2502.16857
作者: Avinash Trivedi,Sangeetha Sivanesan
机构: National Institute of Technology, Tiruchirapalli (国立技术学院，蒂鲁吉亚帕拉姆)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: AAAI-25 DEFACTIFY 4.0 Workshop AI generated text detection (1st Rank)

点击查看摘要

[NLP-72] Improving LLM General Preference Alignment via Optimistic Online Mirror Descent

链接: https://arxiv.org/abs/2502.16852
作者: Yuheng Zhang,Dian Yu,Tao Ge,Linfeng Song,Zhichen Zeng,Haitao Mi,Nan Jiang,Dong Yu
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-73] “Actionable Help” in Crises: A Novel Dataset and Resource-Efficient Models for Identifying Request and Offer Social Media Posts

链接: https://arxiv.org/abs/2502.16839
作者: Rabindra Lamsal,Maria Rodriguez Read,Shanika Karunasekera,Muhammad Imran
机构: The University of Melbourne (墨尔本大学), Australia; Qatar Computing Research Institute (卡塔尔计算研究学院), Qatar
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-74] REGen: A Reliable Evaluation Framework for Generative Event Argument Extraction

链接: https://arxiv.org/abs/2502.16838
作者: Omar Sharif,Joseph Gatto,Madhusudan Basak,Sarah M. Preum
机构: Department of Computer Science, Dartmouth College (计算机科学系，达特茅斯学院)
类目: Computation and Language (cs.CL)
备注: 20 pages, 9 figures, 13 tables

点击查看摘要

[NLP-75] Finding the Sweet Spot: Preference Data Construction for Scaling Preference Optimization

【速读】： 该论文旨在通过扩大策略模型生成的样本数量来提升大语言模型（LLMs）的对齐性能。然而，传统方法在使用Direct Preference Optimization (DPO)过程中，选择最高奖励作为被选响应，最低奖励作为拒绝响应的做法，在样本量增加时会导致性能下降。论文的关键解决方案在于重新定义拒绝响应的选择标准：通过探索奖励分布的正态分布特性，发现选择奖励位于 (\mu - 2\sigma) 的响应作为拒绝响应，而非最低奖励响应，能够显著优化模型性能，并提出了一种可扩展的偏好数据构建策略，以确保模型性能随着样本规模的增加而持续提升。

链接: https://arxiv.org/abs/2502.16825
作者: Yao Xiao,Hai Ye,Linyao Chen,Hwee Tou Ng,Lidong Bing,Xiaoli Li,Roy Ka-wei Lee
机构: Singapore University of Technology and Design; National University of Singapore; The University of Tokyo; Shanda AI Research Institute; Institute for Infocomm Research, A*Star, Singapore
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

Abstract:Iterative data generation and model retraining are widely used to align large language models (LLMs). It typically involves a policy model to generate on-policy responses and a reward model to guide training data selection. Direct Preference Optimization (DPO) further enhances this process by constructing preference pairs of chosen and rejected responses. In this work, we aim to \emphscale up the number of on-policy samples via repeated random sampling to improve alignment performance. Conventional practice selects the sample with the highest reward as chosen and the lowest as rejected for DPO. However, our experiments reveal that this strategy leads to a \emphdecline in performance as the sample size increases. To address this, we investigate preference data construction through the lens of underlying normal distribution of sample rewards. We categorize the reward space into seven representative points and systematically explore all 21 ( C_7^2 ) pairwise combinations. Through evaluations on four models using AlpacaEval 2, we find that selecting the rejected response at reward position \mu - 2\sigma rather than the minimum reward, is crucial for optimal performance. We finally introduce a scalable preference data construction strategy that consistently enhances model performance as the sample scale increases.
zh

[NLP-76] Uncertainty Quantification of Large Language Models through Multi-Dimensional Responses

链接: https://arxiv.org/abs/2502.16820
作者: Tiejin Chen,Xiaoou Liu,Longchao Da,Xiaoou Liu,Vagelis Papalexakis,Hua Wei
机构: Arizona State University(Tempe)(亚利桑那州立大学); University of California, Riverside(Riverside)(加州大学河滨分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-77] Grounded Persuasive Language Generation for Automated Marketing

链接: https://arxiv.org/abs/2502.16810
作者: Jibang Wu,Chenghao Yang,Simon Mahns,Chaoqi Wang,Hao Zhu,Fei Fang,Haifeng Xu
机构: University of Chicago (芝加哥大学); Stanford University (斯坦福大学); Carnegie Mellon University (卡内基梅隆大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); General Economics (econ.GN)
备注:

点击查看摘要

[NLP-78] CoT2Align: Cross-Chain of Thought Distillation via Optimal Transport Alignment for Language Models with Different Tokenizers

链接: https://arxiv.org/abs/2502.16806
作者: Anh Duc Le,Tu Vu,Nam Le Hai,Nguyen Thi Ngoc Diep,Linh Ngo Van,Trung Le,Thien Huu Nguyen
机构: Hanoi University of Science and Technology(河内工业大学); Oraichain Labs Inc., US(美国Oraichain实验室有限公司); ByteDance Inc.(字节跳动有限公司); VNU University of Engineering and Technology(越南国立工程大学); Monash University(蒙纳士大学); University of Oregon(俄勒冈大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-79] Unsupervised Topic Models are Data Mixers for Pre-training Language Models

链接: https://arxiv.org/abs/2502.16802
作者: Jiahui Peng,Xinlin Zhuang,Qiu Jiantao,Ren Ma,Jing Yu,Tianyi Bai,Conghui He
机构: Shanghai AI Laboratory; School of Computer Science and Technology, East China Normal University
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 18 pages,7 figures

点击查看摘要

[NLP-80] MobileSteward: Integrating Multiple App-Oriented Agents with Self-Evolution to Automate Cross-App Instructions KDD2025

链接: https://arxiv.org/abs/2502.16796
作者: Yuxuan Liu,Hongda Sun,Wei Liu,Jian Luan,Bo Du,Rui Yan
机构: Gaoling School of Artificial Intelligence, Renmin University of China(高瓴人工智能学院, 中国人民大学); Xiaomi AI Lab(小米人工智能实验室); School of Computer Science, Wuhan University(计算机科学学院, 武汉大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: Accepted by KDD2025 Research Track

点击查看摘要

[NLP-81] AAD-LLM : Neural Attention-Driven Auditory Scene Understanding

链接: https://arxiv.org/abs/2502.16794
作者: Xilin Jiang,Sukru Samet Dindar,Vishal Choudhari,Stephan Bickel,Ashesh Mehta,Guy M McKhann,Adeen Flinker,Daniel Friedman,Nima Mesgarani
机构: Columbia University (哥伦比亚大学); Mortimer B. Zuckerman Mind Brain Behavior Institute (莫蒂默·B·扎克伯格脑行为研究所); Hofstra Northwell School of Medicine (霍夫斯特拉北威尔医学院); The Feinstein Institutes for Medical Research (菲因斯坦医学研究所); New York University (纽约大学)
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); Audio and Speech Processing (eess.AS)
备注:

点击查看摘要

[NLP-82] he Role of Sparsity for Length Generalization in Transformers

链接: https://arxiv.org/abs/2502.16792
作者: Noah Golowich,Samy Jelassi,David Brandfonbrener,Sham M. Kakade,Eran Malach
机构: MIT(麻省理工学院); Harvard University(哈佛大学); Kempner Institute at Harvard University(哈佛大学Kempner研究所)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-83] Are Large Language Models Good Data Preprocessors?

链接: https://arxiv.org/abs/2502.16790
作者: Elyas Meguellati,Nardiena Pratama,Shazia Sadiq,Gianluca Demartini
机构: The University of Queensland(昆士兰大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-84] MultiOCR-QA: Dataset for Evaluating Robustness of LLM s in Question Answering on Multilingual OCR Texts

链接: https://arxiv.org/abs/2502.16781
作者: Bhawna Piryani,Jamshid Mozafari,Abdelrahman Abdallah,Antoine Doucet,Adam Jatowt
机构: University of Innsbruck(因斯布鲁克大学); University of La Rochelle(勒阿弗尔大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-85] AISafetyLab: A Comprehensive Framework for AI Safety Evaluation and Improvement

链接: https://arxiv.org/abs/2502.16776
作者: Zhexin Zhang,Leqi Lei,Junxiao Yang,Xijie Huang,Yida Lu,Shiyao Cui,Renmiao Chen,Qinglin Zhang,Xinyuan Wang,Hao Wang,Hao Li,Xianqi Lei,Chengwei Pan,Lei Sha,Hongning Wang,Minlie Huang
机构: The Conversational AI (CoAI) group, DCST, Tsinghua University (清华大学); Beihang University (北京航空航天大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 13 pages

点击查看摘要

[NLP-86] LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint

链接: https://arxiv.org/abs/2502.16770
作者: Qianli Ma,Dongrui Liu,Qian Chen,Linfeng Zhang,Jing Shao
机构: Shanghai Jiao Tong University(上海交通大学); Shanghai AI Laboratory(上海人工智能实验室); East China Normal University(华东师范大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-87] A Hybrid Approach to Information Retrieval and Answer Generation for Regulatory Texts

链接: https://arxiv.org/abs/2502.16767
作者: Jhon Rayo,Raul de la Rosa,Mario Garrido
机构: Universidad de los Andes(安第斯大学)
类目: Computation and Language (cs.CL)
备注: 5 pages; Workshop paper; Proceedings of the 1st Regulatory NLP Workshop (RegNLP 2025)

点击查看摘要

[NLP-88] ATEB: Evaluating and Improving Advanced NLP Tasks for Text Embedding Models

链接: https://arxiv.org/abs/2502.16766
作者: Simeng Han,Frank Palma Gomez,Tu Vu,Zefei Li,Daniel Cer,Hansi Zeng,Chris Tar,Arman Cohan,Gustavo Hernandez Abrego
机构: Google Deepmind(谷歌深梦); Yale University(耶鲁大学); University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-89] Language Model Fine-Tuning on Scaled Survey Data for Predicting Distributions of Public Opinions

链接: https://arxiv.org/abs/2502.16761
作者: Joseph Suh,Erfan Jahanparast,Suhong Moon,Minwoo Kang,Serina Chang
机构: University of California, Berkeley(加州大学伯克利分校); Microsoft Research(微软研究)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-90] Entailment-Preserving First-order Logic Representations in Natural Language Entailment

链接: https://arxiv.org/abs/2502.16757
作者: Jinu Lee,Qi Liu,Runzhi Ma,Vincent Han,Ziqi Wang,Heng Ji,Julia Hockenmaier
机构: University of Illinois Urbana-Champaign
类目: Computation and Language (cs.CL)
备注: 14 pages (8 pages of main content), 8 figures

点击查看摘要

[NLP-91] SQLong: Enhanced NL2SQL for Longer Contexts with LLM s

链接: https://arxiv.org/abs/2502.16747
作者: Dai Quoc Nguyen,Cong Duy Vu Hoang,Duy Vu,Gioacchino Tangari,Thanh Tien Vu,Don Dharmasiri,Yuan-Fang Li,Long Duong
机构: Oracle Corporation (甲骨文公司) Australia
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注:

点击查看摘要

[NLP-92] Layer-Wise Evolution of Representations in Fine-Tuned Transformers: Insights from Sparse AutoEncoders

链接: https://arxiv.org/abs/2502.16722
作者: Suneel Nadipalli
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 14 pages, 5 figures

点击查看摘要

[NLP-93] Speed and Conversational Large Language Models : Not All Is About Tokens per Second

链接: https://arxiv.org/abs/2502.16721
作者: Javier Conde,Miguel González,Pedro Reviriego,Zhen Gao,Shanshan Liu,Fabrizio Lombardi
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-94] Beyond Pattern Recognition: Probing Mental Representations of LMs

链接: https://arxiv.org/abs/2502.16717
作者: Moritz Miller,Kumar Shridhar
机构: unknown
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-95] DISC: Dynamic Decomposition Improves LLM Inference Scaling

链接: https://arxiv.org/abs/2502.16706
作者: Jonathan Light,Wei Cheng,Wu Yue,Masafumi Oyamada,Mengdi Wang,Santiago Paternain,Haifeng Chen
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE)
备注:

点击查看摘要

[NLP-96] Can ChatGPT Learn to Count Letters?

链接: https://arxiv.org/abs/2502.16705
作者: Javier Conde,Gonzalo Martínez,Pedro Reviriego,Zhen Gao,Shanshan Liu,Fabrizio Lombardi
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-97] Code Summarization Beyond Function Level ICSE’25

链接: https://arxiv.org/abs/2502.16704
作者: Vladimir Makharev,Vladimir Ivanov
机构: Innopolis University(因诺波利斯大学); AIRI(未知)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted to LLM4Code @ ICSE’25; 8 pages, 3 figures, 4 tables

点击查看摘要

[NLP-98] Uncovering the Hidden Threat of Text Watermarking from Users with Cross-Lingual Knowledge

链接: https://arxiv.org/abs/2502.16699
作者: Mansour Al Ghanim,Jiaqi Xue,Rochana Prih Hastuti,Mengxin Zheng,Yan Solihin,Qian Lou
机构: University of Central Florida (中佛罗里达大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 9 pages

点击查看摘要

[NLP-99] oward Responsible Federated Large Language Models : Leverag ing a Safety Filter and Constitutional AI

链接: https://arxiv.org/abs/2502.16691
作者: Eunchung Noh,Jeonghun Baek
机构: Samsung Electronics(Samsung电子); The University of Tokyo(东京大学)
类目: Computation and Language (cs.CL); Distributed, Parallel, and Cluster Computing (cs.DC); Multiagent Systems (cs.MA)
备注: 5 pages, 3 figures

点击查看摘要

[NLP-100] WildLong: Synthesizing Realistic Long-Context Instruction Data at Scale

链接: https://arxiv.org/abs/2502.16684
作者: Jiaxi Li,Xingxing Zhang,Xun Wang,Xiaolong Huang,Li Dong,Liang Wang,Si-Qing Chen,Wei Lu,Furu Wei
机构: Microsoft Research (微软研究); Singapore University of Technology and Design (新加坡科技设计大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-101] Automatic Input Rewriting Improves Translation with Large Language Models

链接: https://arxiv.org/abs/2502.16682
作者: Dayeon Ki,Marine Carpuat
机构: University of Maryland (马里兰大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 27 pages, 8 figures

点击查看摘要

[NLP-102] MimeQA: Towards Socially-Intelligent Nonverbal Foundation Models

链接: https://arxiv.org/abs/2502.16671
作者: Hengzhi Li,Megan Tjandrasuwita,Yi R. Fung,Armando Solar-Lezama,Paul Pu Liang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[NLP-103] CODESYNC: Synchronizing Large Language Models with Dynamic Code Evolution at Scale

链接: https://arxiv.org/abs/2502.16645
作者: Chenlong Wang,Zhaoyang Chu,Zhengxiang Cheng,Xuyi Yang,Kaiyue Qiu,Yao Wan,Zhou Zhao,Xuanhua Shi,Dongping Chen
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:

点击查看摘要

[NLP-104] Retrieval-Augmented Visual Question Answering via Built-in Autoregressive Search Engines AAAI-25

链接: https://arxiv.org/abs/2502.16641
作者: Xinwei Long,Zhiyuan Ma,Ermo Hua,Kaiyan Zhang,Biqing Qi,Bowen Zhou
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: AAAI-25

点击查看摘要

[NLP-105] Visual-RAG : Benchmarking Text-to-Image Retrieval Augmented Generation for Visual Knowledge Intensive Queries

链接: https://arxiv.org/abs/2502.16636
作者: Yin Wu,Quanyu Long,Jing Li,Jianfei Yu,Wenya Wang
机构: Nanyang Technological University (南洋理工大学); Harbin Institute of Technology (Shenzhen) (哈尔滨工业大学（深圳）); Nanjing University of Science and Technology (南京理工大学)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: 23 pages, 6 figures

点击查看摘要

[NLP-106] Can Large Vision-Language Models Detect Images Copyright Infringement from GenAI?

链接: https://arxiv.org/abs/2502.16618
作者: Qipan Xu,Zhenting Wang,Xiaoxiao He,Ligong Han,Ruixiang Tang
机构: Rutgers University
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-107] CodeCriticBench: A Holistic Code Critique Benchmark for Large Language Models

链接: https://arxiv.org/abs/2502.16614
作者: Alexander Zhang,Marcus Dong,Jiaheng Liu,Wei Zhang,Yejie Wang,Jian Yang,Ge Zhang,Tianyu Liu,Zhongyuan Peng,Yingshui Tan,Yuanxing Zhang,Zhexu Wang,Weixun Wang,Yancheng He,Ken Deng,Wangchunshu Zhou,Wenhao Huang,Zhaoxiang Zhang
机构: NJU(南京大学); M-A-P; Alibaba(阿里巴巴); BUAA(北京航空航天大学); CASIA(中科院自动化所); BUPT(北京邮电大学); Kuaishou(快手); OPPO(OPPO)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-108] MemeIntel: Explainable Detection of Propagandistic and Hateful Memes

链接: https://arxiv.org/abs/2502.16612
作者: Mohamed Bayan Kmainasi,Abul Hasnat,Md Arid Hasan,Ali Ezzat Shahroor,Firoj Alam
机构: Qatar University (卡塔尔大学); Blackbird.AI (Blackbird.AI); APAVI.AI (APAVI.AI); University of New Brunswick (新不伦瑞克大学); Qatar Computing Research Institute (卡塔尔计算研究学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: disinformation, misinformation, factuality, harmfulness, fake news, propaganda, hateful meme, multimodality, text, images

点击查看摘要

[NLP-109] Revealing the Prag matic Dilemma for Moral Reasoning Acquisition in Language Models

链接: https://arxiv.org/abs/2502.16600
作者: Guangliang Liu,Lei Jiang,Xitong Zhang,Kristen Marie Johnson
机构: Michigan State University (密歇根州立大学); University of Illinois (伊利诺伊大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-110] Audio-FLAN: A Preliminary Release

链接: https://arxiv.org/abs/2502.16584
作者: Liumeng Xue,Ziya Zhou,Jiahao Pan,Zixuan Li,Shuai Fan,Yinghao Ma,Sitong Cheng,Dongchao Yang,Haohan Guo,Yujia Xiao,Xinsheng Wang,Zixuan Shen,Chuanbo Zhu,Xinshen Zhang,Tianchi Liu,Ruibin Yuan,Zeyue Tian,Haohe Liu,Emmanouil Benetos,Ge Zhang,Yike Guo,Wei Xue
机构: 未知
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multimedia (cs.MM); Audio and Speech Processing (eess.AS)
备注:

点击查看摘要

[NLP-111] he Hidden Strength of Disagreement: Unraveling the Consensus-Diversity Tradeoff in Adaptive Multi-Agent Systems

链接: https://arxiv.org/abs/2502.16565
作者: Zengqing Wu,Takayuki Ito
机构: 未知
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: Source codes are available at this https URL

点击查看摘要

[NLP-112] Analysis of Emotion in Rumour Threads on Social Media

链接: https://arxiv.org/abs/2502.16560
作者: Rui Xing,Boyang Sun,Kun Zhang,Timothy Baldwin,Jey Han Lau
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Social and Information Networks (cs.SI)
备注: 11 pages, 10 figures

点击查看摘要

[NLP-113] Beyond Words: How Large Language Models Perform in Quantitative Management Problem-Solving

链接: https://arxiv.org/abs/2502.16556
作者: Jonathan Kuzmanko
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Applications (stat.AP)
备注: 28 pages, 5 figures, 19 tables

点击查看摘要

[NLP-114] Reasoning About Persuasion: Can LLM s Enable Explainable Propaganda Detection?

链接: https://arxiv.org/abs/2502.16550
作者: Maram Hasanain,Md Arid Hasan,Mohamed Bayan Kmainasi,Elisa Sartori,Ali Ezzat Shahroor,Giovanni Da San Martino,Firoj Alam
机构: Qatar Computing Research Institute(卡塔尔计算研究研究院), Qatar University(卡塔尔大学), University of New Brunswick(新布伦瑞克大学), University of Padova(帕多瓦大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-115] Advanced Chain-of-Thought Reasoning for Parameter Extraction from Documents Using Large Language Models

链接: https://arxiv.org/abs/2502.16540
作者: Hong Cai Chen,Yi Pin Xu,Yang Zhang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Machine Learning (cs.LG)
备注: 9 pages, 9 figures

点击查看摘要

[NLP-116] Multilingual != Multicultural: Evaluating Gaps Between Multilingual Capabilities and Cultural Alignment in LLM s

链接: https://arxiv.org/abs/2502.16534
作者: Jonathan Rystrøm,Hannah Rose Kirk,Scott Hale
机构: Oxford Internet Institute, University of Oxford (牛津大学互联网研究院, 牛津大学), UK
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:

点击查看摘要

[NLP-117] Retrieval-Augmented Fine-Tuning With Preference Optimization For Visual Program Generation

【速读】： 该论文旨在解决利用大语言模型（LLMs）从用户指令生成工业级梯形图（Ladder Diagram, LD）代码准确性较低的问题。论文的关键解决方案是提出了一种基于两阶段训练策略的方法：首先，通过检索增强微调（retrieval-augmented fine-tuning）利用工业视觉编程语言中常见的子程序重复使用；其次，采用直接偏好优化（Direct Preference Optimization, DPO），通过系统生成的偏好对，通过对图编辑操作进行指导，进一步引导模型输出准确的结果。该方法在真实世界LD数据上的实验表明，其在程序级准确性方面比监督微调提高了超过10%，展示了其在提升工业自动化领域的潜力。

链接: https://arxiv.org/abs/2502.16529
作者: Deokhyung Kang,Jeonghun Cho,Yejin Jeon,Sunbin Jang,Minsub Lee,Jawoon Cho,Gary Geunbae Lee
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

Abstract:Visual programming languages (VPLs) allow users to create programs through graphical interfaces, which results in easier accessibility and their widespread usage in various domains. To further enhance this accessibility, recent research has focused on generating VPL code from user instructions using large language models (LLMs). Specifically, by employing prompting-based methods, these studies have shown promising results. Nevertheless, such approaches can be less effective for industrial VPLs such as Ladder Diagram (LD). LD is a pivotal language used in industrial automation processes and involves extensive domain-specific configurations, which are difficult to capture in a single prompt. In this work, we demonstrate that training-based methods outperform prompting-based methods for LD generation accuracy, even with smaller backbone models. Building on these findings, we propose a two-stage training strategy to further enhance VPL generation. First, we employ retrieval-augmented fine-tuning to leverage the repetitive use of subroutines commonly seen in industrial VPLs. Second, we apply direct preference optimization (DPO) to further guide the model toward accurate outputs, using systematically generated preference pairs through graph editing operations. Extensive experiments on real-world LD data demonstrate that our approach improves program-level accuracy by over 10% compared to supervised fine-tuning, which highlights its potential to advance industrial automation.
zh

[NLP-118] Pay Attention to Real World Perturbations! Natural Robustness Evaluation in Machine Reading Comprehension

链接: https://arxiv.org/abs/2502.16523
作者: Yulong Wu,Viktor Schlegel,Riza Batista-Navarro
机构: Department of Computer Science, University of Manchester, United Kingdom (曼彻斯特大学计算机科学系); Imperial College London, Imperial Global Singapore (伦敦帝国学院全球新加坡)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-119] GraphCheck: Breaking Long-Term Text Barriers with Extracted Knowledge Graph-Powered Fact-Checking

链接: https://arxiv.org/abs/2502.16514
作者: Yingjian Chen,Haoran Liu,Yinhong Liu,Rui Yang,Han Yuan,Yanran Fu,Pengyuan Zhou,Qingyu Chen,James Caverlee,Irene Li
机构: University of Tokyo(东京大学); Texas A&M University(德克萨斯农工大学); University of Cambridge(剑桥大学); Duke-NUS Medical School(杜克-新加坡国立大学医学学院); Aarhus University(奥胡斯大学); Yale University(耶鲁大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-120] FanChuan: A Multilingual and Graph-Structured Benchmark For Parody Detection and Analysis

链接: https://arxiv.org/abs/2502.16503
作者: Yilun Zheng,Sha Li,Fangkun Wu,Yang Ziyi,Lin Hongchao,Zhichao Hu,Cai Xinjun,Ziming Wang,Jinxuan Chen,Sitao Luan,Jiahao Xu,Lihui Chen
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-121] Intrinsic Model Weaknesses: How Priming Attacks Unveil Vulnerabilities in Large Language Models

链接: https://arxiv.org/abs/2502.16491
作者: Yuyi Huang,Runzhe Zhan,Derek F. Wong,Lidia S. Chao,Ailin Tao
机构: NLP2CT Lab, Department of Computer and Information Science, University of Macau(澳门大学计算机与信息科学系NLP2CT实验室); The Second Affiliated Hospital, Guangdong Provincial Key Laboratory of Allergy and Clinical Immunology, Guangzhou Medical University(广州医科大学变态反应与临床免疫广东省重点实验室第二附属医院)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-122] All That Glitters is Not Novel: Plagiarism in AI Generated Research

链接: https://arxiv.org/abs/2502.16487
作者: Tarun Gupta,Danish Pruthi
机构: Indian Institute of Science (印度科学研究所)
类目: Computation and Language (cs.CL)
备注: Preprint

点击查看摘要

[NLP-123] A Fine-Tuning Approach for T5 Using Knowledge Graphs to Address Complex Tasks

链接: https://arxiv.org/abs/2502.16484
作者: Xiaoxuan Liao,Binrong Zhu,Jacky He,Guiran Liu,Hongye Zheng,Jia Gao
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-124] owards Fully-Automated Materials Discovery via Large-Scale Synthesis Dataset and Expert-Level LLM -as-a-Judge

链接: https://arxiv.org/abs/2502.16457
作者: Heegyu Kim,Taeyang Jeon,Seungtaek Choi,Jihoon Hong,Dongwon Jeon,Sungbum Cho,Ga-Yeon Baek,Kyung-Won Kwak,Dong-Hee Lee,Sun-Jin Choi,Jisu Bae,Chihoon Lee,Yunseo Kim,Jinsung Park,Hyunsouk Cho
机构: Ajou University (建国大学); Department of Artificial Intelligence (人工智能系); Department of Software and Computer Engineering (软件与计算机工程系); Department of Materials Science and Engineering and Department of Energy Systems Research (材料科学与工程系和能源系统研究系); Division of Materials Science and Engineering, Hanyang University (汉阳大学材料科学与工程系)
类目: Computation and Language (cs.CL)
备注: under review

点击查看摘要

[NLP-125] Contrastive Learning of English Language and Crystal Graphs for Multimodal Representation of Materials Knowledge

链接: https://arxiv.org/abs/2502.16451
作者: Yang Jeong Park,Mayank Kumaran,Chia-Wei Hsu,Elsa Olivetti,Ju Li
机构: Massachusetts Institute of Technology(麻省理工学院); University of Illinois Urbana-Champaign(伊利诺伊大学香槟分校)
类目: Computation and Language (cs.CL)
备注: 24 pages, 14 figure

点击查看摘要

[NLP-126] Make Literature-Based Discovery Great Again through Reproducible Pipelines

链接: https://arxiv.org/abs/2502.16450
作者: Bojan Cestnik,Andrej Kastrin,Boshko Koloski,Nada Lavrač
机构: 未知
类目: Computation and Language (cs.CL)
备注: 13 pages, 3 tables. Accepted for the Symposium on Intelligent Data Analysis (IDA 2025)

点击查看摘要

[NLP-127] VisFactor: Benchmarking Fundamental Visual Cognition in Multimodal Large Language Models

链接: https://arxiv.org/abs/2502.16435
作者: Jen-Tse Huang,Dasen Dai,Jen-Yuan Huang,Youliang Yuan,Xiaoyuan Liu,Wenxuan Wang,Wenxiang Jiao,Pinjia He,Zhaopeng Tu
机构: The Chinese University of Hong Kong; Peking University; The Chinese University of Hong Kong, Shenzhen; Tencent AI Lab
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: Working in Progress

点击查看摘要

[NLP-128] Sequence-level Large Language Model Training with Contrastive Preference Optimization

链接: https://arxiv.org/abs/2502.16433
作者: Zhili Feng,Dhananjay Ram,Cole Hawkins,Aditya Rawal,Jinman Zhao,Sheng Zha
机构: Carnegie Mellon University (卡内基梅隆大学); AWS AI (AWS AI); AGI Foundations (AGI基金会); Amazon (亚马逊)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-129] Ensemble ToT of LLM s and Its Application to Automatic Grading System for Supporting Self-Learning

链接: https://arxiv.org/abs/2502.16399
作者: Yuki Ito,Qiang Ma
机构: Kyoto University(京都大学); Kyoto Institute of Technology(京都工艺纤维大学)
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 33 pages, 25 figures

点击查看摘要

[NLP-130] An Analyst-Inspector Framework for Evaluating Reproducibility of LLM s in Data Science

链接: https://arxiv.org/abs/2502.16395
作者: Qiuhai Zeng,Claire Jin,Xinyue Wang,Yuhan Zheng,Qunhua Li
机构: Pennsylvania State University (宾夕法尼亚州立大学); Carnegie Mellon University (卡内基梅隆大学); International Monetary Fund (国际货币基金组织)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注:

点击查看摘要

[NLP-131] Automatic Detection of Research Values from Scientific Abstracts Across Computer Science Subfields

链接: https://arxiv.org/abs/2502.16390
作者: Hang Jiang,Tal August,Luca Soldaini,Kyle Lo,Maria Antoniak
机构: Massachusetts Institute of Technology(MIT); Allen Institute for AI
类目: Computation and Language (cs.CL); Digital Libraries (cs.DL)
备注: Paper accepted and presented at the 3rd International Conference of Science of Science Innovation (ICSSI 2024) as an extended abstract

点击查看摘要

[NLP-132] oward a Flexible Framework for Linear Representation Hypothesis Using Maximum Likelihood Estimation

链接: https://arxiv.org/abs/2502.16385
作者: Trung Nguyen,Yan Leng
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-133] Instruction-Tuning LLM s for Event Extraction with Annotation Guidelines

链接: https://arxiv.org/abs/2502.16377
作者: Saurabh Srivastava,Sweta Pati,Ziyu Yao
机构: George Mason University (乔治梅森大学)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-134] A generative approach to LLM harmfulness detection with special red flag tokens

链接: https://arxiv.org/abs/2502.16366
作者: Sophie Xhonneux,David Dobre,Mehrnaz Mohfakhami,Leo Schwinn,Gauthier Gidel
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 13 pages, 6 figures

点击查看摘要

[NLP-135] Wrong Answers Can Also Be Useful: PlausibleQA – A Large-Scale QA Dataset with Answer Plausibility Scores SIGIR2025

链接: https://arxiv.org/abs/2502.16358
作者: Jamshid Mozafari,Abdelrahman Abdallah,Bhawna Piryani,Adam Jatowt
机构: University of Innsbruck (因斯布鲁克大学)
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: Submitted to SIGIR 2025

点击查看摘要

[NLP-136] LegalBench.PT: A Benchmark for Portuguese Law

链接: https://arxiv.org/abs/2502.16357
作者: Beatriz Canaverde,Telmo Pessoa Pires,Leonor Melo Ribeiro,André F. T. Martins
机构: Instituto Superior Técnico (葡萄牙技术研究院); Equall; Georgetown University Law Center (乔治城大学法学院)
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-137] Dynamic Coalition Structure Detection in Natural Language-based Interactions AAMAS2025

链接: https://arxiv.org/abs/2502.16339
作者: Abhishek N. Kulkarni,Andy Liu,Jean-Raphael Gaglione,Daniel Fried,Ufuk Topcu
机构: The University of Texas at Austin(德克萨斯大学奥斯汀分校); Carnegie Mellon University(卡内基梅隆大学)
类目: Multiagent Systems (cs.MA); Computation and Language (cs.CL); Computer Science and Game Theory (cs.GT)
备注: Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025)

点击查看摘要

[NLP-138] Iterative Auto-Annotation for Scientific Named Entity Recognition Using BERT-Based Models

链接: https://arxiv.org/abs/2502.16312
作者: Kartik Gupta
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 9 pages

点击查看摘要

[NLP-139] Interrogating LLM design under a fair learning doctrine

链接: https://arxiv.org/abs/2502.16290
作者: Johnny Tian-Zheng Wei,Maggie Wang,Ameya Godbole,Jonathan H. Choi,Robin Jia
机构: University of Southern California(南加州大学); Princeton University(普林斯顿大学)
类目: Computers and Society (cs.CY); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-140] Fine-Tuning Qwen 2.5 3B for Realistic Movie Dialogue Generation

链接: https://arxiv.org/abs/2502.16274
作者: Kartik Gupta
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 5 pages, 2 figures

点击查看摘要

[NLP-141] hinkBench: Dynamic Out-of-Distribution Evaluation for Robust LLM Reasoning

【速读】： 该论文旨在解决大型语言模型（Large Language Models, LLMs）评估中的显著挑战，特别是数据污染和正确答案泄露的问题。解决方案的关键在于引入ThinkBench，这是一种新型的评估框架，通过动态生成OOD数据集来构建分布外数据，从而实现对LLMs推理能力的稳健评估。ThinkBench提供了包含2,912个样本的OOD数据集，并且统一了对推理模型和非推理模型的评估，证明了大多数LLMs的表现远未达到鲁棒性标准，并存在一定程度的数据泄露问题。通过动态生成OOD数据集，ThinkBench有效地提供了对LLMs的可靠评估，减少了数据污染的影响。

链接: https://arxiv.org/abs/2502.16268
作者: Shulin Huang,Linyi Yang,Yan Song,Shuang Chen,Leyang Cui,Ziyu Wan,Qingcheng Zeng,Ying Wen,Kun Shao,Weinan Zhang,Jun Wang,Yue Zhang
机构: Zhejiang University; Westlake University; University College London; Shanghai Jiao Tong University; Northwestern University; Huawei Noah’s Ark Lab
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

Abstract:Evaluating large language models (LLMs) poses significant challenges, particularly due to issues of data contamination and the leakage of correct answers. To address these challenges, we introduce ThinkBench, a novel evaluation framework designed to evaluate LLMs’ reasoning capability robustly. ThinkBench proposes a dynamic data generation method for constructing out-of-distribution (OOD) datasets and offers an OOD dataset that contains 2,912 samples drawn from reasoning tasks. ThinkBench unifies the evaluation of reasoning models and non-reasoning models. We evaluate 16 LLMs and 4 PRMs under identical experimental conditions and show that most of the LLMs’ performance are far from robust and they face a certain level of data leakage. By dynamically generating OOD datasets, ThinkBench effectively provides a reliable evaluation of LLMs and reduces the impact of data contamination.
zh

[NLP-142] Conflicts of Interest in Published NLP Research 2000-2024

链接: https://arxiv.org/abs/2502.16218
作者: Maarten Bosten,Bennett Kleinberg
机构: Tilburg University (蒂尔堡大学); University College London (伦敦大学学院)
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR); Social and Information Networks (cs.SI)
备注:

点击查看摘要

[NLP-143] IPO: Your Language Model is Secretly a Preference Classifier

链接: https://arxiv.org/abs/2502.16182
作者: Shivank Garg,Ayush Singh,Shweta Singh,Paras Chopra
机构: Indian Institute of Technology Roorkee(印度理工学院鲁尔基); Lossfunk
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-144] BiDeV: Bilateral Defusing Verification for Complex Claim Fact-Checking AAAI2025

链接: https://arxiv.org/abs/2502.16181
作者: Yuxuan Liu,Hongda Sun,Wenya Guo,Xinyan Xiao,Cunli Mao,Zhengtao Yu,Rui Yan
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted by AAAI2025

点击查看摘要

[NLP-145] OrderSum: Semantic Sentence Ordering for Extractive Summarization

链接: https://arxiv.org/abs/2502.16180
作者: Taewan Kwon,Sangyong Lee
机构: 未知
类目: Computation and Language (cs.CL)
备注: 29 pages, 12 pages for the main body, 5 figures

点击查看摘要

[NLP-146] Maybe I Should Not Answer That but… Do LLM s Understand The Safety of Their Inputs?

链接: https://arxiv.org/abs/2502.16174
作者: Maciej Chrabąszcz,Filip Szatkowski,Bartosz Wójcik,Jan Dubiński,Tomasz Trzciński
机构: NASK - National Research Institute (国家研究所); Warsaw University of Technology (华沙工业大学); IDEAS NCBR; Jagiellonian University (雅盖隆大学); Tooploox
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注:

点击查看摘要

[NLP-147] Mapping 1000 Language Models via the Log-Likelihood Vector

链接: https://arxiv.org/abs/2502.16173
作者: Momose Oyama,Hiroaki Yamagiwa,Yusuke Takase,Hidetoshi Shimodaira
机构: Kyoto University(京都大学); RIKEN
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-148] EPERM: An Evidence Path Enhanced Reasoning Model for Knowledge Graph Question and Answering

链接: https://arxiv.org/abs/2502.16171
作者: Xiao Long,Liansheng Zhuang,Aodi Li,Minghong Yao,Shafei Wang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-149] OmniParser V2: Structured-Points-of-Thought for Unified Visual Text Parsing and Its Generality to Multimodal Large Language Models

链接: https://arxiv.org/abs/2502.16161
作者: Wenwen Yu,Zhibo Yang,Jianqiang Wan,Sibo Song,Jun Tang,Wenqing Cheng,Yuliang Liu,Xiang Bai
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-150] Number Representations in LLM s: A Computational Parallel to Human Perception

链接: https://arxiv.org/abs/2502.16147
作者: H.V. AlquBoj,Hilal AlQuabeh,Velibor Bojkovic,Tatsuya Hiraoka,Ahmed Oumar El-Shangiti,Munachiso Nwadike,Kentaro Inui
机构: 未知
类目: Computation and Language (cs.CL)
备注: The number line of LLMs

点击查看摘要

[NLP-151] he Law of Knowledge Overshadowing: Towards Understanding Predicting and Preventing LLM Hallucination

链接: https://arxiv.org/abs/2502.16143
作者: Yuji Zhang,Sha Li,Cheng Qian,Jiateng Liu,Pengfei Yu,Chi Han,Yi R. Fung,Kathleen McKeown,Chengxiang Zhai,Manling Li,Heng Ji
机构: 未知
类目: Computation and Language (cs.CL)
备注: 19 pages, 5 figures

点击查看摘要

[NLP-152] Understanding Zero-shot Rare Word Recognition Improvements Through LLM Integration

链接: https://arxiv.org/abs/2502.16142
作者: Haoxuan Wang
机构: 未知
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注:

点击查看摘要

[NLP-153] Chain-of-Description: What I can understand I can put into words

链接: https://arxiv.org/abs/2502.16137
作者: Jiaxin Guo,Daimeng Wei,Zongyao Li,Hengchao Shang,Yuanchang Luo,Hao Yang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-154] PlanGEN: A Multi-Agent Framework for Generating Planning and Reasoning Trajectories for Complex Problem Solving

链接: https://arxiv.org/abs/2502.16111
作者: Mihir Parmar,Xin Liu,Palash Goyal,Yanfei Chen,Long Le,Swaroop Mishra,Hossein Mobahi,Jindong Gu,Zifeng Wang,Hootan Nakhost,Chitta Baral,Chen-Yu Lee,Tomas Pfister,Hamid Palangi
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 30 pages

点击查看摘要

[NLP-155] Be a Multitude to Itself: A Prompt Evolution Framework for Red Teaming

链接: https://arxiv.org/abs/2502.16109
作者: Rui Li,Peiyi Wang,Jingyuan Ma,Di Zhang,Lei Sha,Zhifang Sui
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-156] Echo: A Large Language Model with Temporal Episodic Memory

链接: https://arxiv.org/abs/2502.16090
作者: WenTao Liu,Ruohua Zhang,Aimin Zhou,Feng Gao,JiaLi Liu
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-157] Moving Beyond Medical Exam Questions: A Clinician-Annotated Dataset of Real-World Tasks and Ambiguity in Mental Healthcare

链接: https://arxiv.org/abs/2502.16051
作者: Max Lamparth,Declan Grabb,Amy Franks,Scott Gershan,Kaitlyn N. Kunstman,Aaron Lulla,Monika Drummond Roots,Manu Sharma,Aryan Shrivastava,Nina Vasan,Colleen Waickman
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-158] Inference Computation Scaling for Feature Augmentation in Recommendation Systems

链接: https://arxiv.org/abs/2502.16040
作者: Weihao Liu,Zhaocheng Du,Haiyuan Zhao,Wenbo Zhang,Xiaoyan Zhao,Gang Wang,Zhenhua Dong,Jun Xu
机构: 未知
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-159] Enhancing LLM s for Identifying and Prioritizing Important Medical Jargons from Electronic Health Record Notes Utilizing Data Augmentation

链接: https://arxiv.org/abs/2502.16022
作者: Won Seok Jang,Sharmin Sultana,Zonghai Yao,Hieu Tran,Zhichao Yang,Sunjae Kwon,Hong Yu
机构: 未知
类目: Computation and Language (cs.CL)
备注: 21pages, 5 figures, 4 tables

点击查看摘要

[NLP-160] KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse

链接: https://arxiv.org/abs/2502.16002
作者: Jingbo Yang,Bairu Hou,Wei Wei,Yujia Bao,Shiyu Chang
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-161] Med-gte-hybrid: A contextual embedding transformer model for extracting actionable information from clinical texts

链接: https://arxiv.org/abs/2502.15996
作者: Aditya Kumar,Simon Rauch,Mario Cypko,Oliver Amft
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 11 pages, 4 figures, 2 tables

点击查看摘要

[NLP-162] Sparsity May Be All You Need: Sparse Random Parameter Adaptation

链接: https://arxiv.org/abs/2502.15975
作者: Jesus Rios,Pierre Dognin,Ronny Luss,Karthikeyan N. Ramamurthy
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-163] Forgotten Polygons: Multimodal Large Language Models are Shape-Blind

链接: https://arxiv.org/abs/2502.15969
作者: William Rudman,Michal Golovanesky,Amir Bar,Vedant Palit,Yann LeCun,Carsten Eickhoff,Ritambhara Singh
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-164] Minions: Cost-efficient Collaboration Between On-device and Cloud Language Models

链接: https://arxiv.org/abs/2502.15964
作者: Avanika Narayan,Dan Biderman,Sabri Eyuboglu,Avner May,Scott Linderman,James Zou,Christopher Re
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-165] R3Mem: Bridging Memory Retention and Retrieval via Reversible Compression

链接: https://arxiv.org/abs/2502.15957
作者: Xiaoqiang Wang,Suyuchen Wang,Yun Zhu,Bang Liu
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Work in progress

点击查看摘要

[NLP-166] MMRAG : Multi-Mode Retrieval-Augmented Generation with Large Language Models for Biomedical In-Context Learning

链接: https://arxiv.org/abs/2502.15954
作者: Zaifu Zhan,Jun Wang,Shuang Zhou,Jiawen Deng,Rui Zhang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Submitted to JAMIA

点击查看摘要

[NLP-167] Optimizing Pre-Training Data Mixtures with Mixtures of Data Expert Models

链接: https://arxiv.org/abs/2502.15950
作者: Lior Belenki,Alekh Agarwal,Tianze Shi,Kristina Toutanova
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-168] AutoMedPrompt: A New Framework for Optimizing LLM Medical Prompts Using Textual Gradients

链接: https://arxiv.org/abs/2502.15944
作者: Sean Wu,Michael Koo,Fabien Scalzo,Ira Kurtz
机构: 未知
类目: Computation and Language (cs.CL)
备注: 14 pages

点击查看摘要

[NLP-169] Straight to Zero: Why Linearly Decaying the Learning Rate to Zero Works Best for LLM s ICLR2025

链接: https://arxiv.org/abs/2502.15938
作者: Shane Bergsma,Nolan Dey,Gurpreet Gosal,Gavia Gray,Daria Soboleva,Joel Hestness
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: ICLR 2025

点击查看摘要

[NLP-170] CVE-LLM : Ontology-Assisted Automatic Vulnerability Evaluation Using Large Language Models

链接: https://arxiv.org/abs/2502.15932
作者: Rikhiya Ghosh,Hans-Martin von Stockhausen,Martin Schmitt,George Marica Vasile,Sanjeev Kumar Karn,Oladimeji Farri
机构: 未知
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注: arXiv admin note: substantial text overlap with arXiv:2407.14640

点击查看摘要

[NLP-171] Improving Consistency in Large Language Models through Chain of Guidance

链接: https://arxiv.org/abs/2502.15924
作者: Harsh Raj,Vipul Gupta,Domenic Rosati,Subhabrata Majumdar
机构: 未知
类目: Computation and Language (cs.CL)
备注: Accepted at Transactions of Machine Learning Research (TMLR) 2025

点击查看摘要

[NLP-172] Self-Taught Agent ic Long Context Understanding

链接: https://arxiv.org/abs/2502.15920
作者: Yufan Zhuang,Xiaodong Yu,Jialian Wu,Ximeng Sun,Ze Wang,Jiang Liu,Yusheng Su,Jingbo Shang,Zicheng Liu,Emad Barsoum
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-173] Mind the Gap! Static and Interactive Evaluations of Large Audio Models

链接: https://arxiv.org/abs/2502.15919
作者: Minzhi Li,William Barr Held,Michael J Ryan,Kunat Pipatanakul,Potsawee Manakul,Hao Zhu,Diyi Yang
机构: 未知
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注:

点击查看摘要

[NLP-174] he Esethu Framework: Reimagining Sustainable Dataset Governance and Curation for Low-Resource Languages

链接: https://arxiv.org/abs/2502.15916
作者: Jenalea Rajab,Anuoluwapo Aremu,Everlyn Asiko Chimoto,Dale Dunbar,Graham Morrissey,Fadel Thior,Luandrie Potgieter,Jessico Ojo,Atnafu Lambebo Tonja,Maushami Chetty,Onyothi Nekoto,Pelonomi Moiloa,Jade Abbott,Vukosi Marivate,Benjamin Rosman
机构: 未知
类目: Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-175] Modality-Aware Neuron Pruning for Unlearning in Multimodal Large Language Models

链接: https://arxiv.org/abs/2502.15910
作者: Zheyuan Liu,Guangyao Dou,Xiangchi Yuan,Chunhui Zhang,Zhaoxuan Tan,Meng Jiang
机构: 未知
类目: Computation and Language (cs.CL)
备注: 19 pages

点击查看摘要

[NLP-176] LLM s in Mobile Apps: Practices Challenges and Opportunities

链接: https://arxiv.org/abs/2502.15908
作者: Kimberly Hau,Safwat Hassan,Shurui Zhou
机构: 未知
类目: oftware Engineering (cs.SE); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-177] IPAD: Inverse Prompt for AI Detection – A Robust and Explainable LLM -Generated Text Detector

链接: https://arxiv.org/abs/2502.15902
作者: Zheng Chen,Yushi Feng,Changyang He,Yue Deng,Hongxi Pu,Bo Li
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:

点击查看摘要

[NLP-178] Directional Gradient Projection for Robust Fine-Tuning of Foundation Models ICLR2025

【速读】： 该论文旨在解决在适应大规模基础模型到下游任务的过程中保持其对分布偏移的鲁棒性的问题。现有方法主要集中在基于微调权重与预训练权重之间幅度关系来约束和投影当前模型，这通常需要广泛的超参数调整且有时会导致欠拟合。论文的关键解决方案是提出了Directional Gradient Projection (DiGraP)，这是一种新颖的分层可训练方法，通过结合梯度的方向信息来弥合正则化和多目标优化之间的差距。实验结果表明，DiGraP在图像分类和视觉问答（VQA）任务中均优于现有基线，不仅提高了在分布内（ID）的泛化能力，还增强了分布外（OOD）的鲁棒性。

链接: https://arxiv.org/abs/2502.15895
作者: Chengyue Huang,Junjiao Tian,Brisa Maneechotesuwan,Shivang Chopra,Zsolt Kira
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ICLR 2025

点击查看摘要

Abstract:Robust fine-tuning aims to adapt large foundation models to downstream tasks while preserving their robustness to distribution shifts. Existing methods primarily focus on constraining and projecting current model towards the pre-trained initialization based on the magnitudes between fine-tuned and pre-trained weights, which often require extensive hyper-parameter tuning and can sometimes result in underfitting. In this work, we propose Directional Gradient Projection (DiGraP), a novel layer-wise trainable method that incorporates directional information from gradients to bridge regularization and multi-objective optimization. Besides demonstrating our method on image classification, as another contribution we generalize this area to the multi-modal evaluation settings for robust fine-tuning. Specifically, we first bridge the uni-modal and multi-modal gap by performing analysis on Image Classification reformulated Visual Question Answering (VQA) benchmarks and further categorize ten out-of-distribution (OOD) VQA datasets by distribution shift types and degree (i.e. near versus far OOD). Experimental results show that DiGraP consistently outperforms existing baselines across Image Classfication and VQA tasks with discriminative and generative backbones, improving both in-distribution (ID) generalization and OOD robustness.
zh

[NLP-179] A Close Look at Decomposition-based XAI-Methods for Transformer Language Models

链接: https://arxiv.org/abs/2502.15886
作者: Leila Arras,Bruno Puri,Patrick Kahardipraja,Sebastian Lapuschkin,Wojciech Samek
机构: 未知
类目: Computation and Language (cs.CL)
备注: 9 pages, 3 figures

点击查看摘要

[NLP-180] MutaGReP: Execution-Free Repository-Grounded Plan Search for Code-Use

【速读】： 该论文旨在解决在请求大型语言模型（LLM）完成基于大代码仓库的任务时，如何提供代码仓库中的相关上下文信息。论文的关键解决方案是提出了一种名为MutaGReP的方法，它通过神经树搜索和符号检索器来探索将用户请求分解为基于代码库的自然语言步骤的计划空间。这种方法允许生成的计划仅使用有限的上下文窗口即可达到接近使用整个代码仓库作为上下文的性能水平，从而有效提升在复杂任务上的表现。

链接: https://arxiv.org/abs/2502.15872
作者: Zaid Khan,Ali Farhadi,Ranjay Krishna,Luca Weihs,Mohit Bansal,Tanmay Gupta
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: Project page: this http URL

点击查看摘要

Abstract:When a human requests an LLM to complete a coding task using functionality from a large code repository, how do we provide context from the repo to the LLM? One approach is to add the entire repo to the LLM’s context window. However, most tasks involve only fraction of symbols from a repo, longer contexts are detrimental to the LLM’s reasoning abilities, and context windows are not unlimited. Alternatively, we could emulate the human ability to navigate a large repo, pick out the right functionality, and form a plan to solve the task. We propose MutaGReP (Mutation-guided Grounded Repository Plan Search), an approach to search for plans that decompose a user request into natural language steps grounded in the codebase. MutaGReP performs neural tree search in plan space, exploring by mutating plans and using a symbol retriever for grounding. On the challenging LongCodeArena benchmark, our plans use less than 5% of the 128K context window for GPT-4o but rival the coding performance of GPT-4o with a context window filled with the repo. Plans produced by MutaGReP allow Qwen 2.5 Coder 32B and 72B to match the performance of GPT-4o with full repo context and enable progress on the hardest LongCodeArena tasks. Project page: this http URL
zh

[NLP-181] C3AI: Crafting and Evaluating Constitutions for Constitutional AI

链接: https://arxiv.org/abs/2502.15861
作者: Yara Kyrychenko,Ke Zhou,Edyta Bogucka,Daniele Quercia
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: This has been accepted for the Web Conference 2025

点击查看摘要

[NLP-182] Synthetic vs. Gold: The Role of LLM -Generated Labels and Data in Cyberbullying Detection

链接: https://arxiv.org/abs/2502.15860
作者: Arefeh Kazemi,Sri Balaaji Natarajan Kalaivendan,Joachim Wagner,Hamza Qadeer,Brian Davis
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-183] PPC-GPT : Federated Task-Specific Compression of Large Language Models via Pruning and Chain-of-Thought Distillation

链接: https://arxiv.org/abs/2502.15857
作者: Tao Fan,Guoqiang Ma,Yuanfeng Song,Lixin Fan,Kai Chen,Qiang Yang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-184] Enhancing Domain-Specific Retrieval-Augmented Generation: Synthetic Data Generation and Evaluation using Reasoning Models

链接: https://arxiv.org/abs/2502.15854
作者: Aryan Jadon,Avinash Patil,Shashank Kumar
机构: IEEE(电气电子工程师协会); CA, USA(加利福尼亚州, 美国); University of Florida(佛罗里达大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8 Pages

点击查看摘要

[NLP-185] Control Illusion: The Failure of Instruction Hierarchies in Large Language Models

链接: https://arxiv.org/abs/2502.15851
作者: Yilin Geng,Haonan Li,Honglin Mu,Xudong Han,Timothy Baldwin,Omri Abend,Eduard Hovy,Lea Frermann
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-186] Forecasting Frontier Language Model Agent Capabilities

链接: https://arxiv.org/abs/2502.15850
作者: Govind Pimpale,Axel Højmark,Jérémy Scheurer,Marius Hobbhahn
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[NLP-187] Learning to Reason from Feedback at Test-Time

链接: https://arxiv.org/abs/2502.15771
作者: Yanyang Li,Michael Lyu,Liwei Wang
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: The code is at this https URL

点击查看摘要

[NLP-188] On the Effectiveness of Large Language Models in Automating Categorization of Scientific Texts

链接: https://arxiv.org/abs/2502.15745
作者: Gautam Kishore Shahi,Oliver Hummel
机构: 未知
类目: Computation and Language (cs.CL); Digital Libraries (cs.DL); Machine Learning (cs.LG)
备注:

点击查看摘要

[NLP-189] Detection of LLM -Generated Java Code Using Discretized Nested Bigrams

链接: https://arxiv.org/abs/2502.15740
作者: Timothy Paek,Chilukuri Mohan
机构: 未知
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: This preprint precedes the final peer-reviewed version, which will be published in Springer’s CSCI 2024 proceedings

点击查看摘要

[NLP-190] Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented Generation SIGMOD2025

链接: https://arxiv.org/abs/2502.15734
作者: Shubham Agarwal,Sai Sundaresan,Subrata Mitra,Debabrata Mahapatra,Archit Gupta,Rounak Sharma,Nirmal Joshua Kapu,Tong Yu,Shiv Saini
机构: 未知
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Operating Systems (cs.OS)
备注: Accepted at SIGMOD 2025

点击查看摘要

[NLP-191] own Hall Debate Prompting: Enhancing Logical Reasoning in LLM s through Multi-Persona Interaction

链接: https://arxiv.org/abs/2502.15725
作者: Vivaan Sandwar,Bhav Jain,Rishan Thangaraj,Ishaan Garg,Michael Lam,Kevin Zhu
机构: 未知
类目: Computation and Language (cs.CL)
备注: Accepted to SoCal NLP Symposium 2024

点击查看摘要

[NLP-192] rustDataFilter:Leverag ing Trusted Knowledge Base Data for More Effective Filtering of Unknown Information

链接: https://arxiv.org/abs/2502.15714
作者: Jinghong Zhang,Yidong Cui,Weiling Wang,Xianyou Cheng
机构: 未知
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 12 pages, 8 figures, submitted to IEEE Transactions on Knowledge and Data Engineering

点击查看摘要

[NLP-193] Predicting Liquidity-Aware Bond Yields using Causal GANs and Deep Reinforcement Learning with LLM Evaluation

链接: https://arxiv.org/abs/2502.17011
作者: Jaskaran Singh Walia,Aarush Sinha,Srinitish Srinivasan,Srihari Unnikrishnan
机构: 未知
类目: Computational Finance (q-fin.CP); Computational Engineering, Finance, and Science (cs.CE); Computation and Language (cs.CL); Machine Learning (cs.LG); Portfolio Management (q-fin.PM)
备注:

点击查看摘要

计算机视觉

[CV-0] Fractal Generative Models

链接: https://arxiv.org/abs/2502.17437
作者: Tianhong Li,Qinyi Sun,Lijie Fan,Kaiming He
机构: 未知
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-1] owards Hierarchical Rectified Flow ICLR2025

链接: https://arxiv.org/abs/2502.17436
作者: Yichi Zhang,Yici Yan,Alex Schwing,Zhizhen Zhao
机构: University of Illinois Urbana-Champaign
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: ICLR 2025; Project Page: this https URL

点击查看摘要

[CV-2] GCC: Generative Color Constancy via Diffusing a Color Checker

链接: https://arxiv.org/abs/2502.17435
作者: Chen-Wei Chang,Cheng-De Fan,Chia-Che Chang,Yi-Chen Lo,Yu-Chee Tseng,Jiun-Long Huang,Yu-Lun Liu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL

点击查看摘要

[CV-3] V-HOP: Visuo-Haptic 6D Object Pose Tracking

链接: https://arxiv.org/abs/2502.17434
作者: Hongyu Li,Mingxi Jia,Tuluhan Akbulut,Yu Xiang,George Konidaris,Srinath Sridhar
机构: Brown University; UT Dallas
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-4] FACTR: Force-Attending Curriculum Training for Contact-Rich Policy Learning

链接: https://arxiv.org/abs/2502.17432
作者: Jason Jingzhou Liu,Yulong Li,Kenneth Shaw,Tony Tao,Ruslan Salakhutdinov,Deepak Pathak
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Website at this https URL

点击查看摘要

[CV-5] CLIMB-3D: Continual Learning for Imbalanced 3D Instance Segmentation

【速读】：该论文旨在解决3D实例分割在处理新类别随时间出现且存在自然类别不平衡的现实场景时所面临的挑战。现有方法主要受限于数据集中的类别平衡问题以及缺乏对类别不平衡的关注。为了解决这些问题，论文提出了一种名为CLIMB-3D的框架，该框架结合了示范重放（Exemplar Replay, ER）、知识蒸馏（Knowledge Distillation, KD）和一个新的类别不平衡校正模块（Imbalance Correction, IC）。关键在于通过最小化示范重用，利用知识蒸馏防止遗忘，并支持类别不平衡校正模块以平衡学习稀有类别，从而有效应对类别不平衡问题。实验结果表明，该框架实现了最先进的性能，在平均精度（mAP）上比基线提高了多达16.76%。

链接: https://arxiv.org/abs/2502.17429
作者: Vishal Thengane,Jean Lahoud,Hisham Cholakkal,Rao Muhammad Anwer,Lu Yin,Xiatian Zhu,Salman Khan
机构: University of Surrey (萨里大学); Mohamed bin Zayed University of Artificial Intelligence (穆罕默德·本·扎耶德人工智能大学); Australian National University (澳大利亚国立大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Code: this https URL

点击查看摘要

Abstract:While 3D instance segmentation has made significant progress, current methods struggle to address realistic scenarios where new categories emerge over time with natural class imbalance. This limitation stems from existing datasets, which typically feature few well-balanced classes. Although few datasets include unbalanced class annotations, they lack the diverse incremental scenarios necessary for evaluating methods under incremental settings. Addressing these challenges requires frameworks that handle both incremental learning and class imbalance. However, existing methods for 3D incremental segmentation rely heavily on large exemplar replay, focusing only on incremental learning while neglecting class imbalance. Moreover, frequency-based tuning for balanced learning is impractical in these setups due to the lack of prior class statistics. To overcome these limitations, we propose a framework to tackle both \textbfContinual \textbfLearning and class \textbfImbalance for \textbf3D instance segmentation (\textbfCLIMB-3D). Our proposed approach combines Exemplar Replay (ER), Knowledge Distillation (KD), and a novel Imbalance Correction (IC) module. Unlike prior methods, our framework minimizes ER usage, with KD preventing forgetting and supporting the IC module in compiling past class statistics to balance learning of rare classes during incremental updates. To evaluate our framework, we design three incremental scenarios based on class frequency, semantic similarity, and random grouping that aim to mirror real-world dynamics in 3D environments. Experimental results show that our proposed framework achieves state-of-the-art performance, with an increase of up to 16.76% in mAP compared to the baseline. Code will be available at: \hrefthis https URLthis https URL
zh

[CV-6] Introducing Visual Perception Token into Multimodal Large Language Model

链接: https://arxiv.org/abs/2502.17425
作者: Runpeng Yu,Xinyin Ma,Xinchao Wang
机构: National University of Singapore
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-7] X-Dancer: Expressive Music to Human Dance Video Generation

链接: https://arxiv.org/abs/2502.17414
作者: Zeyuan Chen,Hongyi Xu,Guoxian Song,You Xie,Chenxu Zhang,Xin Chen,Chao Wang,Di Chang,Linjie Luo
机构: ByteDance(字节跳动); UC San Diego(加州大学圣地亚哥分校); University of Southern California(南加州大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-8] Graph-Guided Scene Reconstruction from Images with 3D Gaussian Splatting ICLR2025

链接: https://arxiv.org/abs/2502.17377
作者: Chong Cheng,Gaochao Song,Yiyang Yao,Qinzheng Zhou,Gangjian Zhang,Hao Wang
机构: HKUST(GZ)(香港科技大学(广州)); HKU(香港大学); SCUT(华南理工大学); UC Berkeley(加州大学伯克利分校)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICLR 2025

点击查看摘要

[CV-9] Experimental validation of UAV search and detection system in real wilderness environment

链接: https://arxiv.org/abs/2502.17372
作者: Stella Dumenčić,Luka Lanča,Karlo Jakac,Stefan Ivić
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Robotics (cs.RO); Systems and Control (eess.SY)
备注: 32 pages, 15 figures

点击查看摘要

[CV-10] KV-Edit: Training-Free Image Editing for Precise Background Preservation

链接: https://arxiv.org/abs/2502.17363
作者: Tianrui Zhu,Shiyi Zhang,Jiawei Shao,Yansong Tang
机构: Shenzhen International Graduate School, Tsinghua University (清华大学深圳国际研究生院); Institute of Artificial Intelligence (TeleAI), China Telecom (中国电信人工智能研究院)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project webpage is available at this https URL

点击查看摘要

[CV-11] DIS-CO: Discovering Copyrighted Content in VLMs Training Data

链接: https://arxiv.org/abs/2502.17358
作者: André V. Duarte,Xuandong Zhao,Arlindo L. Oliveira,Lei Li
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-12] Leverag ing Procedural Knowledge and Task Hierarchies for Efficient Instructional Video Pre-training

链接: https://arxiv.org/abs/2502.17352
作者: Karan Samel,Nitish Sontakke,Irfan Essa
机构: Georgia Insitute of Technology (乔治亚理工学院); Google DeepMind (谷歌深思维)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 3 figures

点击查看摘要

[CV-13] AnyTop: Character Animation Diffusion with Any Topology WWW ATC

链接: https://arxiv.org/abs/2502.17327
作者: Inbar Gat,Sigal Raab,Guy Tevet,Yuval Reshef,Amit H. Bermano,Daniel Cohen-Or
机构: Tel Aviv University(特拉维夫大学)
类目: Graphics (cs.GR); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: Video: this https URL , Project page: this https URL , Code: this https URL

点击查看摘要

[CV-14] A novel approach to navigate the taxonomic hierarchy to address the Open-World Scenarios in Medicinal Plant Classification

链接: https://arxiv.org/abs/2502.17289
作者: Soumen Sinha,Tanisha Rana,Rahul Roy
机构: 未知
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-15] GaussianFlowOcc: Sparse and Weakly Supervised Occupancy Estimation using Gaussian Splatting and Temporal Flow

【速读】：该论文旨在解决三维场景占用估计（Occupancy Estimation）的问题，特别是在自动驾驶领域的应用。现有方法主要依赖于密集体素网格（voxel grids），这导致计算和内存需求高。论文提出的关键解决方案是GaussianFlowOcc，它采用稀疏三维高斯表示（sparse 3D Gaussian representation）替代传统密集体素，并基于高斯变换器（Gaussian Transformer）构建高效的模型架构。这种方法通过消除昂贵的三维卷积运算，显著减少了计算和内存需求。此外，GaussianFlowOcc通过在整体网络训练过程中估计每个高斯点的时间流（temporal flow），有效捕捉场景动态。该方法还具备可扩展性，采用弱监督学习，无需昂贵的密集三维体素标注数据。实验结果表明，GaussianFlowOcc在nuScenes数据集上的性能显著优于现有方法，并且推理速度提高了50倍。

链接: https://arxiv.org/abs/2502.17288
作者: Simon Boeder,Fabian Gigengack,Benjamin Risse
机构: Bosch Research; University of Münster (明斯特大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

Abstract:Occupancy estimation has become a prominent task in 3D computer vision, particularly within the autonomous driving community. In this paper, we present a novel approach to occupancy estimation, termed GaussianFlowOcc, which is inspired by Gaussian Splatting and replaces traditional dense voxel grids with a sparse 3D Gaussian representation. Our efficient model architecture based on a Gaussian Transformer significantly reduces computational and memory requirements by eliminating the need for expensive 3D convolutions used with inefficient voxel-based representations that predominantly represent empty 3D spaces. GaussianFlowOcc effectively captures scene dynamics by estimating temporal flow for each Gaussian during the overall network training process, offering a straightforward solution to a complex problem that is often neglected by existing methods. Moreover, GaussianFlowOcc is designed for scalability, as it employs weak supervision and does not require costly dense 3D voxel annotations based on additional data (e.g., LiDAR). Through extensive experimentation, we demonstrate that GaussianFlowOcc significantly outperforms all previous methods for weakly supervised occupancy estimation on the nuScenes dataset while featuring an inference speed that is 50 times faster than current SOTA.
zh

[CV-16] Continuous Wrist Control on the Hannes Prosthesis: a Vision-based Shared Autonomy Framework ICRA2025

【速读】：该论文旨在解决假肢抓取控制中忽视腕部运动的问题，导致用户需要通过肘部、肩部和髋部进行补偿性动作以适应腕部抓取需求。论文的关键解决方案在于提出了一种基于计算机视觉的系统，该系统在共享自主框架下利用用户与自动系统的协作，实现假肢手臂腕部自由度的连续控制，从而促进更自然的接近抓取动作。此方法允许无缝控制假肢腕部跟随目标物体，并最终根据用户意图对其进行定向以完成抓取任务。

链接: https://arxiv.org/abs/2502.17265
作者: Federico Vasile,Elisa Maiettini,Giulia Pasquale,Nicolò Boccardo,Lorenzo Natale
机构: Istituto Italiano di Tecnologia (意大利技术研究院); Dipartimento di Informatica, Bioingegneria, Robotica e Ingegneria dei Sistemi (DIBRIS), University of Genova (热那亚大学信息工程、生物医学工程、机器人工程和系统系); Open University (开放大学)
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Systems and Control (eess.SY)
备注: Accepted to ICRA 2025. Project website: this https URL

点击查看摘要

Abstract:Most control techniques for prosthetic grasping focus on dexterous fingers control, but overlook the wrist motion. This forces the user to perform compensatory movements with the elbow, shoulder and hip to adapt the wrist for grasping. We propose a computer vision-based system that leverages the collaboration between the user and an automatic system in a shared autonomy framework, to perform continuous control of the wrist degrees of freedom in a prosthetic arm, promoting a more natural approach-to-grasp motion. Our pipeline allows to seamlessly control the prosthetic wrist to follow the target object and finally orient it for grasping according to the user intent. We assess the effectiveness of each system component through quantitative analysis and finally deploy our method on the Hannes prosthetic arm. Code and videos: this https URL.
zh

[CV-17] VideoGrain: Modulating Space-Time Attention for Multi-grained Video Editing ICLR2025

链接: https://arxiv.org/abs/2502.17258
作者: Xiangpeng Yang,Linchao Zhu,Hehe Fan,Yi Yang
机构: ReLER Lab, AAII, University of Technology Sydney (UTS); ReLER Lab, CCAI, Zhejiang University (浙江大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICLR 2025, code and demos are available at this https URL

点击查看摘要

[CV-18] CAR-LOAM: Color-Assisted Robust LiDAR Odometry and Mapping

链接: https://arxiv.org/abs/2502.17249
作者: Yufei Lu,Yuetao Li,Zhizhou Jia,Qun Hao,Shaohui Zhang
机构: School of Optics and Photonics, Beijing Institute of Technology(北京理工大学光电学院)
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-19] UNB StepUP: A footStep database for gait analysis and recognition using Underfoot Pressure

链接: https://arxiv.org/abs/2502.17244
作者: Robyn Larracy,Angkoon Phinyomark,Ala Salehi,Eve MacDonald,Saeed Kazemi,Shikder Shafiul Bashar,Aaron Tabor,Erik Scheme
机构: University of New Brunswick, Institute of Biomedical Engineering (新不伦瑞克大学, 生物医学工程研究所), Fredericton, E3B 5A3, Canada
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-20] MegaLoc: One Retrieval to Place Them All

链接: https://arxiv.org/abs/2502.17237
作者: Gabriele Berton,Carlo Masone
机构: Polytechnic of Turin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Tech Report

点击查看摘要

[CV-21] diness Score-Guided Monte Carlo Tree Search for Visual Tabletop Rearrangement

链接: https://arxiv.org/abs/2502.17235
作者: Hogun Kee,Wooseok Oh,Minjae Kang,Hyemin Ahn,Songhwai Oh
机构: Department of Electrical and Computer Engineering and ASRI, Seoul National University (首尔国立大学); Artificial Intelligence Graduate School (AIGS), Ulsan National Institute of Science and Technology (UNIST) (蔚山科学技术研究院)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 9 pages, 8 figures

点击查看摘要

[CV-22] Dimitra: Audio-driven Diffusion model for Expressive Talking Head Generation

链接: https://arxiv.org/abs/2502.17198
作者: Baptiste Chopin,Tashvik Dhamija,Pranav Balaji,Yaohui Wang,Antitza Dantcheva
机构: Université Côte d’Azur (蔚蓝海岸大学), Inria (法国国家信息与自动化研究所), STARS Team (STARS团队), France (法国); Shanghai Artificial Intelligence Laboratory (上海人工智能实验室), China (中国)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages + 2 pages for supplementary material, 2 figures

点击查看摘要

[CV-23] Disentangling Visual Transformers: Patch-level Interpretability for Image Classification

链接: https://arxiv.org/abs/2502.17196
作者: Guillaume Jeanneret,Loïc Simon,Frédéric Jurie
机构: ISIR - Sorbonne Universite (ISIR - 索邦大学); Normandy University, ENSICAEN, UNICAEN, CNRS, GREYC (诺曼底大学, ENSICAEN, UNICAEN, CNRS, GREYC)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-24] A Prag matic Note on Evaluating Generative Models with Fréchet Inception Distance for Retinal Image Synthesis

链接: https://arxiv.org/abs/2502.17160
作者: Yuli Wu,Fucheng Liu,Rüveyda Yilmaz,Henning Konermann,Peter Walter,Johannes Stegmaier
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-25] Parameter Efficient Merging for Multimodal Large Language Models with Complementary Parameter Adaptation

链接: https://arxiv.org/abs/2502.17159
作者: Fanhu Zeng,Haiyang Guo,Fei Zhu,Li Shen,Hao Tang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-26] DICEPTION: A Generalist Diffusion Model for Visual Perceptual Tasks

链接: https://arxiv.org/abs/2502.17157
作者: Canyu Zhao,Mingyu Liu,Huanyi Zheng,Muzhi Zhu,Zhiyue Zhao,Hao Chen,Tong He,Chunhua Shen
机构: Zhejiang University (浙江大学); Shanghai AI Laboratory (上海人工智能实验室)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 29 pages, 20 figures

点击查看摘要

[CV-27] MaxGlaViT: A novel lightweight vision transformer-based approach for early diagnosis of glaucoma stages from fundus images

链接: https://arxiv.org/abs/2502.17154
作者: Mustafa Yurdakul,Kubra Uyar,Sakir Tasdemir
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-28] SFLD: Reducing the content bias for AI-generated Image Detection WACV2025

链接: https://arxiv.org/abs/2502.17105
作者: Seoyeon Gye,Junwon Ko,Hyounguk Shon,Minchan Kwon,Junmo Kim
机构: School of Electrical Engineering, KAIST(电气工程学院, KAIST), South Korea(韩国)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: IEEE/CVF WACV 2025, Oral

点击查看摘要

[CV-29] Improved Diffusion-based Generative Model with Better Adversarial Robustness ICLR2025

链接: https://arxiv.org/abs/2502.17099
作者: Zekun Wang,Mingyang Yi,Shuchen Xue,Zhenguo Li,Ming Liu,Bing Qin,Zhi-Ming Ma
机构: Harbin Institute of Technology (哈尔滨工业大学); Renmin University of China (中国人民大学); Academy of Mathematics and Systems Science, Chinese Academy of Sciences (中国科学院数学与系统科学研究院); University of Chinese Academy of Sciences (中国科学院大学); Huawei Noah’s Ark Lab (华为诺亚方舟实验室)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: ICLR 2025

点击查看摘要

[CV-30] Enhancing Image Matting in Real-World Scenes with Mask-Guided Iterative Refinement

链接: https://arxiv.org/abs/2502.17093
作者: Rui Liu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9pages

点击查看摘要

[CV-31] Shakti-VLMs: Scalable Vision-Language Models for Enterprise AI

链接: https://arxiv.org/abs/2502.17092
作者: Syed Abdul Gaffar Shakhadri,Kruthika KR,Kartik Basavaraj Angadi
机构: SandLogic Technologies Pvt Ltd.
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-32] Imprinto: Enhancing Infrared Inkjet Watermarking for Human and Machine Perception

链接: https://arxiv.org/abs/2502.17089
作者: Martin Feick,Xuxin Tang,Raul Garcia-Martin,Alexandru Luchianov,Roderick Wei Xiao Huang,Chang Xiao,Alexa Siu,Mustafa Doga Dogan
机构: DFKI(德国人工智能研究中心); Saarland University(萨尔兰大学); Virginia Tech(弗吉尼亚理工大学); Universidad Carlos III de Madrid(卡洛斯三世马德里大学); MIT CSAIL(麻省理工学院计算机科学与人工智能实验室); Adobe Research(Adobe研究院)
类目: Human-Computer Interaction (cs.HC); Computer Vision and Pattern Recognition (cs.CV); Emerging Technologies (cs.ET)
备注: 18 pages, 13 figures. To appear in the Proceedings of the 2025 ACM CHI Conference on Human Factors in Computing Systems. this https URL

点击查看摘要

[CV-33] Pleno-Generation: A Scalable Generative Face Video Compression Framework with Bandwidth Intelligence

链接: https://arxiv.org/abs/2502.17085
作者: Bolin Chen,Hanwei Zhu,Shanzhi Yin,Lingyu Zhu,Jie Chen,Ru-Ling Liao,Shiqi Wang,Yan Ye
机构: Department of Computer Science, City University of Hong Kong, Hong Kong(香港城市大学计算机科学系); Alibaba-NTU Global e-Sustainability CorpLab, Nanyang Technological University, SG(新加坡南洋理工大学阿里巴巴-南大全球电子可持续性公司实验室); DAMO Academy, Alibaba Group, Beijing, China(中国阿里巴巴达摩院); DAMO Academy, Alibaba Group, Sunnyvale, CA 94085 USA(美国阿里巴巴达摩院)
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注:

点击查看摘要

[CV-34] VR-Pipe: Streamlining Hardware Graphics Pipeline for Volume Rendering HPCA2025

链接: https://arxiv.org/abs/2502.17078
作者: Junseo Lee,Jaisung Kim,Junyong Park,Jaewoong Sim
机构: Seoul National University (首尔国立大学)
类目: Graphics (cs.GR); Hardware Architecture (cs.AR); Computer Vision and Pattern Recognition (cs.CV)
备注: To appear at the 31st International Symposium on High-Performance Computer Architecture (HPCA 2025)

点击查看摘要

[CV-35] DUNIA: Pixel-Sized Embeddings via Cross-Modal Alignment for Earth Observation Applications

链接: https://arxiv.org/abs/2502.17066
作者: Ibrahim Fayad,Max Zimmer,Martin Schwartz,Philippe Ciais,Fabian Gieseke,Gabriel Belouze,Sarah Brood,Aurelien De Truchis,Alexandre d’Aspremont
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 26 pages, 8 figures

点击查看摘要

[CV-36] SpecDM: Hyperspectral Dataset Synthesis with Pixel-level Semantic Annotations

链接: https://arxiv.org/abs/2502.17056
作者: Wendi Liu,Pei Yang,Wenhui Hong,Xiaoguang Mei,Jiayi Ma
机构: Electronic Information School, Wuhan University (电子信息学院, 武汉大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-37] PointSea: Point Cloud Completion via Self-structure Augmentation

链接: https://arxiv.org/abs/2502.17053
作者: Zhe Zhu,Honghua Chen,Xing He,Mingqiang Wei
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by International Journal of Computer Vision. arXiv admin note: text overlap with arXiv:2307.08492

点击查看摘要

[CV-38] LCV2I: Communication-Efficient and High-Performance Collaborative Perception Framework with Low-Resolution LiDAR

链接: https://arxiv.org/abs/2502.17039
作者: Xinxin Feng,Haoran Sun,Haifeng Zheng,Huacong Chen,Wenqiang Chen
机构: Fujian Key Lab for Intelligent Processing and Wireless Transmission of Media Information (媒体信息智能处理与无线传输福建省重点实验室), College of Physics and Information Engineering (物理与信息工程学院), Fuzhou University (福州大学), Fuzhou 350108, China (中国)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-39] Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems

链接: https://arxiv.org/abs/2502.17019
作者: Maksim Zhdanov,Max Welling,Jan-Willem van de Meent
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-40] Improving the Transferability of Adversarial Examples by Inverse Knowledge Distillation

链接: https://arxiv.org/abs/2502.17003
作者: Wenyuan Wu,Zheng Liu,Yong Chen,Chao Su,Dezhong Peng,Xu Wang
机构: College of Computer Science, Sichuan University, China(四川大学计算机学院，中国); Sichuan Newstrong UHD Video Technology Company Ltd., China(四川新强超高清晰度视频技术有限公司，中国); Institute of Optics and Electronics, Chinese Academy of Sciences, China(中国科学院光电研究院，中国)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-41] An Enhanced Large Language Model For Cross Modal Query Understanding System Using DL-KeyBERT Based CAZSSCL-MPGPT

链接: https://arxiv.org/abs/2502.17000
作者: Shreya Singh
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 26 pages, 7 figures

点击查看摘要

[CV-42] PQDAST: Depth-Aware Arbitrary Style Transfer for Games via Perceptual Quality-Guided Distillation

【速读】：该论文旨在解决在游戏管道中集成任意风格迁移方法的挑战，主要难点在于内存使用和处理速度的要求。论文的关键解决方案是提出了一种名为PQDAST的方法，采用感知质量引导的知识蒸馏框架，并利用FLIP评估器训练压缩模型，从而在有限影响风格化质量的前提下显著减少内存使用和处理时间。此外，通过在合成数据集上进行深度和时间方面的考虑来更好地保留深度信息和细节，进一步将模型注入渲染管道以增强时间一致性并避免后处理效果的减弱。

链接: https://arxiv.org/abs/2502.16996
作者: Eleftherios Ioannou,Steve Maddock
机构: The University of Sheffield (谢菲尔德大学), UK
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 12 pages, 10 figures

点击查看摘要

Abstract:Artistic style transfer is concerned with the generation of imagery that combines the content of an image with the style of an artwork. In the realm of computer games, most work has focused on post-processing video frames. Some recent work has integrated style transfer into the game pipeline, but it is limited to single styles. Integrating an arbitrary style transfer method into the game pipeline is challenging due to the memory and speed requirements of games. We present PQDAST, the first solution to address this. We use a perceptual quality-guided knowledge distillation framework and train a compressed model using the FLIP evaluator, which substantially reduces both memory usage and processing time with limited impact on stylisation quality. For better preservation of depth and fine details, we utilise a synthetic dataset with depth and temporal considerations during training. The developed model is injected into the rendering pipeline to further enforce temporal stability and avoid diminishing post-process effects. Quantitative and qualitative experiments demonstrate that our approach achieves superior performance in temporal consistency, with comparable style transfer quality, to state-of-the-art image, video and in-game methods.
zh

[CV-43] Semantic Neural Radiance Fields for Multi-Date Satellite Data WACV2025

链接: https://arxiv.org/abs/2502.16992
作者: Valentin Wagner,Sebastian Bullinger,Christoph Bodensteiner,Michael Arens
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the CV4EO Workshop at WACV 2025

点击查看摘要

[CV-44] ask-Oriented 6-DoF Grasp Pose Detection in Clutters ICRA2025

链接: https://arxiv.org/abs/2502.16976
作者: An-Lan Wang,Nuo Chen,Kun-Yu Lin,Li Yuan-Ming,Wei-Shi Zheng
机构: Sun Yat-sen University (中山大学); Key Laboratory of Machine Intelligence and Advanced Computing, Ministry of Education, China (中国教育部机器智能与先进计算重点实验室)
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ICRA 2025

点击查看摘要

[CV-45] raFlow: Trajectory Distillation on Pre-Trained Rectified Flow

链接: https://arxiv.org/abs/2502.16972
作者: Zhangkai Wu,Xuhui Fan,Hongyu Wu,Longbing Cao
机构: mq.edu.au(麦考瑞大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-46] Autoregressive Image Generation Guided by Chains of Thought

链接: https://arxiv.org/abs/2502.16965
作者: Miaomiao Cai,Guanjie Wang,Wei Li,Zhijun Tu,Hanting Chen,Shaohui Lin,Jie Hu
机构: University of Science and Technology of China(中国科学技术大学); Huawei Noah’s Ark Lab(华为诺亚方舟实验室); East China Normal University(华东师范大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-47] A survey of datasets for computer vision in agriculture ALT

链接: https://arxiv.org/abs/2502.16950
作者: Nico Heider,Lorenz Gunreben,Sebastian Zürner,Martin Schieck
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 2 figures, published in the proceedings of the 45th GIL Annual Conference (GIL-Jahrestagung), Digitale Infrastrukturen für eine nachhaltige Land-, Forst- und Ernährungswirtschaft (2025)

点击查看摘要

[CV-48] MAD-AD: Masked Diffusion for Unsupervised Brain Anomaly Detection

链接: https://arxiv.org/abs/2502.16943
作者: Farzad Beizaee,Gregory Lodygensky,Christian Desrosiers,Jose Dolz
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注:

点击查看摘要

[CV-49] Gaussian Difference: Find Any Change Instance in 3D Scenes ICASSP2025

链接: https://arxiv.org/abs/2502.16941
作者: Binbin Jiang,Rui Huang,Qingyi Zhao,Yuxiang Zhang
机构: College of Computer Science and Technology, Civil Aviation University of China (中国民航大学计算机科学与技术学院), Tianjin (天津), P.R. China (中华人民共和国天津)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICASSP 2025

点击查看摘要

[CV-50] Multi-Dimensional Quality Assessment for Text-to-3D Assets: Dataset and Model

链接: https://arxiv.org/abs/2502.16915
作者: Kang Fu,Huiyu Duan,Zicheng Zhang,Xiaohong Liu,Xiongkuo Min,Jia Wang,Guangtao Zhai
机构: Shanghai Jiao Tong University(上海交通大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-51] HVIS: A Human-like Vision and Inference System for Human Motion Prediction

链接: https://arxiv.org/abs/2502.16913
作者: Kedi Lyu,Haipeng Chen,Zhenguang Liu,Yifang Yin,Yukang Lin,Yingying Jiao
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-52] SPARC: Score Prompting and Adaptive Fusion for Zero-Shot Multi-Label Recognition in Vision-Language Models

链接: https://arxiv.org/abs/2502.16911
作者: Kevin Miller,Samarth Mishra,Aditya Gangrade,Kate Saenko,Venkatesh Saligrama
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-53] MambaFlow: A Novel and Flow-guided State Space Model for Scene Flow Estimation

链接: https://arxiv.org/abs/2502.16907
作者: Jiehao Luo,Jintao Cheng,Xiaoyu Tang,Qingwen Zhang,Bohuan Xue,Rui Fan
机构: School of Data Science and Engineering, and Xingzhi College, South China Normal University(华南师范大学数据科学与工程学院和兴智学院); School of Electronics and Information Engineering, and Xingzhi College, South China Normal University(华南师范大学电子与信息工程学院和兴智学院); Division of Robotics, Perception, and Learning (RPL), KTH Royal Institute of Technology(皇家理工学院机器人、感知和学习部); College of Electronics & Information Engineering, Shanghai Research Institute for Intelligent Autonomous Systems, the State Key Laboratory of Intelligent Autonomous Systems, and Frontiers Science Center for Intelligent Autonomous Systems, Tongji University(同济大学电子与信息工程学院、上海自主智能无人系统科学中心、智能自主系统重点实验室和自主智能前沿科学中心)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-54] Culture-TRIP: Culturally-Aware Text-to-Image Generation with Iterative Prompt Refinment NAACL2025

链接: https://arxiv.org/abs/2502.16902
作者: Suchae Jeong,Inseong Choi,Youngsik Yun,Jihie Kim
机构: Department of Computer Science and Engineering, Dongguk University (东国大学计算机工程系); Department of Computer Science and Artificial Intelligence, Dongguk University (东国大学计算机科学与人工智能系)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 31 pages, 23 figures, Accepted by NAACL 2025

点击查看摘要

[CV-55] Unveiling Institution-Specific Bias in Pathology Foundation Models: Detriments Causes and Potential Solutions

【速读】：该论文旨在解决病理学基础模型（Pathology Foundation Models, PFMs）在提取特征过程中受到诊断无关信息污染的问题，特别是机构特定特征的影响。这种污染会导致虚假相关性，削弱模型在实际临床应用中的泛化能力。论文的关键解决方案是提出一种简单而有效的策略来减轻无关信息的影响，从而提升模型的泛化性能。

链接: https://arxiv.org/abs/2502.16889
作者: Weiping Lin,Shen Liu,Runchen Zhu,Liansheng Wang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages,1 figure,14 tables

点击查看摘要

Abstract:Pathology foundation models (PFMs) extract valuable discriminative features from images for downstream clinical tasks. PFMs have simplified the development of deep learning models, effectively leveraging prior knowledge to improve diagnostic accuracy in diverse scenarios. However, we find that PFMs sometimes struggle with certain challenges. Specifically, features extracted by PFMs are often contaminated by diagnosis-irrelevant information, i.e., institution-specific features associated with the images. This contamination can lead to spurious correlations, undermining the models’ generalization ability when applied in real-world clinical settings. In this work, we first reveal the issue of feature contamination in PFMs, demonstrate the presence of institution-specific features, thoroughly investigate its negative impacts, analyze the underlying causes, and provide insights into potential solutions. Specifically, we find that institution-specific information is embedded in pathological images and can be readily captured by current PFMs. Through extensive experiments, we demonstrate the detrimental impact of this irrelevant information, particularly in out-of-distribution (OOD) settings, where reliance on contaminated features leads to significant performance degradation. This indicates that the models are being misled by non-diagnostic information. We further delve into the reasons PFMs extract such institution-specific information and validate our findings. Finally, we propose a simple yet effective solution to mitigate the influence of irrelevant information. This study is not intended to criticize existing PFMs, as they have indeed greatly advanced the development of computational pathology. our aim is to inspire future research to focus on innovative training strategies, rather than relying exclusively on scaling laws, to realize more generalized PFMs.
zh

[CV-56] Mitigating Hallucinations in Diffusion Models through Adaptive Attention Modulation

链接: https://arxiv.org/abs/2502.16872
作者: Trevine Oorloff,Yaser Yacoob,Abhinav Shrivastava
机构: University of Maryland(马里兰大学),
College Park, Maryland, USA
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-57] A Survey of fMRI to Image Reconstruction

链接: https://arxiv.org/abs/2502.16861
作者: Weiyu Guo,Guoying Sun,JianXiang He,Tong Shao,Shaoguang Wang,Ziyang Chen,Meisheng Hong,Ying Sun,Hui Xiong
机构: Thrust of Artificial Intelligence, The Hong Kong University of Science and Technology (Guangzhou)(人工智能推力，香港科技大学（广州）); College of Computer Science and Technology, Harbin Institute of Technology (Shenzhen)(计算机科学与技术学院，哈尔滨工业大学（深圳）); School of Control Science and Engineering, Shandong University(控制科学与工程学院，山东大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-58] Characterizing Structured versus Unstructured Environments based on Pedestrians and Vehicles Motion Trajectories

链接: https://arxiv.org/abs/2502.16847
作者: Mahsa Golchoubian,Moojan Ghafurian,Nasser Lashgarian Azad,Kerstin Dautenhahn
机构: University of Waterloo(滑铁卢大学)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-59] Exploring Causes and Mitigation of Hallucinations in Large Vision Language Models

链接: https://arxiv.org/abs/2502.16842
作者: Yaqi Sun,Kyohei Atarashi,Koh Takeuchi,Hisashi Kashima
机构: Kyoto University (京都大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 6 figures

点击查看摘要

[CV-60] Fair Foundation Models for Medical Image Analysis: Challenges and Perspectives

链接: https://arxiv.org/abs/2502.16841
作者: Dilermando Queiroz,Anderson Carlos,André Anjos,Lilian Berton
机构: Federal University of São Paulo(圣保罗联邦大学); Federal Institute of Goiás(戈亚斯联邦技术学院); Idiap Research Institute(伊迪亚普研究学院)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-61] FedBM: Stealing Knowledge from Pre-trained Language Models for Heterogeneous Federated Learning

链接: https://arxiv.org/abs/2502.16832
作者: Meilu Zhu,Qiushi Yang,Zhifan Gao,Yixuan Yuan,Jun Liu
机构: City University of Hong Kong; Sun Yat-sen University; Chinese University of Hong Kong (香港中文大学); The University of Hong Kong (香港大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by MedIA 2025

点击查看摘要

[CV-62] Noise2Score3D:Unsupervised Tweedies Approach for Point Cloud Denoising

链接: https://arxiv.org/abs/2502.16826
作者: Xiangbin Wei
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-63] CLIP-SENet: CLIP-based Semantic Enhancement Network for Vehicle Re-identification

链接: https://arxiv.org/abs/2502.16815
作者: Liping Lu,Zihao Fu,Duanfeng Chu,Wei Wang,Bingrong Xu
机构: School of Computer Science and Artificial Intelligence, Wuhan University of Technology, China(武汉理工大学计算机科学与人工智能学院，中国); Intelligent Transportation Systems Research Center, Wuhan University of Technology, China(武汉理工大学智能交通运输系统研究中心，中国); School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-sen University, China(中山大学深圳校区网络科学与技术学院，中国); School of Automation, Wuhan University of Technology, China(武汉理工大学自动化学院，中国)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-64] CRTrack: Low-Light Semi-Supervised Multi-object Tracking Based on Consistency Regularization

链接: https://arxiv.org/abs/2502.16809
作者: Zijing Zhao,Jianlong Yu,Lin Zhang,Shunli Zhang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-65] Hierarchical Semantic Compression for Consistent Image Semantic Restoration

链接: https://arxiv.org/abs/2502.16799
作者: Shengxi Li,Zifu Zhang,Mai Xu,Lai Jiang,Yufan Liu,Ce Zhu
机构: School of Electronic and Information Engineering, Beihang University(北京航空航天大学); State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所); School of Electronic and Information Engineering, University of Electronic Science and Technology of China(电子科技大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 12 figures

点击查看摘要

[CV-66] Continuous Patch Stitching for Block-wise Image Compression

链接: https://arxiv.org/abs/2502.16795
作者: Zifu Zhang,Shengxi Li,Henan Liu,Mai Xu,Ce Zhu
机构: School of Electronic and Information Engineering, Beihang University (北京航空航天大学); School of Electronic and Information Engineering, University of Electronic Science and Technology of China (电子科技大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: 5 pages, 8 figures

点击查看摘要

[CV-67] SwimVG: Step-wise Multimodal Fusion and Adaption for Visual Grounding

链接: https://arxiv.org/abs/2502.16786
作者: Liangtao Shi,Ting Liu,Xiantao Hu,Yue Hu,Quanjun Yin,Richang Hong
机构: Key Laboratory of Knowledge Engineering with Big Data, Hefei University of Technology, Hefei 230009, China (合肥工业大学大数据知识工程实验室);
Ministry of Education and School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230009, China (教育部计算机科学与信息工程学院, 合肥工业大学);
School of systems engineering, National University of Defense Technology, Changsha, Hunan Province, 410073, China (国防科技大学系统工程学院);
Department of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210014, China (南京理工大学计算机科学与工程学院)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 7 this http URL code is available at this https URL

点击查看摘要

[CV-68] Unposed Sparse Views Room Layout Reconstruction in the Age of Pretrain Model

链接: https://arxiv.org/abs/2502.16779
作者: Yaxuan Huang,Xili Dai,Jianan Wang,Xianbiao Qi,Yixing Yuan,Xiangyu Yue
机构: Hong Kong Center for Construction Robotics, The Hong Kong University of Science and Technology (香港科技大学); The Hong Kong University of Science and Technology (Guangzhou) (香港科技大学（广州）); Astribot (未知); Intellifusion Inc. (未知); MMLab, The Chinese University of Hong Kong (香港中文大学MMLab)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-69] A Transformer-in-Transformer Network Utilizing Knowledge Distillation for Image Recognition

链接: https://arxiv.org/abs/2502.16762
作者: Dewan Tauhid Rahman,Yeahia Sarker,Antar Mazumder,Md. Shamim Anower
机构: Department of Computer Science, University of Miami (迈阿密大学); Department of Mechatronics Engineering, Rajshahi University of Engineering & Technology (鲁特科技大学); Department of Electrical & Electronic Engineering, Rajshahi University of Engineering & Technology (鲁特科技大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-70] GS-TransUNet: Integrated 2D Gaussian Splatting and Transformer UNet for Accurate Skin Lesion Analysis

链接: https://arxiv.org/abs/2502.16748
作者: Anand Kumar,Kavinder Roghit Kanthen,Josna John
机构: University of California, San Diego(加州大学圣地亚哥分校)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 7 figures, SPIE Medical Imaging 2025

点击查看摘要

[CV-71] DOSE3 : Diffusion-based Out-of-distribution detection on SE(3) trajectories

链接: https://arxiv.org/abs/2502.16725
作者: Hongzhe Cheng,Tianyou Zheng,Tianyi Zhang,Matthew Johnson-Roberson,Weiming Zhi
机构: Robotics Institute (机器人研究所); Carnegie Mellon University (卡内基梅隆大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-72] Interpretable Retinal Disease Prediction Using Biology-Informed Heterogeneous Graph Representations

链接: https://arxiv.org/abs/2502.16697
作者: Laurin Lux,Alexander H. Berger,Maria Romeo Tricas,Alaa E. Fayed,Sobha Sivaprasada,Linus Kreitner,Jonas Weidner,Martin J. Menten,Daniel Rueckert,Johannes C. Paetzold
机构: J. C. P. and D. R. are affiliated with BioMedIA, Imperial College London, United Kingdom (英国); L.L., A.B., M.R.T., L. K., J.W., M. J. M., and D.R. are with the Lab for AI in Medicine, Klinikum rechts der Isar (MRI), Technical University of Munich (TUM), Germany; L.L., M.J.M., and D.R. are with the Munich Center for Machine Learning (MCML), Munich, Germany; S.S. is with University College London, UK (英国); A. E. F. is with the Dept. of Ophthalmology, Kasr Al-Ainy School of Medicine, Cairo University, Egypt (埃及)
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-73] AeroReformer: Aerial Referring Transformer for UAV-based Referring Image Segmentation

链接: https://arxiv.org/abs/2502.16680
作者: Rui Li
机构: Intelligent Control & Smart Energy (ICSE) Research Group, School of Engineering, University of Warwick (华威大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-74] VPNeXt – Rethinking Dense Decoding for Plain Vision Transformer

【速读】：该论文旨在解决两个主要问题：一是是否有必要使用复杂的Transformer掩码解码器架构来获得良好的表示；二是Plain Vision Transformer (ViT) 是否真的需要依赖模拟金字塔特征进行上采样。为了解决这些问题，VPNeXt引入了Visual Context Replay (VCR) 和 ViTUp模块。VCR高效地实现了与Transformer解码器类似的效果，而ViTUp模块则充分利用了被忽视的ViT真实金字塔特征，从而在语义分割任务中取得了优于模拟金字塔特征的上采样结果。这些改进使得VPNeXt在VOC2012数据集上显著超越了长期存在的mIoU壁垒，达到了最新的技术水平。

链接: https://arxiv.org/abs/2502.16654
作者: Xikai Tang,Ye Huang,Guangqiang Yin,Lixin Duan
机构: School of Information and Software Engineering, University of Electronic Science and Technology of China(电子科技大学信息与软件工程学院); Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China(中国电子科技大学深圳先进研究院)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Tech report

点击查看摘要

Abstract:We present VPNeXt, a new and simple model for the Plain Vision Transformer (ViT). Unlike the many related studies that share the same homogeneous paradigms, VPNeXt offers a fresh perspective on dense representation based on ViT. In more detail, the proposed VPNeXt addressed two concerns about the existing paradigm: (1) Is it necessary to use a complex Transformer Mask Decoder architecture to obtain good representations? (2) Does the Plain ViT really need to depend on the mock pyramid feature for upsampling? For (1), we investigated the potential underlying reasons that contributed to the effectiveness of the Transformer Decoder and introduced the Visual Context Replay (VCR) to achieve similar effects efficiently. For (2), we introduced the ViTUp module. This module fully utilizes the previously overlooked ViT real pyramid feature to achieve better upsampling results compared to the earlier mock pyramid feature. This represents the first instance of such functionality in the field of semantic segmentation for Plain ViT. We performed ablation studies on related modules to verify their effectiveness gradually. We conducted relevant comparative experiments and visualizations to show that VPNeXt achieved state-of-the-art performance with a simple and effective design. Moreover, the proposed VPNeXt significantly exceeded the long-established mIoU wall/barrier of the VOC2012 dataset, setting a new state-of-the-art by a large margin, which also stands as the largest improvement since 2015.
zh

[CV-75] Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration

【速读】：该论文旨在解决开放词汇3D场景理解中的问题，特别是通过利用3D高斯点阵(3D Gaussian Splatting)实现更全面的理解。其关键在于一种语言特征注册技术，该技术将与语言对齐的CLIP嵌入直接关联到3D高斯分布上，而非依赖于渲染过程。此外，通过使用在大规模图像数据上训练的产品量化(Product Quantization)技术来紧凑表示这些嵌入，从而避免了每场景优化的需求。实验表明，该方法在3D感知基准测试，如开放词汇3D语义分割、3D物体定位及3D物体选择任务中显著优于现有方法。

链接: https://arxiv.org/abs/2502.16652
作者: Kim Jun-Seong,GeonU Kim,Kim Yu-Ji,Yu-Chiang Frank Wang,Jaesung Choe,Tae-Hyun Oh
机构: POSTECH; NVIDIA; KAIST
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 20 pages

点击查看摘要

Abstract:We introduce Dr. Splat, a novel approach for open-vocabulary 3D scene understanding leveraging 3D Gaussian Splatting. Unlike existing language-embedded 3DGS methods, which rely on a rendering process, our method directly associates language-aligned CLIP embeddings with 3D Gaussians for holistic 3D scene understanding. The key of our method is a language feature registration technique where CLIP embeddings are assigned to the dominant Gaussians intersected by each pixel-ray. Moreover, we integrate Product Quantization (PQ) trained on general large-scale image data to compactly represent embeddings without per-scene optimization. Experiments demonstrate that our approach significantly outperforms existing approaches in 3D perception benchmarks, such as open-vocabulary 3D semantic segmentation, 3D object localization, and 3D object selection tasks. For video results, please visit : this https URL
zh

[CV-76] Automatic Joint Structured Pruning and Quantization for Efficient Neural Network Training and Compression

链接: https://arxiv.org/abs/2502.16638
作者: Xiaoyi Qu,David Aponte,Colby Banbury,Daniel P. Robinson,Tianyu Ding,Kazuhito Koishida,Ilya Zharkov,Tianyi Chen
机构: Microsoft(微软); Lehigh University(莱斯大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-77] AdverX-Ray: Ensuring X-Ray Integrity Through Frequency-Sensitive Adversarial VAEs

链接: https://arxiv.org/abs/2502.16610
作者: Francisco Caetano,Christiaan Viviers,Lena Filatova,Peter H. N. de With,Fons van der Sommen
机构: Eindhoven University of Technology(埃因霍温科技大学); Philips IGT(飞利浦IGT); Eindhoven University of Technology(埃因霍温科技大学); Eindhoven University of Technology(埃因霍温科技大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: SPIE Medical Imaging 2025 Runner-up 2025 Robert F. Wagner All-Conference Best Student Paper Award

点击查看摘要

[CV-78] VidLBEval: Benchmarking and Mitigating Language Bias in Video-Involved LVLMs

【速读】：该论文旨在解决现有包含视频的大型视觉语言模型（Large Vision-Language Models, LVLMs）中存在的语言偏见问题，即模型倾向于过度依赖语言信息而忽视视频内容，从而导致错误响应。为了解决这一问题，论文提出了一种名为多分支对比解码（Multi-branch Contrastive Decoding, MCD）的方法，通过引入两个专家分支来同时对抗由业余文本分支可能产生的语言偏见。实验结果表明，MCD能够有效缓解语言偏见问题，并在不进行额外再训练或改变模型架构的情况下保持模型的通用能力。

链接: https://arxiv.org/abs/2502.16602
作者: Yiming Yang,Yangyang Guo,Hui Lu,Yan Wang
机构: Nanyang Technological University; National University of Singapore; Sichuan University
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

Abstract:Recently, Large Vision-Language Models (LVLMs) have made significant strides across diverse multimodal tasks and benchmarks. This paper reveals a largely under-explored problem from existing video-involved LVLMs - language bias, where models tend to prioritize language over video and thus result in incorrect responses. To address this research gap, we first collect a Video Language Bias Evaluation Benchmark, which is specifically designed to assess the language bias in video-involved LVLMs through two key tasks: ambiguous video contrast and interrogative question probing. Accordingly, we design accompanied evaluation metrics that aim to penalize LVLMs being biased by language. In addition, we also propose Multi-branch Contrastive Decoding (MCD), introducing two expert branches to simultaneously counteract language bias potentially generated by the amateur text-only branch. Our experiments demonstrate that i) existing video-involved LVLMs, including both proprietary and open-sourced, are largely limited by the language bias problem; ii) our MCD can effectively mitigate this issue and maintain general-purpose capabilities in various video-involved LVLMs without any additional retraining or alteration to model architectures.
zh

[CV-79] SelaVPR: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition

链接: https://arxiv.org/abs/2502.16601
作者: Feng Lu,Tong Jin,Xiangyuan Lan,Lijun Zhang,Yunpeng Liu,Yaowei Wang,Chun Yuan
机构: Tsinghua Shenzhen International Graduate School, Tsinghua University (清华大学深圳国际研究生院); Pengcheng Laboratory (鹏城实验室); Shenyang Institute of Automation, Chinese Academy of Sciences (中国科学院沈阳自动化研究所); Pazhou Laboratory (黄埔)(琶洲实验室); Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences (中国科学院重庆绿色智能技术研究院)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-80] Improving Monocular Visual-Inertial Initialization with Structureless Visual-Inertial Bundle Adjustment ICRA2025

链接: https://arxiv.org/abs/2502.16598
作者: Junlin Song,Antoine Richard,Miguel Olivares-Mendez
机构: Space Robotics (SpaceR) Research Group (太空机器人研究小组), Int. Centre for Security, Reliability and Trust (SnT) (安全、可靠性和信任国际中心), University of Luxembourg (卢森堡大学), Luxembourg (卢森堡).
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ICRA 2025

点击查看摘要

[CV-81] Co-MTP: A Cooperative Trajectory Prediction Framework with Multi-Temporal Fusion for Autonomous Driving ICRA2025

链接: https://arxiv.org/abs/2502.16589
作者: Xinyu Zhang,Zewei Zhou,Zhaoyi Wang,Yangjie Ji,Yanjun Huang,Hong Chen
机构: Tongji University (同济大学), School of Automotive Studies (汽车学院), Shanghai, China (中国)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 8 pages, 3 figures, ICRA 2025

点击查看摘要

[CV-82] Multimodal Large Language Models for Text-rich Image Understanding: A Comprehensive Review

链接: https://arxiv.org/abs/2502.16586
作者: Pei Fu,Tongkun Guan,Zining Wang,Zhentao Guo,Chen Duan,Hao Sun,Boming Chen,Jiayao Ma,Qianyi Jiang,Kai Zhou,Junfeng Luo
机构: Meituan(美团); Shanghai Jiao Tong University(上海交通大学); Beijing Institute of Technology(北京理工大学); MAIS & NLPR, Institute of Automation, Chinese Academy of Sciences(中科院自动化所模式识别国家重点实验室)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-83] Anatomical grounding pre-training for medical phrase grounding

链接: https://arxiv.org/abs/2502.16585
作者: Wenjun Zhang,Shakes Chandra,Aaron Nicolson
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-84] Efficient 4D Gaussian Stream with Low Rank Adaptation

链接: https://arxiv.org/abs/2502.16575
作者: Zhenhuan Liu,Shuai Liu,Yidong Lu,Yirui Chen,Jie Yang,Wei Liu
机构: Dept. of Automation (自动化系), Shanghai Jiao Tong University (上海交通大学), Shanghai (上海), China (中国)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 3 pages draft

点击查看摘要

[CV-85] Entropy-Lens: The Information Signature of Transformer Computations

链接: https://arxiv.org/abs/2502.16570
作者: Riccardo Ali,Francesco Caso,Christopher Irwin,Pietro Liò
机构: University of Cambridge(剑桥大学); Sapienza University of Rome (罗马一大); University of Eastern Piedmont(东皮埃蒙特大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-86] Benchmarking Online Object Trackers for Underwater Robot Position Locking Applications

链接: https://arxiv.org/abs/2502.16569
作者: Ali Safa,Waqas Aman,Ali Al-Zawqari,Saif Al-Kuwari
机构: Hamad Bin Khalifa University (哈马德·本·哈利法大学); Vrije Universiteit Brussels (布鲁塞尔自由大学)
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-87] Multi-Target Federated Backdoor Attack Based on Feature Aggregation

链接: https://arxiv.org/abs/2502.16545
作者: Lingguag Hao,Kuangrong Hao,Bing Wei,Xue-song Tang
机构: Donghua University (东华大学); College of Information Science and Technology, Donghua University (东华大学信息科学与技术学院); Engineering Research Center of Digitized Textile & Apparel Technology, Ministry of Education (教育部数字化纺织服装技术工程研究中心)
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-88] EDocNet: Efficient Datasheet Layout Analysis Based on Focus and Global Knowledge Distillation

链接: https://arxiv.org/abs/2502.16541
作者: Hong Cai Chen,Longchang Wu,Yang Zhang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 9 pages, 6 figures

点击查看摘要

[CV-89] Color Information-Based Automated Mask Generation for Detecting Underwater Atypical Glare Areas

链接: https://arxiv.org/abs/2502.16538
作者: Mingyu Jeon,Yeonji Paeng,Sejin Lee
机构: Kongju National University (公州国立大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 7pages, 6 figures

点击查看摘要

[CV-90] Deep unrolling for learning optimal spatially varying regularisation parameters for Total Generalised Variation

链接: https://arxiv.org/abs/2502.16532
作者: Thanh Trung Vu,Andreas Kofler,Kostas Papafitsoros
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Optimization and Control (math.OC)
备注:

点击查看摘要

[CV-91] Subpixel Edge Localization Based on Converted Intensity Summation under Stable Edge Region

链接: https://arxiv.org/abs/2502.16502
作者: Yingyuan Yang,Guoyuan Liang,Xianwen Wang,Kaiming Wang,Can Wang,Xiaojun Wu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-92] runk-branch Contrastive Network with Multi-view Deformable Aggregation for Multi-view Action Recognition

链接: https://arxiv.org/abs/2502.16493
作者: Yingyuan Yang,Guoyuan Liang,Can Wang,Xiaojun Wu
机构: Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences (深圳先进技术研究院, 中国科学院); University of Chinese Academy of Sciences (中国科学院大学); Guangdong Provincial Key Laboratory of Robotics and Intelligent System (广东省机器人与智能系统重点实验室); School of Mechanical Engineering and Automation, Harbin Institute of Technology at Shenzhen (哈尔滨工业大学（深圳）机械工程与自动化学院)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-93] On Computational Limits of FlowAR Models: Expressivity and Efficiency

链接: https://arxiv.org/abs/2502.16490
作者: Chengyue Gong,Yekun Ke,Xiaoyu Li,Yingyu Liang,Zhizhou Sha,Zhenmei Shi,Zhao Song
机构: The University of Texas at Austin(德克萨斯大学奥斯汀分校); University of New South Wales(新南威尔士大学); The University of Hong Kong(香港大学); University of Wisconsin-Madison(威斯康星大学麦迪逊分校); Tsinghua University(清华大学); The Simons Institute for the Theory of Computing at UC Berkeley(伯克利加州大学西蒙斯计算理论研究所)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Complexity (cs.CC); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-94] Geometry-Aware 3D Salient Object Detection Network

链接: https://arxiv.org/abs/2502.16488
作者: Chen Wang,Liyuan Zhang,Le Hui,Qi Liu,Yuchao Dai
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-95] MQADet: A Plug-and-Play Paradigm for Enhancing Open-Vocabulary Object Detection via Multimodal Question Answering

链接: https://arxiv.org/abs/2502.16486
作者: Caixiong Li,Xiongwei Zhao,Jinhang Zhang,Xing Zhang,Zhou Wu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-96] Drag en3D: Multiview Geometry Consistent 3D Gaussian Generation with Drag -Based Control

链接: https://arxiv.org/abs/2502.16475
作者: Jinbo Yan,Alan Zhao,Yixin Hu
机构: Tencent(腾讯) China; Tencent(腾讯) China; Tencent America(腾讯美国) USA
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-97] Feature Space Perturbation: A Panacea to Enhanced Transferability Estimation

链接: https://arxiv.org/abs/2502.16471
作者: Prafful Kumar Khoba,Zijian Wang,Chetan Arora,Mahsa Baktashmotlagh
机构: The University of Queensland (昆士兰大学); Indian Institute of Technology Delhi (印度理工学院德里分校); UQ–IITD Research Academy (UQ–IITD 研究院)
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-98] Cross-domain Few-shot Object Detection with Multi-modal Textual Enrichment

【速读】：该论文致力于解决跨域多模态少样本目标检测（Cross-Domain Multi-Modal Few-Shot Object Detection, CDMM-FSOD）中的性能退化问题。为应对这一挑战，论文提出的关键解决方案包括：(i) 多模态特征聚合模块，用于对齐视觉与语言特征嵌入，确保多模态之间的整合一致性；(ii) 丰富的文本语义校正模块，通过双向文本特征生成来优化多模态特征对齐，从而增强对语言及其在目标检测中应用的理解。这些组件共同作用以实现有效的领域适应。

链接: https://arxiv.org/abs/2502.16469
作者: Zeyu Shangguan,Daniel Seita,Mohammad Rostami
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: arXiv admin note: substantial text overlap with arXiv:2403.16188

点击查看摘要

Abstract:Advancements in cross-modal feature extraction and integration have significantly enhanced performance in few-shot learning tasks. However, current multi-modal object detection (MM-OD) methods often experience notable performance degradation when encountering substantial domain shifts. We propose that incorporating rich textual information can enable the model to establish a more robust knowledge relationship between visual instances and their corresponding language descriptions, thereby mitigating the challenges of domain shift. Specifically, we focus on the problem of Cross-Domain Multi-Modal Few-Shot Object Detection (CDMM-FSOD) and introduce a meta-learning-based framework designed to leverage rich textual semantics as an auxiliary modality to achieve effective domain adaptation. Our new architecture incorporates two key components: (i) A multi-modal feature aggregation module, which aligns visual and linguistic feature embeddings to ensure cohesive integration across modalities. (ii) A rich text semantic rectification module, which employs bidirectional text feature generation to refine multi-modal feature alignment, thereby enhancing understanding of language and its application in object detection. We evaluate the proposed method on common cross-domain object detection benchmarks and demonstrate that it significantly surpasses existing few-shot object detection approaches.
zh

[CV-99] Iterative Flow Matching – Path Correction and Gradual Refinement for Enhanced Generative Modeling

链接: https://arxiv.org/abs/2502.16445
作者: Eldad Haber,Shadab Ahamed,Md. Shahriar Rahim Siddiqui,Niloufar Zakariaei,Moshe Eliasof
机构: The University of British Columbia(英属哥伦比亚大学), Vancouver, BC, Canada; University of Cambridge(剑桥大学), Cambridge, United Kingdom
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML)
备注: 17 pages, 8 figures

点击查看摘要

[CV-100] Visual Reasoning Evaluation of Grok Deepseek Janus Gemini Qwen Mistral and ChatGPT

链接: https://arxiv.org/abs/2502.16428
作者: Nidhal Jegham,Marwan Abdelatti,Abdeltawab Hendawi
机构: Department of Computer Science Providence College (普罗维登斯学院计算机科学系), RI, USA
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-101] Fine-Grained Video Captioning through Scene Graph Consolidation

链接: https://arxiv.org/abs/2502.16427
作者: Sanghyeok Chu,Seonguk Seo,Bohyung Han
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-102] Unified Prompt Attack Against Text-to-Image Generation Models

链接: https://arxiv.org/abs/2502.16423
作者: Duo Peng,Qiuhong Ke,Mark He Huang,Ping Hu,Jun Liu
机构: Singapore University of Technology and Design (新加坡科技大学); Monash University (蒙纳士大学); University of Electronic Science and Technology of China (电子科技大学); Lancaster University (兰卡斯特大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by IEEE T-PAMI 2025

点击查看摘要

[CV-103] High-resolution Rainy Image Synthesis: Learning from Rendering

链接: https://arxiv.org/abs/2502.16421
作者: Kaibin Zhou,Shengjie Zhao,Hao Deng,Lin Zhang
机构: School of Software Engineering, Tongji University (同济大学软件工程学院), Shanghai, China
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-104] AnyDexGrasp: General Dexterous Grasping for Different Hands with Human-level Learning Efficiency

链接: https://arxiv.org/abs/2502.16420
作者: Hao-Shu Fang,Hengxu Yan,Zhenyu Tang,Hongjie Fang,Chenxi Wang,Cewu Lu
机构: Shanghai Jiao Tong University(上海交通大学); School of Artificial Intelligence, Shanghai Jiao Tong University(上海交通大学人工智能学院)
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Project website: this https URL

点击查看摘要

[CV-105] DeProPose: Deficiency-Proof 3D Human Pose Estimation via Adaptive Multi-View Fusion

链接: https://arxiv.org/abs/2502.16419
作者: Jianbin Jiao,Xina Cheng,Kailun Yang,Xiangrong Zhang,Licheng Jiao
机构: School of Artificial Intelligence, Xidian University, China(西安电子科技大学); School of Robotics, Hunan University, China(湖南大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO); Image and Video Processing (eess.IV)
备注: The source code will be available at this https URL

点击查看摘要

[CV-106] A Survey on Industrial Anomalies Synthesis

链接: https://arxiv.org/abs/2502.16412
作者: Xichen Xu,Yanshu Wang,Yawen Huang,Jiaqi Liu,Xiaoning Lei,Guoyang Xie,Guannan Jiang,Zhichao Lu
机构: Shanghai Jiao Tong University; Jarvis Research Center, Tencent Youtu Lab; City University of Hong Kong; Department of Intelligent Manufacturing, CATL
类目: Computer Vision and Pattern Recognition (cs.CV); Computational Engineering, Finance, and Science (cs.CE)
备注:

点击查看摘要

[CV-107] An Expert Ensemble for Detecting Anomalous Scenes Interactions and Behaviors in Autonomous Driving

链接: https://arxiv.org/abs/2502.16389
作者: Tianchen Ji,Neeloy Chakraborty,Andre Schreiber,Katherine Driggs-Campbell
机构: Coordinated Science Laboratory, University of Illinois at Urbana-Champaign (伊利诺伊大学香槟分校协调科学实验室)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Robotics (cs.RO)
备注: Accepted by International Journal of Robotics Research (IJRR)

点击查看摘要

[CV-108] Concept Corrector: Erase concepts on the fly for text-to-image diffusion models

链接: https://arxiv.org/abs/2502.16368
作者: Zheling Meng,Bo Peng,Xiaochuan Jin,Yueming Lyu,Wei Wang,Jing Dong
机构: NLPR, Institute of Automation CAS(自动化所模式识别国家重点实验室); School of Artificial Intelligence UCAS(中国科学院大学人工智能学院); Nanjing University(南京大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-109] Audio Visual Segmentation Through Text Embeddings

链接: https://arxiv.org/abs/2502.16359
作者: Kyungbok Lee,You Zhang,Zhiyao Duan
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Multimedia (cs.MM)
备注:

点击查看摘要

[CV-110] AquaNeRF: Neural Radiance Fields in Underwater Media with Distractor Removal

链接: https://arxiv.org/abs/2502.16351
作者: Luca Gough,Adrian Azzarelli,Fan Zhang,Nantheera Anantrasirichai
机构: Visual Information Laboratory, University of Bristol (视觉信息实验室, 布里斯托尔大学), UK
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by 2025 IEEE International Symposium on Circuits and Systems

点击查看摘要

[CV-111] Pointmap Association and Piecewise-Plane Constraint for Consistent and Compact 3D Gaussian Segmentation Field

链接: https://arxiv.org/abs/2502.16303
作者: Wenhao Hu,Wenhao Chai,Shengyu Hao,Xiaotong Cui,Xuexiang Wen,Jenq-Neng Hwang,Gaoang Wang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-112] DualNeRF: Text-Driven 3D Scene Editing via Dual-Field Representation

链接: https://arxiv.org/abs/2502.16302
作者: Yuxuan Xiong,Yue Shi,Yishun Dou,Bingbing Ni
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-113] MOB-GCN: A Novel Multiscale Object-Based Graph Neural Network for Hyperspectral Image Classification

链接: https://arxiv.org/abs/2502.16289
作者: Tuan-Anh Yang,Truong-Son Hy,Phuong D. Dao
机构: University of Alabama at Birmingham (阿拉巴马大学伯明翰分校); Vietnam National University (越南国立大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-114] DiffFake: Exposing Deepfakes using Differential Anomaly Detection WACV2025

【速读】：该论文旨在解决传统深度伪造检测器在处理未见过的深度伪造技术生成的假图像时效果不佳的问题。关键解决方案在于提出了一种名为DiffFake的新型深度伪造检测方法，它将检测问题视为异常检测任务。具体而言，DiffFake通过利用差分异常检测框架学习同一个人面部图像之间自然变化，结合深层人脸嵌入对进行训练。此外，还提出通过伪深度伪造图像（带有全局和局部伪影）训练特征提取器，以提取有意义且可泛化的特征，进而训练异常检测模型。

链接: https://arxiv.org/abs/2502.16247
作者: Sotirios Stamnas,Victor Sanchez
机构: University of Warwick (华威大学)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at WACV 2025 AI4MFDD Workshop

点击查看摘要

Abstract:Traditional deepfake detectors have dealt with the detection problem as a binary classification task. This approach can achieve satisfactory results in cases where samples of a given deepfake generation technique have been seen during training, but can easily fail with deepfakes generated by other techniques. In this paper, we propose DiffFake, a novel deepfake detector that approaches the detection problem as an anomaly detection task. Specifically, DiffFake learns natural changes that occur between two facial images of the same person by leveraging a differential anomaly detection framework. This is done by combining pairs of deep face embeddings and using them to train an anomaly detection model. We further propose to train a feature extractor on pseudo-deepfakes with global and local artifacts, to extract meaningful and generalizable features that can then be used to train the anomaly detection model. We perform extensive experiments on five different deepfake datasets and show that our method can match and sometimes even exceed the performance of state-of-the-art competitors.
zh

[CV-115] Prompt as Knowledge Bank: Boost Vision-language model via Structural Representation for zero-shot medical detection ICLR2025

链接: https://arxiv.org/abs/2502.16223
作者: Yuguang Yang,Tongfei Chen,Haoyu Huang,Linlin Yang,Chunyu Xie,Dawei Leng,Xianbin Cao,Baochang Zhang
机构: School of Electronic Information Engineering, Beihang University, China(北京航空航天大学电子信息工程学院); 360 AI Research, Qihoo 360, China(三六零人工智能研究院); School of Artificial Intelligence, Beihang University, China(北京航空航天大学人工智能学院); State Key Laboratory of Media Convergence and Communication, Communication University of China, China(中国传媒大学媒体融合与传播国家重点实验室); National Superior College for Engineers, Beihang University, China(北京航空航天大学卓越工程师学院); Artificial Intelligence Research Center, Lobachevsky State University, Nizhny Novgorod 603022, Russia(罗巴切夫斯基州立大学人工智能研究中心)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted as ICLR 2025 conference paper

点击查看摘要

[CV-116] SalM2: An Extremely Lightweight Saliency Mamba Model for Real-Time Cognitive Awareness of Driver Attention

链接: https://arxiv.org/abs/2502.16214
作者: Chunyu Zhao,Wentao Mu,Xian Zhou,Wenbo Liu,Fei Yan,Tao Deng
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-117] Mojito: LLM -Aided Motion Instructor with Jitter-Reduced Inertial Tokens

链接: https://arxiv.org/abs/2502.16175
作者: Ziwei Shan,Yaoyu He,Chengfeng Zhao,Jiashen Du,Jingyan Zhang,Qixuan Zhang,Jingyi Yu,Lan Xu
机构: ShanghaiTech University(上海科技大学); Deemos Technology(迪觅思科技)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注: First three authors contribute equally. Project page: this https URL

点击查看摘要

[CV-118] PersGuard: Preventing Malicious Personalization via Backdoor Attacks on Pre-trained Text-to-Image Diffusion Models

链接: https://arxiv.org/abs/2502.16167
作者: Xinwei Liu,Xiaojun Jia,Yuan Xun,Hua Zhang,Xiaochun Cao
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-119] A Deep Learning Framework with Geographic Information Adaptive Loss for Remote Sensing Images based UAV Self-Positioning

链接: https://arxiv.org/abs/2502.16164
作者: Mingkun Li,Ziming Wang,Guang Huo,Wei Chen,Xiaoning Zhao
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-120] USegMix: Unsupervised Segment Mix for Efficient Data Augmentation in Pathology Images

【速读】：该论文旨在解决计算病理学中标签病理数据集稀缺的问题。解决方案的关键在于提出了一种名为USegMix的高效数据增强方法。USegMix通过两个阶段生成新的合成图像：首先，以自动化且无监督的方式使用超像素和Segment Anything Model (SAM)构建组织片段池；其次，从目标图像中选择一个候选片段，并用片段池中的相似片段替换，再利用预训练的扩散模型进行融合。这种方法能够生成多样化且逼真的病理图像，从而提升癌症分类性能。

链接: https://arxiv.org/abs/2502.16160
作者: Jiamu Wang,Jin Tae Kwak
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

Abstract:In computational pathology, researchers often face challenges due to the scarcity of labeled pathology datasets. Data augmentation emerges as a crucial technique to mitigate this limitation. In this study, we introduce an efficient data augmentation method for pathology images, called USegMix. Given a set of pathology images, the proposed method generates a new, synthetic image in two phases. In the first phase, USegMix constructs a pool of tissue segments in an automated and unsupervised manner using superpixels and the Segment Anything Model (SAM). In the second phase, USegMix selects a candidate segment in a target image, replaces it with a similar segment from the segment pool, and blends them by using a pre-trained diffusion model. In this way, USegMix can generate diverse and realistic pathology images. We rigorously evaluate the effectiveness of USegMix on two pathology image datasets of colorectal and prostate cancers. The results demonstrate improvements in cancer classification performance, underscoring the substantial potential of USegMix for pathology image analysis.
zh

[CV-121] Robust Dynamic Facial Expression Recognition

链接: https://arxiv.org/abs/2502.16129
作者: Feng Liu,Hanyang Wang,Siyuan Shen
机构: Shanghai Jiao Tong University(上海交通大学); Midea Group Inc.(美的集团); Baidu Inc.(百度公司)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-122] Detecting OOD Samples via Optimal Transport Scoring Function ICASSP

链接: https://arxiv.org/abs/2502.16115
作者: Heng Gao,Zhuolin He,Jian Pu
机构: Fudan University (复旦大学)
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML)
备注: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025

点击查看摘要

[CV-123] AnxietyFaceTrack: A Smartphone-Based Non-Intrusive Approach for Detecting Social Anxiety Using Facial Features

链接: https://arxiv.org/abs/2502.16106
作者: Nilesh Kumar Sahu,Snehil Gupta,Haroon R Lone
机构: IISER Bhopal(印度科学院理工学院博帕尔分校); AIIMS Bhopal(人工智能与医学科学研究所博帕尔分校)
类目: Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-124] NeurFlow: Interpreting Neural Networks through Neuron Groups and Functional Interactions

链接: https://arxiv.org/abs/2502.16105
作者: Tue M. Cao,Nhat X. Hoang,Hieu H. Pham,Phi Le Nguyen,My T. Thai
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: The Thirteenth International Conference on Learning Representations

点击查看摘要

[CV-125] Set a Thief to Catch a Thief: Combating Label Noise through Noisy Meta Learning

链接: https://arxiv.org/abs/2502.16104
作者: Hanxuan Wang,Na Lu,Xueying Zhao,Yuxuan Yan,Kaipeng Ma,Kwoh Chee Keong,Gustavo Carneiro
机构: Department of Automation Science and Engineering, Xi’an Jiaotong University, China(西安交通大学自动化科学与工程学院，中国); School of Electrical Engineering, Xi’an Jiaotong University, China(西安交通大学电气工程学院，中国); College of Computing and Data Science, Nanyang Technological University, Singapore(南洋理工大学计算与数据科学学院，新加坡); Centre for Vision, Speech and Signal Processing, University of Surrey, UK(萨里大学视觉、语音和信号处理中心，英国)
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-126] Good Representation Better Explanation: Role of Convolutional Neural Networks in Transformer-Based Remote Sensing Image Captioning

链接: https://arxiv.org/abs/2502.16095
作者: Swadhin Das,Saarthak Gupta,and Kamal Kumar,Raksha Sharma
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:

点击查看摘要

[CV-127] MedForge: Building Medical Foundation Models Like Open Source Software Development

链接: https://arxiv.org/abs/2502.16055
作者: Zheling Tan,Kexin Ding,Jin Gao,Mu Zhou,Dimitris Metaxas,Shaoting Zhang,Dequan Wang
机构: Shanghai Jiao Tong University (上海交通大学); Rutgers University (罗格斯大学)
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-128] A Multi-Scale Isolation Forest Approach for Real-Time Detection and Filtering of FGSM Adversarial Attacks in Video Streams of Autonomous Vehicles

链接: https://arxiv.org/abs/2502.16044
作者: Richard Abhulimhen,Negash Begashaw,Gurcan Comert,Chunheng Zhao,Pierluigi Pisu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
备注: 17 pages, 7 figures

点击查看摘要

[CV-129] Clinical Inspired MRI Lesion Segmentation

链接: https://arxiv.org/abs/2502.16032
作者: Lijun Yan,Churan Wang,Fangwei Zhong,Yizhou Wang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-130] Real Time Offside Detection using a Single Camera in Soccer

链接: https://arxiv.org/abs/2502.16030
作者: Shounak Desai
机构: Rochester Institute of Technology (罗切斯特理工学院)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 5 pages 11 figures

点击查看摘要

[CV-131] FeatSharp: Your Vision Model Features Sharper

链接: https://arxiv.org/abs/2502.16025
作者: Mike Ranzinger,Greg Heinrich,Pavlo Molchanov,Jan Kautz,Bryan Catanzaro,Andrew Tao
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-132] Cross-Model Transferability of Adversarial Patches in Real-time Segmentation for Autonomous Driving

链接: https://arxiv.org/abs/2502.16012
作者: Prashant Shekhar,Bidur Devkota,Dumindu Samaraweera,Laxima Niure Kandel,Manoj Babu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-133] Improved Partial Differential Equation and Fast Approximation Algorithm for Hazy/Underwater/Dust Storm Image Enhancement

链接: https://arxiv.org/abs/2502.15986
作者: Uche A. Nnolim
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 11 figures

点击查看摘要

[CV-134] Visual Zero-Shot E-Commerce Product Attribute Value Extraction NAACL2025

链接: https://arxiv.org/abs/2502.15979
作者: Jiaying Gong,Ming Cheng,Hongda Shen,Pierre-Yves Vandenbussche,Janet Jenq,Hoda Eldardiry
机构: 未知
类目: Information Retrieval (cs.IR); Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 4 figures, accepted for publication in NAACL 2025 Industry Track

点击查看摘要

[CV-135] Multi-Agent Multimodal Models for Multicultural Text to Image Generation

链接: https://arxiv.org/abs/2502.15972
作者: Parth Bhalerao,Mounika Yalamarty,Brian Trinh,Oana Ignat
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-136] Human Motion Prediction Reconstruction and Generation

链接: https://arxiv.org/abs/2502.15956
作者: Canxuan Gang,Yiran Wang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Tech report

点击查看摘要

[CV-137] Dissecting Human Body Representations in Deep Networks Trained for Person Identification

链接: https://arxiv.org/abs/2502.15934
作者: Thomas M Metz,Matthew Q Hill,Blake Myers,Veda Nandan Gandi,Rahul Chilakapati,Alice J O’Toole
机构: School of Behavioral and Brain Science, The University of Texas at Dallas (行为与脑科学学院，德克萨斯大学达拉斯分校)

注：原文中没有明确提及其他机构，因此只提取了给出的信息。
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-138] Graph Attention Convolutional U-NET: A Semantic Segmentation Model for Identifying Flooded Areas

链接: https://arxiv.org/abs/2502.15907
作者: Muhammad Umair Danish,Madhushan Buwaneswaran,Tehara Fonseka,Katarina Grolinger
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-139] RIFLEx: A Free Lunch for Length Extrapolation in Video Diffusion Transformers

链接: https://arxiv.org/abs/2502.15894
作者: Min Zhao,Guande He,Yixiao Chen,Hongzhou Zhu,Chongxuan Li,Jun Zhu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-140] Understanding and Evaluating Hallucinations in 3D Visual Language Models

链接: https://arxiv.org/abs/2502.15888
作者: Ruiying Peng,Kaiyuan Li,Weichen Zhang,Chen Gao,Xinlei Chen,Yong Li
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-141] DOEI: Dual Optimization of Embedding Information for Attention-Enhanced Class Activation Maps

链接: https://arxiv.org/abs/2502.15885
作者: Hongjie Zhu,Zeyu Zhang,Guansong Pang,Xu Wang,Shimin Wen,Yu Bai,Daji Ergu,Ying Cai,Yang Zhao
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-142] Generative AI Framework for 3D Object Generation in Augmented Reality

链接: https://arxiv.org/abs/2502.15869
作者: Majid Behravan
机构: 未知
类目: Graphics (cs.GR); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注:

点击查看摘要

[CV-143] A Critical Assessment of Modern Generative Models Ability to Replicate Artistic Styles

链接: https://arxiv.org/abs/2502.15856
作者: Andrea Asperti,Franky George,Tiberio Marras,Razvan Ciprian Stricescu,Fabio Zanotti
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:

点击查看摘要

[CV-144] Exploring the Role of Artificial Intelligence and Machine Learning in Process Optimization for Chemical Industry

链接: https://arxiv.org/abs/2502.15768
作者: Zishuo Lin,Jiajie Wang,Zhe Yan,Peiyong Ma
机构: 未知
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Optics (physics.optics)
备注:

点击查看摘要

[CV-145] Detecting Content Rating Violations in Android Applications: A Vision-Language Approach

链接: https://arxiv.org/abs/2502.15739
作者: D. Denipitiyage,B. Silva,S. Seneviratne,A. Seneviratne,S. Chawla
机构: 未知
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 8 figures

点击查看摘要

[CV-146] A Performance Analysis of You Only Look Once Models for Deployment on Constrained Computational Edge Devices in Drone Applications

链接: https://arxiv.org/abs/2502.15737
作者: Lucas Rey,Ana M. Bernardos,Andrzej D. Dobrzycki,David Carramiñana,Luca Bergesio,Juan A. Besada,José Ramón Casar
机构: 未知
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: This manuscript consists of 24 pages, 7 figures, and 7 tables

点击查看摘要

[CV-147] M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation

链接: https://arxiv.org/abs/2502.17029
作者: Boris Shirokikh,Anvar Kurmukov,Mariia Donskova,Valentin Samokhin,Mikhail Belyaev,Ivan Oseledets
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages,7 figures,11 tables

点击查看摘要

[CV-148] DiffKAN-Inpainting: KAN-based Diffusion model for brain tumor inpainting

链接: https://arxiv.org/abs/2502.16771
作者: Tianli Tao,Ziyang Wang,Han Zhang,Theodoros N. Arvanitis,Le Zhang
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-149] FedDA-TSformer: Federated Domain Adaptation with Vision TimeSformer for Left Ventricle Segmentation on Gated Myocardial Perfusion SPECT Image

【速读】：该论文旨在解决多中心心脏左心室功能评估中的左心室分割精度和患者数据隐私保护问题。关键解决方案在于将联邦领域自适应（Federated Domain Adaptation）与TimeSformer模型相结合，命名为FedDA-TSformer。FedDA-TSformer通过空间注意力（spatial attention）、时间注意力（temporal attention）和联邦学习（federated learning）机制，从多中心门控心肌灌注显像（gated myocardial perfusion, MPS）数据中提取时空特征，同时确保不同医院间数据的安全性和一致性。

链接: https://arxiv.org/abs/2502.16709
作者: Yehong Huang,Chen Zhao,Rochak Dhakal,Min Zhao,Guang-Uei Hung,Zhixin Jiang,Weihua Zhou
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

Abstract:Background and Purpose: Functional assessment of the left ventricle using gated myocardial perfusion (MPS) single-photon emission computed tomography relies on the precise extraction of the left ventricular contours while simultaneously ensuring the security of patient data. Methods: In this paper, we introduce the integration of Federated Domain Adaptation with TimeSformer, named ‘FedDA-TSformer’ for left ventricle segmentation using MPS. FedDA-TSformer captures spatial and temporal features in gated MPS images, leveraging spatial attention, temporal attention, and federated learning for improved domain adaptation while ensuring patient data security. In detail, we employed Divide-Space-Time-Attention mechanism to extract spatio-temporal correlations from the multi-centered MPS datasets, ensuring that predictions are spatio-temporally consistent. To achieve domain adaptation, we align the model output on MPS from three different centers using local maximum mean discrepancy (LMMD) loss. This approach effectively addresses the dual requirements of federated learning and domain adaptation, enhancing the model’s performance during training with multi-site datasets while ensuring the protection of data from different hospitals. Results: Our FedDA-TSformer was trained and evaluated using MPS datasets collected from three hospitals, comprising a total of 150 subjects. Each subject’s cardiac cycle was divided into eight gates. The model achieved Dice Similarity Coefficients (DSC) of 0.842 and 0.907 for left ventricular (LV) endocardium and epicardium segmentation, respectively. Conclusion: Our proposed FedDA-TSformer model addresses the challenge of multi-center generalization, ensures patient data privacy protection, and demonstrates effectiveness in left ventricular (LV) segmentation.
zh

[CV-150] Diagnosing COVID-19 Severity from Chest X-Ray Images Using ViT and CNN Architectures

链接: https://arxiv.org/abs/2502.16622
作者: Luis Lara,Lucia Eve Berger,Rajesh Raju,Shawn Whitfield
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-151] Deep learning approaches to surgical video segmentation and object detection: A Scoping Review

链接: https://arxiv.org/abs/2502.16459
作者: Devanish N. Kamtam,Joseph B. Shrager,Satya Deepya Malla,Nicole Lin,Juan J. Cardona,Jake J. Kim,Clarence Hu
机构: 未知
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 38 pages, 2 figures

点击查看摘要

[CV-152] Category-Selective Neurons in Deep Networks: Comparing Purely Visual and Visual-Language Models

链接: https://arxiv.org/abs/2502.16456
作者: Zitong Lu,Yuxin Wang
机构: 未知
类目: Neurons and Cognition (q-bio.NC); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-153] Revealing Microscopic Objects in Fluorescence Live Imaging by Video-to-video Translation Based on A Spatial-temporal Generative Adversarial Network

链接: https://arxiv.org/abs/2502.16342
作者: Yang Jiao,Mei Yang,Mo Weng
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-154] Large Language Model for Lossless Image Compression with Visual Prompts

链接: https://arxiv.org/abs/2502.16163
作者: Junhao Du,Chuqin Zhou,Ning Cao,Gang Chen,Yunuo Chen,Zhengxue Cheng,Li Song,Guo Lu,Wenjun Zhang
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

[CV-155] Patch Stitching Data Augmentation for Cancer Classification in Pathology Images

链接: https://arxiv.org/abs/2502.16162
作者: Jiamu Wang,Chang-Su Kim,Jin Tae Kwak
机构: 未知
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:

点击查看摘要

人工智能

[AI-0] From System 1 to System 2: A Survey of Reasoning Large Language Models

链接: https://arxiv.org/abs/2502.17419
作者: Zhong-Zhi Li,Duzhen Zhang,Ming-Liang Zhang,Jiaxin Zhang,Zengyan Liu,Yuxuan Yao,Haotian Xu,Junhao Zheng,Pei-Jie Wang,Xiuyi Chen,Yingying Zhang,Fei Yin,Jiahua Dong,Zhijiang Guo,Le Song,Cheng-Lin Liu
类目: Artificial Intelligence (cs.AI)
*备注: Slow-thinking, Large Language Models, Human-like Reasoning, Decision Making in AI, AGI

点击查看摘要

[AI-1] he Empirical Impact of Reducing Symmetries on the Performance of Deep Ensembles and MoE

链接: https://arxiv.org/abs/2502.17391
作者: Andrei Chernov,Oleg Novitskij
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: preprint

点击查看摘要

[AI-2] me series forecasting based on optimized LLM for fault prediction in distribution power grid insulators

链接: https://arxiv.org/abs/2502.17341
作者: João Pedro Matos-Carvalho,Stefano Frizzo Stefenon,Valderi Reis Quietinho Leithardt,Kin-Choong Yow
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
*备注:

点击查看摘要

Abstract:Surface contamination on electrical grid insulators leads to an increase in leakage current until an electrical discharge occurs, which can result in a power system shutdown. To mitigate the possibility of disruptive faults resulting in a power outage, monitoring contamination and leakage current can help predict the progression of faults. Given this need, this paper proposes a hybrid deep learning (DL) model for predicting the increase in leakage current in high-voltage insulators. The hybrid structure considers a multi-criteria optimization using tree-structured Parzen estimation, an input stage filter for signal noise attenuation combined with a large language model (LLM) applied for time series forecasting. The proposed optimized LLM outperforms state-of-the-art DL models with a root-mean-square error equal to 2.24 \times10^-4 for a short-term horizon and 1.21 \times10^-3 for a medium-term horizon.

[AI-3] DMPBC: Self-Imitative Reinforcement Learning for Humanoid Robot Control

链接: https://arxiv.org/abs/2502.17322
作者: Zifeng Zhuang,Diyuan Shi,Runze Suo,Xiao He,Hongyin Zhang,Ting Wang,Shangke Lyu,Donglin Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-4] Benchmarking Retrieval-Augmented Generation in Multi-Modal Contexts

链接: https://arxiv.org/abs/2502.17297
作者: Zhenghao Liu,Xingsheng Zhu,Tianshuo Zhou,Xinyi Zhang,Xiaoyuan Yi,Yukun Yan,Yu Gu,Ge Yu,Maosong Sun
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

Abstract:This paper introduces Multi-Modal Retrieval-Augmented Generation (M^2RAG), a benchmark designed to evaluate the effectiveness of Multi-modal Large Language Models (MLLMs) in leveraging knowledge from multi-modal retrieval documents. The benchmark comprises four tasks: image captioning, multi-modal question answering, multi-modal fact verification, and image reranking. All tasks are set in an open-domain setting, requiring RAG models to retrieve query-relevant information from a multi-modal document collection and use it as input context for RAG modeling. To enhance the context utilization capabilities of MLLMs, we also introduce Multi-Modal Retrieval-Augmented Instruction Tuning (MM-RAIT), an instruction tuning method that optimizes MLLMs within multi-modal contexts. Our experiments show that MM-RAIT improves the performance of RAG systems by enabling them to effectively learn from multi-modal contexts. All data and code are available at this https URL.

[AI-5] Detecting Benchmark Contamination Through Watermarking

链接: https://arxiv.org/abs/2502.17259
作者: Tom Sander,Pierre Fernandez,Saeed Mahloujifar,Alain Durmus,Chuan Guo
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-6] IGDA: Interactive Graph Discovery through Large Language Model Agents

链接: https://arxiv.org/abs/2502.17189
作者: Alex Havrilla,David Alvarez-Melis,Nicolo Fusi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-7] ology-Driven Affective Computing: A Causal Framework for Sustained Well-Being

链接: https://arxiv.org/abs/2502.17172
作者: Bin Yin,Chong-Yi Liu,Liya Fu,Jinkun Zhang
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
*备注: 24 pages, 7 figures

点击查看摘要

[AI-8] CodeSwift: Accelerating LLM Inference for Efficient Code Generation

链接: https://arxiv.org/abs/2502.17139
作者: Qianhui Zhao,Li Zhang,Fang Liu,Xiaoli Lian,Qiaoyuanhe Meng,Ziqian Jiao,Zetong Zhou,Borui Zhang,Runlin Guo,Jia Li
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
*备注:

点击查看摘要

[AI-9] Evaluating the Effectiveness of Large Language Models in Automated News Article Summarization

链接: https://arxiv.org/abs/2502.17136
作者: Lionel Richy Panlap Houamegni,Fatih Gedikli
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
*备注:

点击查看摘要

[AI-10] Applications of Large Models in Medicine

链接: https://arxiv.org/abs/2502.17132
作者: YunHe Su,Zhengyang Lu,Junhui Liu,Ke Pang,Haoran Dai,Sa Liu Yuxin Jia,Lujia Ge,Jing-min Yang
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-11] Low-distortion and GPU-compatible Tree Embeddings in Hyperbolic Space

链接: https://arxiv.org/abs/2502.17130
作者: Max van Spengler,Pascal Mettes
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-12] Adversarial Training for Defense Against Label Poisoning Attacks ICLR2025

链接: https://arxiv.org/abs/2502.17121
作者: Melis Ilayda Bal,Volkan Cevher,Michael Muehlebach
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: Accepted at the International Conference on Learning Representations (ICLR 2025)

点击查看摘要

[AI-13] Diffusion Models for Tabular Data: Challenges Current Progress and Future Directions

链接: https://arxiv.org/abs/2502.17119
作者: Zhong Li,Qi Huang,Lincen Yang,Jiayang Shi,Zhao Yang,Niki van Stein,Thomas Bäck,Matthijs van Leeuwen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-14] Strength Estimation and Human-Like Strength Adjustment in Games ICLR2025

链接: https://arxiv.org/abs/2502.17109
作者: Chun Jung Chen,Chung-Chin Shih,Ti-Rong Wu
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
*备注: Accepted by the Thirteenth International Conference on Learning Representations (ICLR 2025)

点击查看摘要

Abstract:Strength estimation and adjustment are crucial in designing human-AI interactions, particularly in games where AI surpasses human players. This paper introduces a novel strength system, including a strength estimator (SE) and an SE-based Monte Carlo tree search, denoted as SE-MCTS, which predicts strengths from games and offers different playing strengths with human styles. The strength estimator calculates strength scores and predicts ranks from games without direct human interaction. SE-MCTS utilizes the strength scores in a Monte Carlo tree search to adjust playing strength and style. We first conduct experiments in Go, a challenging board game with a wide range of ranks. Our strength estimator significantly achieves over 80% accuracy in predicting ranks by observing 15 games only, whereas the previous method reached 49% accuracy for 100 games. For strength adjustment, SE-MCTS successfully adjusts to designated ranks while achieving a 51.33% accuracy in aligning to human actions, outperforming a previous state-of-the-art, with only 42.56% accuracy. To demonstrate the generality of our strength system, we further apply SE and SE-MCTS to chess and obtain consistent results. These results show a promising approach to strength estimation and adjustment, enhancing human-AI interactions in games. Our code is available at this https URL.

[AI-15] Generative Models in Decision Making: A Survey

链接: https://arxiv.org/abs/2502.17100
作者: Yinchuan Li,Xinyu Shao,Jianping Zhang,Haozhi Wang,Leo Maxime Brunswic,Kaiwen Zhou,Jiqian Dong,Kaiyang Guo,Xiu Li,Zhitang Chen,Jun Wang,Jianye Hao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-16] Forgetting Any Data at Any Time: A Theoretically Certified Unlearning Framework for Vertical Federated Learning

链接: https://arxiv.org/abs/2502.17081
作者: Linian Wang,Leye Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: 18 pages

点击查看摘要

[AI-17] LLM -QE: Improving Query Expansion by Aligning Large Language Models with Ranking Preferences

链接: https://arxiv.org/abs/2502.17057
作者: Sijia Yao,Pengcheng Huang,Zhenghao Liu,Yu Gu,Yukun Yan,Shi Yu,Ge Yu
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
*备注: 13 pages, 5 tables, 4 figures

点击查看摘要

[AI-18] Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam

链接: https://arxiv.org/abs/2502.17055
作者: Tianjin Huang,Haotian Hu,Zhenyu Zhang,Gaojie Jin,Xiang Li,Li Shen,Tianlong Chen,Lu Liu,Qingsong Wen,Zhangyang Wang,Shiwei Liu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-19] abulaTime: A Novel Multimodal Deep Learning Framework for Advancing Acute Coronary Syndrome Prediction through Environmental and Clinical Data Integration

链接: https://arxiv.org/abs/2502.17049
作者: Xin Zhang,Liangxiu Han,Stephen White,Saad Hassan,Philip A Kalra,James Ritchie,Carl Diver,Jennie Shorley
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-20] Distributional Vision-Language Alignment by Cauchy-Schwarz Divergence

链接: https://arxiv.org/abs/2502.17028
作者: Wenzhe Yin,Zehao Xiao,Pan Zhou,Shujian Yu,Jiayi Shen,Jan-Jakob Sonke,Efstratios Gavves
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-21] Class-Dependent Perturbation Effects in Evaluating Time Series Attributions

链接: https://arxiv.org/abs/2502.17022
作者: Gregor Baer,Isel Grau,Chao Zhang,Pieter Van Gorp
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
*备注:

点击查看摘要

[AI-22] Moving Past Single Metrics: Exploring Short-Text Clustering Across Multiple Resolutions

链接: https://arxiv.org/abs/2502.17020
作者: Justin Miller,Tristram Alexander
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
*备注: 11 pages, 3 figures

点击查看摘要

[AI-23] All You Need for Counterfactual Explainability Is Principled and Reliable Estimate of Aleatoric and Epistemic Uncertainty

链接: https://arxiv.org/abs/2502.17007
作者: Kacper Sokol,Eyke Hüllermeier
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
*备注:

点击查看摘要

[AI-24] Lean and Mean: Decoupled Value Policy Optimization with Global Value Guidance

链接: https://arxiv.org/abs/2502.16944
作者: Chenghua Huang,Lu Wang,Fangkai Yang,Pu Zhao,Zhixu Li,Qingwei Lin,Dongmei Zhang,Saravan Rajmohan,Qi Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: 16 pages, 3 figures

点击查看摘要

[AI-25] Supervised contrastive learning from weakly-labeled audio segments for musical version matching

链接: https://arxiv.org/abs/2502.16936
作者: Joan Serrà,R. Oguz Araz,Dmitry Bogdanov,Yuki Mitsufuji
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS); Machine Learning (stat.ML)
*备注: 15 pages, 6 figures, 7 tables; includes Appendix

点击查看摘要

[AI-26] BigMac: A Communication-Efficient Mixture-of-Experts Model Structure for Fast Training and Inference

链接: https://arxiv.org/abs/2502.16927
作者: Zewen Jin,Shengnan Wang,Jiaan Zhu,Hongrui Zhan,Youhui Bai,Lin Zhang,Zhenyu Ming,Cheng Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-27] ENACT-Heart – ENsemble-based Assessment Using CNN and Transformer on Heart Sounds ALT

链接: https://arxiv.org/abs/2502.16914
作者: Jiho Han,Adnan Shaout
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
*备注: Accepted but not published in Global Digital Health Knowledge Exchange Empowerment Conference ( this http URL )

点击查看摘要

[AI-28] When Can We Solve the Weighted Low Rank Approximation Problem in Truly Subquadratic Time? AISTATS2025

链接: https://arxiv.org/abs/2502.16912
作者: Chenyang Li,Yingyu Liang,Zhenmei Shi,Zhao Song
类目: Computational Complexity (cs.CC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: AIStats 2025

点击查看摘要

[AI-29] Zero-shot Load Forecasting for Integrated Energy Systems: A Large Language Model-based Framework with Multi-task Learning

链接: https://arxiv.org/abs/2502.16896
作者: Jiaheng Li,Donghe Li,Ye Yang,Huan Xi,Yu Xiao,Li Sun,Dou An,Qingyu Yang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-30] ReFocus: Reinforcing Mid-Frequency and Key-Frequency Modeling for Multivariate Time Series Forecasting

链接: https://arxiv.org/abs/2502.16890
作者: Guoqi Yu,Yaoming Li,Juncheng Wang,Xiaoyu Guo,Angelica I. Aviles-Rivero,Tong Yang,Shujun Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: Under Review

点击查看摘要

[AI-31] A Multi-LLM -Agent -Based Framework for Economic and Public Policy Analysis

链接: https://arxiv.org/abs/2502.16879
作者: Yuzhi Hao(1),Danyang Xie(2) ((1) Department of Economics, The Hong Kong University of Science and Technology, (2) Thrust of Innovation, Policy, and Entrepreneurship, the Society Hub, The Hong Kong University of Science and Technology (Guangzhou))
类目: Artificial Intelligence (cs.AI); General Economics (econ.GN)
*备注:

点击查看摘要

[AI-32] Utilizing Social Media Analytics to Detect Trends in Saudi Arabias Evolving Market

链接: https://arxiv.org/abs/2502.16871
作者: Kanwal Aalijah
类目: ocial and Information Networks (cs.SI); Artificial Intelligence (cs.AI)
*备注: 8 pages

点击查看摘要

[AI-33] Graphyour Data: Towards End-to-End Modeling Exploring and Generating Report from Raw Data

链接: https://arxiv.org/abs/2502.16868
作者: Longbin Lai,Changwei Luo,Yunkai Lou,Mingchen Ju,Zhengyi Yang
类目: Databases (cs.DB); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
*备注: 4 pages

点击查看摘要

[AI-34] oward Agent ic AI: Generative Information Retrieval Inspired Intelligent Communications and Networking

链接: https://arxiv.org/abs/2502.16866
作者: Ruichen Zhang,Shunpu Tang,Yinqiu Liu,Dusit Niyato,Zehui Xiong,Sumei Sun,Shiwen Mao,Zhu Han
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
*备注: 7 pages, 4 figures

点击查看摘要

[AI-35] PulseBat: A field-accessible dataset for second-life battery diagnostics from realistic histories using multidimensional rapid pulse test

链接: https://arxiv.org/abs/2502.16848
作者: Shengyu Tao,Guangyuan Ma,Huixiong Yang,Minyan Lu,Guodan Wei,Guangmin Zhou,Xuan Zhang
类目: Artificial Intelligence (cs.AI); Computational Physics (physics.comp-ph)
*备注: Extended data descriptor of Nat Commun 15, 10154 (2024), this https URL

点击查看摘要

Abstract:As electric vehicles (EVs) approach the end of their operational life, their batteries retain significant economic value and present promising opportunities for second-life use and material recycling. This is particularly compelling for Global South and other underdeveloped regions, where reliable energy storage is vital to addressing critical challenges posed by weak and even nonexistent power grid and energy infrastructures. However, despite this potential, widespread adoption has been hindered by critical uncertainties surrounding the technical performance, safety, and recertification of second-life batteries. In cases where they have been redeployed, mismatches between estimated and actual performance often render batteries technically unsuitable or hazardous, turning them into liabilities for communities they were intended to benefit. This considerable misalignment exacerbates energy access disparities and undermines the broader vision of energy justice, highlighting an urgent need for robust and scalable solutions to unlock the potential. In the PulseBat Dataset, the authors tested 464 retired lithium-ion batteries, covering 3 cathode material types, 6 historical usages, 3 physical formats, and 6 capacity designs. The pulse test experiments were performed repeatedly for each second-life battery with 10 pulse width, 10 pulse magnitude, multiple state-of-charge, and state-of-health conditions, e.g., from 0.37 to 1.03. The PulseBat Dataset recorded these test conditions and the voltage response as well as the temperature signals that were subject to the injected pulse current, which could be used as a valuable data resource for critical diagnostics tasks such as state-of-charge estimation, state-of-health estimation, cathode material type identification, open-circuit voltage reconstruction, thermal management, and beyond.

[AI-36] In-context learning of evolving data streams with tabular foundational models

链接: https://arxiv.org/abs/2502.16840
作者: Afonso Lourenço,João Gama,Eric P. Xing,Goreti Marreiros
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-37] A Novel Multi-Task Teacher-Student Architecture with Self-Supervised Pretraining for 48-Hour Vasoactive-Inotropic Trend Analysis in Sepsis Mortality Prediction

链接: https://arxiv.org/abs/2502.16834
作者: Houji Jin,Negin Ashrafi,Kamiar Alaei,Elham Pishgar,Greg Placencia,Maryam Pishgar
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-38] Predicting the Energy Landscape of Stochastic Dynamical System via Physics-informed Self-supervised Learning

链接: https://arxiv.org/abs/2502.16828
作者: Ruikun Li,Huandong Wang,Qingmin Liao,Yong Li
类目: Computational Engineering, Finance, and Science (cs.CE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-39] Snoopy: Effective and Efficient Semantic Join Discovery via Proxy Columns

链接: https://arxiv.org/abs/2502.16813
作者: Yuxiang Guo,Yuren Mao,Zhonghao Hu,Lu Chen,Yunjun Gao
类目: Databases (cs.DB); Artificial Intelligence (cs.AI)
*备注: Accepted by TKDE

点击查看摘要

[AI-40] Multi-Agent Autonomous Driving Systems with Large Language Models : A Survey of Recent Advances

链接: https://arxiv.org/abs/2502.16804
作者: Yaozu Wu,Dongyuan Li,Yankai Chen,Renhe Jiang,Henry Peng Zou,Liancheng Fang,Zhen Wang,Philip S. Yu
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-41] VGFL-SA: Vertical Graph Federated Learning Structure Attack Based on Contrastive Learning

链接: https://arxiv.org/abs/2502.16793
作者: Yang Chen,Bin Zhou
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-42] AlphaAgent : LLM -Driven Alpha Mining with Regularized Exploration to Counteract Alpha Decay

链接: https://arxiv.org/abs/2502.16789
作者: Ziyi Tang,Zechuan Chen,Jiarui Yang,Jiayao Mai,Yongsen Zheng,Keze Wang,Jinrui Chen,Liang Lin
类目: Computational Engineering, Finance, and Science (cs.CE); Artificial Intelligence (cs.AI)
*备注: 9 pages

点击查看摘要

[AI-43] he Robustness of Structural Features in Species Interaction Networks

链接: https://arxiv.org/abs/2502.16778
作者: Sanaz Hasanzadeh Fard,Emily Dolson
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
*备注:

点击查看摘要

[AI-44] owards Reinforcement Learning for Exploration of Speculative Execution Vulnerabilities

链接: https://arxiv.org/abs/2502.16756
作者: Evan Lai,Wenjie Xiong,Edward Suh,Mohit Tiwari,Mulong Luo
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-45] Order-Optimal Projection-Free Algorithm for Adversarially Constrained Online Convex Optimization

链接: https://arxiv.org/abs/2502.16744
作者: Yiyang Lu,Mohammad Pedramfar,Vaneet Aggarwal
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
*备注:

点击查看摘要

[AI-46] AUKT: Adaptive Uncertainty-Guided Knowledge Transfer with Conformal Prediction

链接: https://arxiv.org/abs/2502.16736
作者: Rui Liu,Peng Gao,Yu Shen,Ming Lin,Pratap Tokekar
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-47] DeepSeek reshaping healthcare in Chinas tertiary hospitals

链接: https://arxiv.org/abs/2502.16732
作者: Jishizhan Chen,Qingzeng Zhang
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-48] RapidPen: Fully Automated IP-to-Shell Penetration Testing with LLM -based Agents

链接: https://arxiv.org/abs/2502.16730
作者: Sho Nakatani(1) ((1) Security amp; Development Lab)
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-49] NatSGLD: A Dataset with Speech Gesture Logic and Demonstration for Robot Learning in Natural Human-Robot Interaction

链接: https://arxiv.org/abs/2502.16718
作者: Snehesh Shrestha,Yantian Zha,Saketh Banagiri,Ge Gao,Yiannis Aloimonos,Cornelia Fermüller
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
*备注: arXiv admin note: substantial text overlap with arXiv:2403.02274

点击查看摘要

[AI-50] Understanding the Impact of Artificial Intelligence in Academic Writing: Metadata to the Rescue

链接: https://arxiv.org/abs/2502.16713
作者: Javier Conde,Pedro Reviriego,Joaquín Salvachúa,Gonzalo Martínez,José Alberto Hernández,Fabrizio Lombardi
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-51] Exploring Incremental Unlearning: Techniques Challenges and Future Directions

链接: https://arxiv.org/abs/2502.16708
作者: Sadia Qureshi,Thanveer Shaik,Xiaohui Tao,Haoran Xie,Lin Li,Jianming Yong,Xiaohua Jia
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: This work has been submitted to the IEEE for possible publication

点击查看摘要

[AI-52] Reflective Planning : Vision-Language Models for Multi-Stage Long-Horizon Robotic Manipulation

链接: https://arxiv.org/abs/2502.16707
作者: Yunhai Feng,Jiaming Han,Zhuoran Yang,Xiangyu Yue,Sergey Levine,Jianlan Luo
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-53] Beyond Release: Access Considerations for Generative AI Systems

链接: https://arxiv.org/abs/2502.16701
作者: Irene Solaiman,Rishi Bommasani,Dan Hendrycks,Ariel Herbert-Voss,Yacine Jernite,Aviya Skowron,Andrew Trask
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-54] Dynamic LLM Routing and Selection based on User Preferences: Balancing Performance Cost and Ethics

链接: https://arxiv.org/abs/2502.16696
作者: Deepak Babu Piskala,Vijay Raajaa,Sachin Mishra,Bruno Bozza
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

Abstract:With the widespread deployment of large language models (LLMs) such as GPT4, BART, and LLaMA, the need for a system that can intelligently select the most suitable model for specific tasks while balancing cost, latency, accuracy, and ethical considerations has become increasingly important. Recognizing that not all tasks necessitate models with over 100 billion parameters, we introduce OptiRoute, an advanced model routing engine designed to dynamically select and route tasks to the optimal LLM based on detailed user-defined requirements. OptiRoute captures both functional (e.g., accuracy, speed, cost) and non-functional (e.g., helpfulness, harmlessness, honesty) criteria, leveraging lightweight task analysis and complexity estimation to efficiently match tasks with the best-fit models from a diverse array of LLMs. By employing a hybrid approach combining k-nearest neighbors (kNN) search and hierarchical filtering, OptiRoute optimizes for user priorities while minimizing computational overhead. This makes it ideal for real-time applications in cloud-based ML platforms, personalized AI services, and regulated industries.

[AI-55] From Text to Space: Mapping Abstract Spatial Models in LLM s during a Grid-World Navigation Task

链接: https://arxiv.org/abs/2502.16690
作者: Nicolas Martorell
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-56] Are Sparse Autoencoders Useful? A Case Study in Sparse Probing

链接: https://arxiv.org/abs/2502.16681
作者: Subhash Kantamneni,Joshua Engels,Senthooran Rajamanoharan,Max Tegmark,Neel Nanda
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-57] SBSC: Step-By-Step Coding for Improving Mathematical Olympiad Performance ICLR2025 NEURIPS’24

链接: https://arxiv.org/abs/2502.16666
作者: Kunal Singh,Ankan Biswas,Sayandeep Bhowmick,Pradeep Moturi,Siva Kishore Gollapalli
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: Published as a full conference paper at ICLR 2025. Shorter(Early) Version accepted at NeurIPS’24 MATH-AI track

点击查看摘要

[AI-58] Saarthi: The First AI Formal Verification Engineer

链接: https://arxiv.org/abs/2502.16662
作者: Aman Kumar,Deepak Narayan Gadde,Keerthan Kopparam Radhakrishna,Djones Lettnin
类目: Artificial Intelligence (cs.AI)
*备注: Published at the DVCon U.S. 2025

点击查看摘要

[AI-59] BioMaze: Benchmarking and Enhancing Large Language Models for Biological Pathway Reasoning

链接: https://arxiv.org/abs/2502.16660
作者: Haiteng Zhao,Chang Ma,FangZhi Xu,Lingpeng Kong,Zhi-Hong Deng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM)
*备注:

点击查看摘要

[AI-60] Few-shot Continual Relation Extraction via Open Information Extraction

链接: https://arxiv.org/abs/2502.16648
作者: Thiem Nguyen,Anh Nguyen,Quyen Tran,Tu Vu,Diep Nguyen,Linh Ngo,Thien Nguyen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-61] me Series Domain Adaptation via Latent Invariant Causal Mechanism

链接: https://arxiv.org/abs/2502.16637
作者: Ruichu Cai,Junxian Huang,Zhenhui Yang,Zijian Li,Emadeldeen Eldele,Min Wu,Fuchun Sun
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Methodology (stat.ME)
*备注:

点击查看摘要

[AI-62] OptionZero: Planning with Learned Options ICLR2025

链接: https://arxiv.org/abs/2502.16634
作者: Po-Wei Huang,Pei-Chiun Peng,Hung Guei,Ti-Rong Wu
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: Accepted by the Thirteenth International Conference on Learning Representations (ICLR 2025) as oral presentation

点击查看摘要

[AI-63] Energy-Efficient Transformer Inference: Optimization Strategies for Time Series Classification

链接: https://arxiv.org/abs/2502.16627
作者: Arshia Kermani,Ehsan Zeraatkar,Habib Irani
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Performance (cs.PF)
*备注:

点击查看摘要

[AI-64] Intelligent Tutors Beyond K-12: An Observational Study of Adult Learner Engagement and Academic Impact

链接: https://arxiv.org/abs/2502.16613
作者: Adit Gupta,Christopher MacLellan
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-65] arget Speaker Extraction through Comparing Noisy Positive and Negative Audio Enrollm ents

链接: https://arxiv.org/abs/2502.16611
作者: Shitong Xu,Yiyuan Yang,Niki Trigoni,Andrew Markham
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
*备注: 16 pages, 5 figures, appendix included

点击查看摘要

[AI-66] oward Dependency Dynamics in Multi-Agent Reinforcement Learning for Traffic Signal Control

链接: https://arxiv.org/abs/2502.16608
作者: Yuli Zhang,Shangbo Wang,Dongyao Jia,Pengfei Fan,Ruiyuan Jiang,Hankang Gu,Andy H.F. Chow
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
*备注:

点击查看摘要

Abstract:Reinforcement learning (RL) emerges as a promising data-driven approach for adaptive traffic signal control (ATSC) in complex urban traffic networks, with deep neural networks substantially augmenting its learning capabilities. However, centralized RL becomes impractical for ATSC involving multiple agents due to the exceedingly high dimensionality of the joint action space. Multi-agent RL (MARL) mitigates this scalability issue by decentralizing control to local RL agents. Nevertheless, this decentralized method introduces new challenges: the environment becomes partially observable from the perspective of each local agent due to constrained inter-agent communication. Both centralized RL and MARL exhibit distinct strengths and weaknesses, particularly under heavy intersectional traffic conditions. In this paper, we justify that MARL can achieve the optimal global Q-value by separating into multiple IRL (Independent Reinforcement Learning) processes when no spill-back congestion occurs (no agent dependency) among agents (intersections). In the presence of spill-back congestion (with agent dependency), the maximum global Q-value can be achieved by using centralized RL. Building upon the conclusions, we propose a novel Dynamic Parameter Update Strategy for Deep Q-Network (DQN-DPUS), which updates the weights and bias based on the dependency dynamics among agents, i.e. updating only the diagonal sub-matrices for the scenario without spill-back congestion. We validate the DQN-DPUS in a simple network with two intersections under varying traffic, and show that the proposed strategy can speed up the convergence rate without sacrificing optimal exploration. The results corroborate our theoretical findings, demonstrating the efficacy of DQN-DPUS in optimizing traffic signal control.

[AI-67] Reasoning about Affordances: Causal and Compositional Reasoning in LLM s

链接: https://arxiv.org/abs/2502.16606
作者: Magnus F. Gjerde,Vanessa Cheung,David Lagnado
类目: Artificial Intelligence (cs.AI)
*备注: 21 pages, 7 figures, 3 tables

点击查看摘要

[AI-68] racking the Copyright of Large Vision-Language Models through Parameter Learning Adversarial Images ICLR2025

链接: https://arxiv.org/abs/2502.16593
作者: Yubo Wang,Jianting Tang,Chaohu Liu,Linli Xu
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: Accepted to ICLR 2025

点击查看摘要

Abstract:Large vision-language models (LVLMs) have demonstrated remarkable image understanding and dialogue capabilities, allowing them to handle a variety of visual question answering tasks. However, their widespread availability raises concerns about unauthorized usage and copyright infringement, where users or individuals can develop their own LVLMs by fine-tuning published models. In this paper, we propose a novel method called Parameter Learning Attack (PLA) for tracking the copyright of LVLMs without modifying the original model. Specifically, we construct adversarial images through targeted attacks against the original model, enabling it to generate specific outputs. To ensure these attacks remain effective on potential fine-tuned models to trigger copyright tracking, we allow the original model to learn the trigger images by updating parameters in the opposite direction during the adversarial attack process. Notably, the proposed method can be applied after the release of the original model, thus not affecting the model’s performance and behavior. To simulate real-world applications, we fine-tune the original model using various strategies across diverse datasets, creating a range of models for copyright verification. Extensive experiments demonstrate that our method can more effectively identify the original copyright of fine-tuned models compared to baseline methods. Therefore, this work provides a powerful tool for tracking copyrights and detecting unlicensed usage of LVLMs.

[AI-69] LawPal : A Retrieval Augmented Generation Based System for Enhanced Legal Accessibility in India

链接: https://arxiv.org/abs/2502.16573
作者: Dnyanesh Panchal,Aaryan Gole,Vaibhav Narute,Raunak Joshi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-70] Composable Strategy Framework with Integrated Video-Text based Large Language Models for Heart Failure Assessment

链接: https://arxiv.org/abs/2502.16548
作者: Jianzhou Chen,Xiumei Wang,Jinyang Sun,Xi Chen,Heyu Chu,Guo Song,Yuji Luo,Xingping Zhou,Rong Gu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-71] Rebalancing the Scales: A Systematic Mapping Study of Generative Adversarial Networks (GANs) in Addressing Data Imbalance

链接: https://arxiv.org/abs/2502.16535
作者: Pankaj Yadav,Gulshan Sihag,Vivek Vijay
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: 49 pages, 6 figures

点击查看摘要

[AI-72] A Survey of Graph Transformers: Architectures Theories and Applications

链接: https://arxiv.org/abs/2502.16533
作者: Chaohao Yuan,Kangfei Zhao,Ercan Engin Kuruoglu,Liang Wang,Tingyang Xu,Wenbing Huang,Deli Zhao,Hong Cheng,Yu Rong
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-73] Predicting Bad Goods Risk Scores with ARIMA Time Series: A Novel Risk Assessment Approach

链接: https://arxiv.org/abs/2502.16520
作者: Bishwajit Prasad Gond
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-74] Gaussian Process Regression for Improved Underwater Navigation

链接: https://arxiv.org/abs/2502.16510
作者: Nadav Cohen,Itzik Klein
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Signal Processing (eess.SP); Systems and Control (eess.SY)
*备注:

点击查看摘要

[AI-75] PMAT: Optimizing Action Generation Order in Multi-Agent Reinforcement Learning AAMAS2025

链接: https://arxiv.org/abs/2502.16496
作者: Kun Hu,Muning Wen,Xihuai Wang,Shao Zhang,Yiwei Shi,Minne Li,Minglong Li,Ying Wen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
*备注: Accepted by AAMAS 2025

点击查看摘要

[AI-76] A Split-Window Transformer for Multi-Model Sequence Spammer Detection using Multi-Model Variational Autoencoder

链接: https://arxiv.org/abs/2502.16483
作者: Zhou Yang,Yucai Pang,Hongbo Yin,Yunpeng Xiao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multimedia (cs.MM); Social and Information Networks (cs.SI)
*备注:

点击查看摘要

[AI-77] Unmasking Societal Biases in Respiratory Support for ICU Patients through Social Determinants of Health

链接: https://arxiv.org/abs/2502.16477
作者: Mira Moukheiber,Lama Moukheiber,Dana Moukheiber,Hyung-Chul Lee
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-78] Facilitating Emergency Vehicle Passage in Congested Urban Areas Using Multi-agent Deep Reinforcement Learning

链接: https://arxiv.org/abs/2502.16449
作者: Haoran Su
类目: Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
*备注: Ph.D. dissertation in Transportation Systems

点击查看摘要

[AI-79] Auxiliary Discrminator Sequence Generative Adversarial Networks (ADSeqGAN) for Few Sample Molecule Generation

链接: https://arxiv.org/abs/2502.16446
作者: Haocheng Tang,Jing Long,Junmei Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Biomolecules (q-bio.BM)
*备注:

点击查看摘要

[AI-80] abGen-ICL: Residual-Aware In-Context Example Selection for Tabular Data Generation

链接: https://arxiv.org/abs/2502.16414
作者: Liancheng Fang,Aiwei Liu,Hengrui Zhang,Henry Peng Zou,Weizhi Zhang,Philip S. Yu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-81] ool or Tutor? Experimental evidence from AI deployment in cancer diagnosis

链接: https://arxiv.org/abs/2502.16411
作者: Vivianna Fang He,Sihan Li,Phanish Puranam
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-82] rustChain: A Blockchain Framework for Auditing and Verifying Aggregators in Decentralized Federated Learning

链接: https://arxiv.org/abs/2502.16406
作者: Ehsan Hallaji,Roozbeh Razavi-Far,Mehrdad Saif
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
*备注:

点击查看摘要

[AI-83] Navigation-GPT : A Robust and Adaptive Framework Utilizing Large Language Models for Navigation Applications

链接: https://arxiv.org/abs/2502.16402
作者: Feng Ma,Xiu-min Wang,Chen Chen,Xiao-bin Xu,Xin-ping Yan
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-84] FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in FL

链接: https://arxiv.org/abs/2502.16396
作者: Ehsan Hallaji,Roozbeh Razavi-Far,Mehrdad Saif
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
*备注:

点击查看摘要

[AI-85] Understanding Fixed Predictions via Confined Regions

链接: https://arxiv.org/abs/2502.16380
作者: Connor Lawless,Tsui-Wei Weng,Berk Ustun,Madeleine Udell
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
*备注:

点击查看摘要

[AI-86] Does Your AI Agent Get You? A Personalizable Framework for Approximating Human Models from Argumentation-based Dialogue Traces

链接: https://arxiv.org/abs/2502.16376
作者: Yinxu Tang,Stylianos Loukas Vasileiou,William Yeoh
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Logic in Computer Science (cs.LO)
*备注:

点击查看摘要

[AI-87] Personhood Credentials: Human-Centered Design Recommendation Balancing Security Usability and Trust

链接: https://arxiv.org/abs/2502.16375
作者: Ayae Ide,Tanusree Sharma
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
*备注:

点击查看摘要

[AI-88] Exploring Sentiment Manipulation by LLM -Enabled Intelligent Trading Agents

链接: https://arxiv.org/abs/2502.16343
作者: David Byrd
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Multiagent Systems (cs.MA)
*备注:

点击查看摘要

[AI-89] A Gap Between the Gaussian RKHS and Neural Networks: An Infinite-Center Asymptotic Analysis

链接: https://arxiv.org/abs/2502.16331
作者: Akash Kumar,Rahul Parhi,Mikhail Belkin
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
*备注: 22 pages, 1 figure

点击查看摘要

[AI-90] Deep Time Warping for Multiple Time Series Alignment

链接: https://arxiv.org/abs/2502.16324
作者: Alireza Nourbakhsh,Hoda Mohammadzade
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: 32 pages, 13 figures

点击查看摘要

[AI-91] Direct Alignment with Heterogeneous Preferences

链接: https://arxiv.org/abs/2502.16320
作者: Ali Shirali,Arash Nasr-Esfahany,Abdullah Alomar,Parsa Mirtaheri,Rediet Abebe,Ariel Procaccia
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-92] A calibration test for evaluating set-based epistemic uncertainty representations

链接: https://arxiv.org/abs/2502.16299
作者: Mira Jürgens,Thomas Mortier,Eyke Hüllermeier,Viktor Bengs,Willem Waegeman
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
*备注:

点击查看摘要

[AI-93] mePFN: Effective Multivariate Time Series Forecasting with Synthetic Data AAAI-2025

链接: https://arxiv.org/abs/2502.16294
作者: Ege Onur Taga,M. Emrullah Ildiz,Samet Oymak
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: To appear in AAAI-2025 as a conference paper

点击查看摘要

[AI-94] he Design Space of Recent AI-assisted Research Tools for Ideation Sensemaking and Scientific Creativity

链接: https://arxiv.org/abs/2502.16291
作者: Runlong Ye,Matthew Varona,Oliver Huang,Patrick Yung Kang Lee,Michael Liut,Carolina Nobre(University of Toronto)
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-95] Verification of Bit-Flip Attacks against Quantized Neural Networks

链接: https://arxiv.org/abs/2502.16286
作者: Yedi Zhang,Lei Huang,Pengfei Gao,Fu Song,Jun Sun,Jin Song Dong
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: 37 pages, 13 figures, 14 tables

点击查看摘要

[AI-96] MolSpectra: Pre-training 3D Molecular Representation with Multi-modal Energy Spectra ICLR2025

链接: https://arxiv.org/abs/2502.16284
作者: Liang Wang,Shaozhen Liu,Yu Rong,Deli Zhao,Qiang Liu,Shu Wu,Liang Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
*备注: Accepted by ICLR 2025

点击查看摘要

[AI-97] Understanding the Emergence of Multimodal Representation Alignment

链接: https://arxiv.org/abs/2502.16282
作者: Megan Tjandrasuwita,Chanakya Ekbote,Liu Ziyin,Paul Pu Liang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: 21 pages, 22 figures, 3 tables

点击查看摘要

[AI-98] Human Preferences in Large Language Model Latent Space: A Technical Analysis on the Reliability of Synthetic Data in Voting Outcome Prediction

链接: https://arxiv.org/abs/2502.16280
作者: Sarah Ball,Simeon Allmendinger,Frauke Kreuter,Niklas Kühl
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-99] Beyond Trusting Trust: Multi-Model Validation for Robust Code Generation

链接: https://arxiv.org/abs/2502.16279
作者: Bradley McDanel
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
*备注: 3 pages, 2 figures

点击查看摘要

Abstract:This paper explores the parallels between Thompson’s “Reflections on Trusting Trust” and modern challenges in LLM-based code generation. We examine how Thompson’s insights about compiler backdoors take on new relevance in the era of large language models, where the mechanisms for potential exploitation are even more opaque and difficult to analyze. Building on this analogy, we discuss how the statistical nature of LLMs creates novel security challenges in code generation pipelines. As a potential direction forward, we propose an ensemble-based validation approach that leverages multiple independent models to detect anomalous code patterns through cross-model consensus. This perspective piece aims to spark discussion about trust and validation in AI-assisted software development.

[AI-100] Linear Attention for Efficient Bidirectional Sequence Modeling

链接: https://arxiv.org/abs/2502.16249
作者: Arshia Afzal,Elias Abad Rocamora,Leyla Naz Candogan,Pol Puigdemont,Francesco Tonin,Yongtao Wu,Mahsa Shoaran,Volkan Cevher
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-101] Reproducibility Study of Cooperation Competition and Maliciousness: LLM -Stakeholders Interactive Negotiation

链接: https://arxiv.org/abs/2502.16242
作者: Jose L. Garcia,Karolina Hajkova,Maria Marchenko,Carlos Miguel Patiño
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-102] Dynamic Parallel Tree Search for Efficient LLM Reasoning

链接: https://arxiv.org/abs/2502.16235
作者: Yifu Ding,Wentao Jiang,Shunyu Liu,Yongcheng Jing,Jinyang Guo,Yingjie Wang,Jing Zhang,Zengmao Wang,Ziwei Liu,Bo Du,Xianglong Liu,Dacheng Tao
类目: Artificial Intelligence (cs.AI)
*备注: 17 pages, 11 figures

点击查看摘要

[AI-103] Graph Self-Supervised Learning with Learnable Structural and Positional Encodings WWW

链接: https://arxiv.org/abs/2502.16233
作者: Asiri Wijesinghe,Hao Zhu,Piotr Koniusz
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: This paper is accepted by The World Wide Web Conference (WWW) 2025

点击查看摘要

[AI-104] Machine Learning Framework for Early Power Performance and Area Estimation of RTL

链接: https://arxiv.org/abs/2502.16203
作者: Anindita Chattopadhyay,Vijay Kumar Sutrakar
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-105] An Autonomous Network Orchestration Framework Integrating Large Language Models with Continual Reinforcement Learning

链接: https://arxiv.org/abs/2502.16198
作者: Masoud Shokrnezhad,Tarik Taleb
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
*备注: IEEE Communications Magazine

点击查看摘要

[AI-106] Robustness and Cybersecurity in the EU Artificial Intelligence Act

链接: https://arxiv.org/abs/2502.16184
作者: Henrik Nolte,Miriam Rateike,Michèle Finck
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-107] An End-to-End Homomorphically Encrypted Neural Network

链接: https://arxiv.org/abs/2502.16176
作者: Marcos Florencio,Luiz Alencar,Bianca Lima
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-108] Destroy and Repair Using Hyper Graphs for Routing AAAI2025

链接: https://arxiv.org/abs/2502.16170
作者: Ke Li,Fei Liu,Zhengkun Wang,Qingfu Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
*备注: Accepted at AAAI2025

点击查看摘要

[AI-109] Patterns Over Principles: The Frag ility of Inductive Reasoning in LLM s under Noisy Observations

链接: https://arxiv.org/abs/2502.16169
作者: Chunyang Li,Weiqi Wang,Tianshi Zheng,Yangqiu Song
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-110] Heterogeneous Multi-Agent Bandits with Parsimonious Hints AAAI-2025

链接: https://arxiv.org/abs/2502.16128
作者: Amirmahdi Mirfakhar,Xuchuang Wang,Jinhang Zuo,Yair Zick,Mohammad Hajiesmaili
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
*备注: Accepted at AAAI-2025

点击查看摘要

[AI-111] Worse than Zero-shot? A Fact-Checking Dataset for Evaluating the Robustness of RAG Against Misleading Retrievals

链接: https://arxiv.org/abs/2502.16101
作者: Linda Zeng,Rithwik Gupta,Divij Motwani,Diji Yang,Yi Zhang
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
*备注:

点击查看摘要

[AI-112] LitLinker: Supporting the Ideation of Interdisciplinary Contexts with Large Language Models for Teaching Literature in Elementary Schools

链接: https://arxiv.org/abs/2502.16097
作者: Haoxiang Fan,Changshuang Zhou,Hao Yu,Xueyang Wu,Jiangyu Gu,Zhenhui Peng
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-113] Privacy-Aware Joint DNN Model Deployment and Partition Optimization for Delay-Efficient Collaborative Edge Inference

链接: https://arxiv.org/abs/2502.16091
作者: Zhipeng Cheng,Xiaoyu Xia,Hong Wang,Minghui Liwang,Ning Chen,Xuwei Fan,Xianbin Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-114] ogether We Rise: Optimizing Real-Time Multi-Robot Task Allocation using Coordinated Heterogeneous Plays AAMAS2025 AAAI

链接: https://arxiv.org/abs/2502.16079
作者: Aritra Pal,Anandsingh Chauhan,Mayank Baranwal
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
*备注: Accepted to AAMAS 2025 (AAAI Track)

点击查看摘要

[AI-115] Curie: Toward Rigorous and Automated Scientific Experimentation with AI Agents

链接: https://arxiv.org/abs/2502.16069
作者: Patrick Tser Jern Kon,Jiachen Liu,Qiuyi Ding,Yiming Qiu,Zhenning Yang,Yibo Huang,Jayanth Srinivasa,Myungjin Lee,Mosharaf Chowdhury,Ang Chen
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: 21 pages

点击查看摘要

[AI-116] A Survey of Model Extraction Attacks and Defenses in Distributed Computing Environments

链接: https://arxiv.org/abs/2502.16065
作者: Kaixiang Zhao,Lincan Li,Kaize Ding,Neil Zhenqiang Gong,Yue Zhao,Yushun Dong
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-117] Single-Channel EEG Tokenization Through Time-Frequency Modeling

链接: https://arxiv.org/abs/2502.16060
作者: Jathurshan Pradeepkumar,Xihao Piao,Zheng Chen,Jimeng Sun
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
*备注:

点击查看摘要

[AI-118] Human-AI Collaboration in Cloud Security: Cognitive Hierarchy-Driven Deep Reinforcement Learning

链接: https://arxiv.org/abs/2502.16054
作者: Zahra Aref,Sheng Wei,Narayan B. Mandayam
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
*备注:

点击查看摘要

[AI-119] Multimodal Inconsistency Reasoning (MMIR): A New Benchmark for Multimodal Reasoning Models

链接: https://arxiv.org/abs/2502.16033
作者: Qianqi Yan,Yue Fan,Hongquan Li,Shan Jiang,Yang Zhao,Xinze Guan,Ching-Chen Kuo,Xin Eric Wang
类目: Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-120] Hierarchical Residuals Exploit Brain-Inspired Compositionality

链接: https://arxiv.org/abs/2502.16003
作者: Francisco M. López,Jochen Triesch
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注: Accepted at ESANN 2025

点击查看摘要

[AI-121] Automated Query-Product Relevance Labeling using Large Language Models for E-commerce Search

链接: https://arxiv.org/abs/2502.15990
作者: Jayant Sachdev,Sean D Rosario,Abhijeet Phatak,He Wen,Swati Kirti,Chittaranjan Tripathy
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-122] Forecasting Open-Weight AI Model Growth on Hugging Face

链接: https://arxiv.org/abs/2502.15987
作者: Kushal Raj Bhandari,Pin-Yu Chen,Jianxi Gao
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Physics and Society (physics.soc-ph)
*备注: Link to the website for trajectory visualization: this https URL

点击查看摘要

[AI-123] xt-to-SQL Domain Adaptation via Human-LLM Collaborative Data Annotation

链接: https://arxiv.org/abs/2502.15980
作者: Yuan Tian,Daniel Lee,Fei Wu,Tung Mai,Kun Qian,Siddhartha Sahai,Tianyi Zhang,Yunyao Li
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Databases (cs.DB)
*备注: Accepted by IUI’25

点击查看摘要

[AI-124] A Knowledge Distillation-Based Approach to Enhance Transparency of Classifier Models AAAI2025

链接: https://arxiv.org/abs/2502.15959
作者: Yuchen Jiang,Xinyuan Zhao,Yihang Wu,Ahmad Chaddad
类目: Artificial Intelligence (cs.AI)
*备注: Accepted in AAAI 2025

点击查看摘要

[AI-125] Compression Barriers for Autoregressive Transformers

链接: https://arxiv.org/abs/2502.15955
作者: Themistoklis Haris,Krzysztof Onak
类目: Data Structures and Algorithms (cs.DS); Artificial Intelligence (cs.AI); Computational Complexity (cs.CC); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-126] Multi-Objective Optimization of Water Resource Allocation for Groundwater Recharge and Surface Runoff Management in Watershed Systems

链接: https://arxiv.org/abs/2502.15953
作者: Abbas Sharifi,Hajar Kazemi Naeini,Mohsen Ahmadi,Saeed Asadi,Abbas Varmaghani
类目: Artificial Intelligence (cs.AI); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注:

点击查看摘要

[AI-127] Discovery and Deployment of Emergent Robot Swarm Behaviors via Representation Learning and Real2Sim2Real Transfer AAMAS2025

链接: https://arxiv.org/abs/2502.15937
作者: Connor Mattson,Varun Raveendra,Ricardo Vega,Cameron Nowzari,Daniel S. Drew,Daniel S. Brown
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
*备注: 10 pages, 5 figures. To be included in Proc. of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025)

点击查看摘要

[AI-128] Space-O-RAN: Enabling Intelligent Open and Interoperable Non Terrestrial Networks in 6G

链接: https://arxiv.org/abs/2502.15936
作者: Eduardo Baena,Paolo Testolina,Michele Polese,Dimitrios Koutsonikolas,Josep Jornet,Tommaso Melodia
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
*备注:

点击查看摘要

[AI-129] ML-Driven Approaches to Combat Medicare Fraud: Advances in Class Imbalance Solutions Feature Engineering Adaptive Learning and Business Impact

链接: https://arxiv.org/abs/2502.15898
作者: Dorsa Farahmandazad,Kasra Danesh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-130] Practical Principles for AI Cost and Compute Accounting

链接: https://arxiv.org/abs/2502.15873
作者: Stephen Casper,Luke Bailey,Tim Schreier
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
*备注:

点击查看摘要

[AI-131] A Comprehensive Survey on the Trustworthiness of Large Language Models in Healthcare

链接: https://arxiv.org/abs/2502.15871
作者: Manar Aljohani,Jun Hou,Sindhura Kommu,Xuan Wang
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

Abstract:The application of large language models (LLMs) in healthcare has the potential to revolutionize clinical decision-making, medical research, and patient care. As LLMs are increasingly integrated into healthcare systems, several critical challenges must be addressed to ensure their reliable and ethical deployment. These challenges include truthfulness, where models generate misleading information; privacy, with risks of unintentional data retention; robustness, requiring defenses against adversarial attacks; fairness, addressing biases in clinical outcomes; explainability, ensuring transparent decision-making; and safety, mitigating risks of misinformation and medical errors. Recently, researchers have begun developing benchmarks and evaluation frameworks to systematically assess the trustworthiness of LLMs. However, the trustworthiness of LLMs in healthcare remains underexplored, lacking a systematic review that provides a comprehensive understanding and future insights into this area. This survey bridges this gap by providing a comprehensive overview of the recent research of existing methodologies and solutions aimed at mitigating the above risks in healthcare. By focusing on key trustworthiness dimensions including truthfulness, privacy and safety, robustness, fairness and bias, and explainability, we present a thorough analysis of how these issues impact the reliability and ethical use of LLMs in healthcare. This paper highlights ongoing efforts and offers insights into future research directions to ensure the safe and trustworthy deployment of LLMs in healthcare.

[AI-132] Making Sense of AI Limitations: How Individual Perceptions Shape Organizational Readiness for AI Adoption

链接: https://arxiv.org/abs/2502.15870
作者: Thomas Übellacker
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
*备注:

点击查看摘要

[AI-133] AI Governance InternationaL Evaluation Index (AGILE Index)

链接: https://arxiv.org/abs/2502.15859
作者: Yi Zeng,Enmeng Lu,Xin Guan,Cunqing Huangfu,Zizhe Ruan,Ammar Younas
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
*备注: Evaluation Report. 85 pages, 30 Figures

点击查看摘要

[AI-134] Generative AI Training and Copyright Law

链接: https://arxiv.org/abs/2502.15858
作者: Tim W. Dornis,Sebastian Stober
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: submitted as an overview article to the Transactions of the International Society for Music Information Retrieval

点击查看摘要

[AI-135] Deriving Representative Structure from Music Corpora

链接: https://arxiv.org/abs/2502.15849
作者: Ilana Shapiro,Ruanqianqian(Lisa)Huang,Zachary Novack,Cheng-i Wang,Hao-Wen Dong,Taylor Berg-Kirkpatrick,Shlomo Dubnov,Sorin Lerner
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO); Audio and Speech Processing (eess.AS)
*备注: 12 pages, 8 figures, 7 tables

点击查看摘要

[AI-136] One for All: A General Framework of LLM s-based Multi-Criteria Decision Making on Human Expert Level

链接: https://arxiv.org/abs/2502.15778
作者: Hui Wang,Fafa Zhang,Chaoxu Mu
类目: Artificial Intelligence (cs.AI)
*备注: 11 pages, 6 figures

点击查看摘要

[AI-137] SS GAZ PTP: Towards Improving Gumbel AlphaZero with Two-stage Self-play for Multi-constrained Electric Vehicle Routing Problems

链接: https://arxiv.org/abs/2502.15777
作者: Hui Wang,Xufeng Zhang,Xiaoyu Zhang,Zhenhuan Ding,Chaoxu Mu
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI)
*备注: 11 pages,9 figures

点击查看摘要

[AI-138] Logic.py: Bridging the Gap between LLM s and Constraint Solvers

链接: https://arxiv.org/abs/2502.15776
作者: Pascal Kesseli,Peter O’Hearn,Ricardo Silveira Cabral
类目: Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
*备注: 11 pages,9 figures

点击查看摘要

[AI-139] Performance Review on LLM for solving leetcode problems

链接: https://arxiv.org/abs/2502.15770
作者: Lun Wang,Chuanqi Shi,Shaoshui Du,Yiyi Tao,Yixian Shen,Hang Zheng,Xinyu Qiu
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-140] Generalized Attention Flow: Feature Attribution for Transformer Models via Maximum Flow

链接: https://arxiv.org/abs/2502.15765
作者: Behrooz Azarkhalili,Maxwell Libbrecht
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-141] High-Throughput Computational Screening and Interpretable Machine Learning of Metal-organic Frameworks for Iodine Capture

链接: https://arxiv.org/abs/2502.15764
作者: Haoyi Tan,Yukun Teng,Guangcun Shan
类目: Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI); Computational Physics (physics.comp-ph)
*备注: 13 page,6 figures, submitted to npjCM

点击查看摘要

[AI-142] Hybrid Offline-online Scheduling Method for Large Language Model Inference Optimization

链接: https://arxiv.org/abs/2502.15763
作者: Bowen Pang,Kai Li,Ruifeng She,Feifan Wang
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-143] SmartEdge: Smart Healthcare End-to-End Integrated Edge and Cloud Computing System for Diabetes Prediction Enabled by Ensemble Machine Learning

链接: https://arxiv.org/abs/2502.15762
作者: Alain Hennebelle,Qifan Dieng,Leila Ismail,Rajkumar Buyya
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-144] LoXR: Performance Evaluation of Locally Executing LLM s on XR Devices

链接: https://arxiv.org/abs/2502.15761
作者: Dawar Khan,Xinyu Liu,Omar Mena,Donggang Jia,Alexandre Kouyoumdjian,Ivan Viola
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Graphics (cs.GR); Human-Computer Interaction (cs.HC)
*备注:

点击查看摘要

[AI-145] Physics-consistent machine learning: output projection onto physical manifolds

链接: https://arxiv.org/abs/2502.15755
作者: Matilde Valente,Tiago C. Dias,Vasco Guerra,Rodrigo Ventura
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Plasma Physics (physics.plasm-ph)
*备注: 22 pages, 5 figures

点击查看摘要

[AI-146] xt2Net: Transforming Plain-text To A Dynamic Interactive Network Simulation Environment

链接: https://arxiv.org/abs/2502.15754
作者: Alireza Marefat,Abbaas Alif Mohamed Nishar,Ashwin Ashok
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注: 7 pages, 9 figures, Accepted at IEEE SoutheastCon 2025

点击查看摘要

[AI-147] CProF:Time-Complexity Prediction SSL Framework NAACL2025

链接: https://arxiv.org/abs/2502.15749
作者: Joonghyuk Hahn,Hyeseon Ahn,Jungin Kim,Soohan Lim,Yo-Sub Han
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
*备注: 26 pages, 13 figures, This paper has been accepted to NAACL 2025

点击查看摘要

[AI-148] DistrEE: Distributed Early Exit of Deep Neural Network Inference on Edge Devices

链接: https://arxiv.org/abs/2502.15735
作者: Xian Peng,Xin Wu,Lianming Xu,Li Wang,Aiguo Fei
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-149] Data Wrangling Task Automation Using Code-Generating Language Models AAAI2025

链接: https://arxiv.org/abs/2502.15732
作者: Ashlesha Akella,Krishnasuri Narayanam
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Databases (cs.DB); Software Engineering (cs.SE)
*备注: Accepted at AAAI 2025 Demo

点击查看摘要

[AI-150] Modular and Integrated AI Control Framework across Fiber and Wireless Networks for 6G

链接: https://arxiv.org/abs/2502.15731
作者: Merim Dzaferagic,Marco Ruffini,Daniel Kilper
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-151] Retrieval Augmented Generation Based LLM Evaluation For Protocol State Machine Inference With Chain-of-Thought Reasoning

链接: https://arxiv.org/abs/2502.15727
作者: Youssef Maklad,Fares Wael,Wael Elsersy,Ali Hamdi
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Information Retrieval (cs.IR)
*备注:

点击查看摘要

[AI-152] Instruction-Based Fine-tuning of Open-Source LLM s for Predicting Customer Purchase Behaviors

链接: https://arxiv.org/abs/2502.15724
作者: Halil Ibrahim Ergul,Selim Balcisoy,Burcin Bozkaya
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-153] Balancing Content Size in RAG -Text2SQL System

链接: https://arxiv.org/abs/2502.15723
作者: Prakhar Gurawa,Anjali Dharmik
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-154] RI-QA: a Toolset for Customized Question-Answer Dataset Generation Using Language Models for Enhanced Scientific Research

链接: https://arxiv.org/abs/2502.15721
作者: Qiming Liu,Zhongzheng Niu,Siting Liu,Mao Tian
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Digital Libraries (cs.DL)
*备注: 13 pages, 3 figures

点击查看摘要

[AI-155] raining AI to be Loyal

链接: https://arxiv.org/abs/2502.15720
作者: Sewoong Oh,Himanshu Tyagi,Pramod Viswanath
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 13 pages 4 figures

点击查看摘要

[AI-156] Governing AI Beyond the Pretraining Frontier

链接: https://arxiv.org/abs/2502.15719
作者: Nicholas A. Caputo
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
*备注: 14 pages

点击查看摘要

[AI-157] Regulating Multifunctionality

链接: https://arxiv.org/abs/2502.15715
作者: Cary Coglianese,Colton R. Crum
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
*备注: Forthcoming in Philipp Hacker, Andreas Engel, Sarah Hammer and Brent Mittelstadt (eds), The Oxford Handbook on the Foundations and Regulation of Generative AI (Oxford University Press)

点击查看摘要

[AI-158] UAV-assisted Internet of Vehicles: A Framework Empowered by Reinforcement Learning and Blockchain

链接: https://arxiv.org/abs/2502.15713
作者: Ahmed Alagha,Maha Kadadha,Rabeb Mizouni,Shakti Singh,Jamal Bentahar,Hadi Otrok
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
*备注:

点击查看摘要

[AI-159] GPUs CPUs and… NICs: Rethinking the Networks Role in Serving Complex AI Pipelines

链接: https://arxiv.org/abs/2502.15712
作者: Mike Wong,Ulysses Butler,Emma Farkash,Praveen Tammana,Anirudh Sivaraman,Ravi Netravali
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Operating Systems (cs.OS)
*备注:

点击查看摘要

[AI-160] Conditional Diffusion-Flow models for generating 3D cosmic density fields: applications to f® cosmologies

链接: https://arxiv.org/abs/2502.17087
作者: Julieth Katherine Riveros,Paola Saavedra,Hector J. Hortua,Jorge Enrique Garcia-Farieta,Ivan Olier
类目: Cosmology and Nongalactic Astrophysics (astro-ph.CO); Artificial Intelligence (cs.AI)
*备注: 13 pages comments welcome

点击查看摘要

[AI-161] Convergence of Shallow ReLU Networks on Weakly Interacting Data

链接: https://arxiv.org/abs/2502.16977
作者: Léo Dana(SIERRA),Francis Bach(SIERRA),Loucas Pillaud-Vivien(ENPC, CERMICS)
类目: Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注:

点击查看摘要

[AI-162] Auto-ADMET: An Effective and Interpretable AutoML Method for Chemical ADMET Property Prediction

链接: https://arxiv.org/abs/2502.16378
作者: Alex G. C. de Sá,David B. Ascher
类目: Biomolecules (q-bio.BM); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

Abstract:Machine learning (ML) has been playing important roles in drug discovery in the past years by providing (pre-)screening tools for prioritising chemical compounds to pass through wet lab experiments. One of the main ML tasks in drug discovery is to build quantitative structure-activity relationship (QSAR) models, associating the molecular structure of chemical compounds with an activity or property. These properties – including absorption, distribution, metabolism, excretion and toxicity (ADMET) – are essential to model compound behaviour, activity and interactions in the organism. Although several methods exist, the majority of them do not provide an appropriate model’s personalisation, yielding to bias and lack of generalisation to new data since the chemical space usually shifts from application to application. This fact leads to low predictive performance when completely new data is being tested by the model. The area of Automated Machine Learning (AutoML) emerged aiming to solve this issue, outputting tailored ML algorithms to the data at hand. Although an important task, AutoML has not been practically used to assist cheminformatics and computational chemistry researchers often, with just a few works related to the field. To address these challenges, this work introduces Auto-ADMET, an interpretable evolutionary-based AutoML method for chemical ADMET property prediction. Auto-ADMET employs a Grammar-based Genetic Programming (GGP) method with a Bayesian Network Model to achieve comparable or better predictive performance against three alternative methods – standard GGP method, pkCSM and XGBOOST model – on 12 benchmark chemical ADMET property prediction datasets. The use of a Bayesian Network model on Auto-ADMET’s evolutionary process assisted in both shaping the search procedure and interpreting the causes of its AutoML performance.

[AI-163] rECGnition_v2.0: Self-Attentive Canonical Fusion of ECG and Patient Data using deep learning for effective Cardiac Diagnostics

链接: https://arxiv.org/abs/2502.16255
作者: Shreya Srivastava,Durgesh Kumar,Ram Jiwari,Sandeep Seth,Deepak Sharma
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[AI-164] Speech Enhancement Using Continuous Embeddings of Neural Audio Codec ICASSP2025

链接: https://arxiv.org/abs/2502.16240
作者: Haoyang Li,Jia Qi Yip,Tianyu Fan,Eng Siong Chng
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Sound (cs.SD)
*备注: Accepted to ICASSP 2025

点击查看摘要

机器学习

[LG-0] S4S: Solving for a Diffusion Model Solver

链接: https://arxiv.org/abs/2502.17423
作者: Eric Frankel,Sitan Chen,Jerry Li,Pang Wei Koh,Lillian J. Ratliff,Sewoong Oh
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-1] COSMOS: A Hybrid Adaptive Optimizer for Memory-Efficient Training of LLM s

链接: https://arxiv.org/abs/2502.17410
作者: Liming Liu,Zhenghao Xu,Zixuan Zhang,Hao Kang,Zichong Li,Chen Liang,Weizhu Chen,Tuo Zhao
类目: Machine Learning (cs.LG)
*备注: 23 pages, 9 figures, 6 tables

点击查看摘要

[LG-2] On the Dichotomy Between Privacy and Traceability in ell_p Stochastic Convex Optimization

链接: https://arxiv.org/abs/2502.17384
作者: Sasha Voitovych,Mahdi Haghifam,Idan Attias,Gintare Karolina Dziugaite,Roi Livni,Daniel M. Roy
类目: Machine Learning (cs.LG)
*备注: 53 Pages

点击查看摘要

[LG-3] Continuous Integration Practices in Machine Learning Projects: The Practitioners` Perspective

链接: https://arxiv.org/abs/2502.17378
作者: João Helis Bernardo,Daniel Alencar da Costa,Filipe Roseiro Cogo,Sérgio Queiróz de Medeiros,Uirá Kulesza
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-4] Sustainable Greenhouse Management: A Comparative Analysis of Recurrent and Graph Neural Networks

链接: https://arxiv.org/abs/2502.17371
作者: Emiliano Seri,Marcello Petitta,Cristina Cornaro
类目: Machine Learning (cs.LG); Applications (stat.AP)
*备注:

点击查看摘要

[LG-5] A Closer Look at TabPFN v2: Strength Limitation and Extension

链接: https://arxiv.org/abs/2502.17361
作者: Han-Jia Ye,Si-Yang Liu,Wei-Lun Chao
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-6] Distributional Scaling Laws for Emergent Capabilities

链接: https://arxiv.org/abs/2502.17356
作者: Rosie Zhao,Tian Qin,David Alvarez-Melis,Sham Kakade,Naomi Saphra
类目: Machine Learning (cs.LG)
*备注: 17 pages

点击查看摘要

[LG-7] Low-rank bias weight decay and model merging in neural networks

链接: https://arxiv.org/abs/2502.17340
作者: Ilja Kuzborskij,Yasin Abbasi Yadkori
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-8] okenized SAEs: Disentangling SAE Reconstructions

链接: https://arxiv.org/abs/2502.17332
作者: Thomas Dooms,Daniel Wilhelm
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-9] Unveiling ECC Vulnerabilities: LSTM Networks for Operation Recognition in Side-Channel Attacks

链接: https://arxiv.org/abs/2502.17330
作者: Alberto Battistello,Guido Bertoni,Michele Corrias,Lorenzo Nava,Davide Rusconi,Matteo Zoia,Fabio Pierazzi,Andrea Lanzi
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 20 pages, 5 figures

点击查看摘要

[LG-10] Survey on Strategic Mining in Blockchain: A Reinforcement Learning Approach

链接: https://arxiv.org/abs/2502.17307
作者: Jichen Li,Lijia Xie,Hanting Huang,Bo Zhou,Binfeng Song,Wanying Zeng,Xiaotie Deng,Xiao Zhang
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
*备注: 10 pages

点击查看摘要

[LG-11] Delta Decompression for MoE-based LLM s Compression

链接: https://arxiv.org/abs/2502.17298
作者: Hao Gu,Wei Li,Lujun Li,Qiyuan Zhu,Mark Lee,Shengjie Sun,Wei Xue,Yike Guo
类目: Machine Learning (cs.LG)
*备注: Work in progress

点击查看摘要

[LG-12] Joint Value Estimation and Bidding in Repeated First-Price Auctions

链接: https://arxiv.org/abs/2502.17292
作者: Yuxiao Wen,Yanjun Han,Zhengyuan Zhou
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT); Information Theory (cs.IT); Methodology (stat.ME); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-13] Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coverag e

链接: https://arxiv.org/abs/2502.17264
作者: Konstantina Bairaktari,Jiayun Wu,Zhiwei Steven Wu
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-14] Robust Federated Learning in Unreliable Wireless Networks: A Client Selection Approach

链接: https://arxiv.org/abs/2502.17260
作者: Yanmeng Wang,Wenkai Ji,Jian Zhou,Fu Xiao,Tsung-Hui Chang
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-15] REINFORCE Adversarial Attacks on Large Language Models : An Adaptive Distributional and Semantic Objective

链接: https://arxiv.org/abs/2502.17254
作者: Simon Geisler,Tom Wollschläger,M. H. I. Abdalla,Vincent Cohen-Addad,Johannes Gasteiger,Stephan Günnemann
类目: Machine Learning (cs.LG)
*备注: 30 pages, 6 figures, 15 tables

点击查看摘要

[LG-16] Overconfident Oracles: Limitations of In Silico Sequence Design Benchmarking

链接: https://arxiv.org/abs/2502.17246
作者: Shikha Surana,Nathan Grinsztajn,Timothy Atkinson,Paul Duckworth,Thomas D. Barrett
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-17] Electrical Load Forecasting over Multihop Smart Metering Networks with Federated Learning

链接: https://arxiv.org/abs/2502.17226
作者: Ratun Rahman,Pablo Moriano,Samee U. Khan,Dinh C. Nguyen
类目: Machine Learning (cs.LG)
*备注: arXiv admin note: text overlap with arXiv:2411.10619

点击查看摘要

[LG-18] Humanoid Whole-Body Locomotion on Narrow Terrain via Dynamic Balance and Reinforcement Learning

链接: https://arxiv.org/abs/2502.17219
作者: Weiji Xie,Chenjia Bai,Jiyuan Shi,Junkai Yang,Yunfei Ge,Weinan Zhang,Xuelong Li
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: This work has been submitted to the IEEE for possible publication

点击查看摘要

[LG-19] Neural Attention: A Novel Mechanism for Enhanced Expressive Power in Transformer Models

链接: https://arxiv.org/abs/2502.17206
作者: Andrew DiGiugno,Ausif Mahmood
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-20] Sparse Hyperparametric Itakura-Saito NMF via Bi-Level Optimization

链接: https://arxiv.org/abs/2502.17123
作者: Laura Selicatoa,Flavia Esposito,Andersen Ang,Nicoletta Del Buono,Rafal Zdunek
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 22 pages, 5 figures, 4 tables

点击查看摘要

[LG-21] A comparative analysis of rank aggregation methods for the partial label ranking problem

链接: https://arxiv.org/abs/2502.17077
作者: Jiayi Wang,Juan C. Alfaro,Viktor Bengs
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-22] Data Analysis Prediction over Multiple Unseen Datasets: A Vector Embedding Approach

链接: https://arxiv.org/abs/2502.17060
作者: Andreas Loizou,Dimitrios Tsoumakos
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-23] Advancing Eurasia Fire Understanding Through Machine Learning Techniques

链接: https://arxiv.org/abs/2502.17023
作者: Boris Kriuk
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 13 pages, 7 figures, 2 tables

点击查看摘要

[LG-24] Unbiased and Sign Compression in Distributed Learning: Comparing Noise Resilience via SDEs AISTATS2025

链接: https://arxiv.org/abs/2502.17009
作者: Enea Monzio Compagnoni,Rustem Islamov,Frank Norbert Proske,Aurelien Lucchi
类目: Machine Learning (cs.LG)
*备注: Accepted at AISTATS 2025 (Oral). arXiv admin note: substantial text overlap with arXiv:2411.15958

点击查看摘要

[LG-25] Atten-Transformer: A Deep Learning Framework for User App Usage Prediction

链接: https://arxiv.org/abs/2502.16957
作者: Longlong Li,Cunquan Qu,Guanghui Wang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-26] MTVHunter: Smart Contracts Vulnerability Detection Based on Multi-Teacher Knowledge Translation

链接: https://arxiv.org/abs/2502.16955
作者: Guokai Sun,Yuan Zhuang,Shuo Zhang,Xiaoyu Feng,Zhenguang Liu,Liguo Zhang
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-27] Deep Minimax Classifiers for Imbalanced Datasets with a Small Number of Minority Samples

链接: https://arxiv.org/abs/2502.16948
作者: Hansung Choi,Daewon Seo
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-28] SUSTeR: Sparse Unstructured Spatio Temporal Reconstruction on Traffic Prediction

链接: https://arxiv.org/abs/2502.16935
作者: Yannick Wölker,Christian Beth,Matthias Renz,Arne Biastoch
类目: Machine Learning (cs.LG)
*备注: accepted and presented at ACM SIGSPATIAL '23

点击查看摘要

[LG-29] Achieving Fair PCA Using Joint Eigenvalue Decomposition

链接: https://arxiv.org/abs/2502.16933
作者: Vidhi Rathore,Naresh Manwani
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-30] Machine learning and high dimensional vector search

链接: https://arxiv.org/abs/2502.16931
作者: Matthijs Douze
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-31] Distributionally Robust Active Learning for Gaussian Process Regression

链接: https://arxiv.org/abs/2502.16870
作者: Shion Takeno,Yoshito Okura,Yu Inatsu,Aoyama Tatsuya,Tomonari Tanaka,Akahane Satoshi,Hiroyuki Hanada,Noriaki Hashimoto,Taro Murayama,Hanju Lee,Shinya Kojima,Ichiro Takeuchi
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 25 pages, 3 figures

点击查看摘要

[LG-32] Leverag ing Large Language Models for Effective and Explainable Multi-Agent Credit Assignment AAMAS2025

链接: https://arxiv.org/abs/2502.16863
作者: Kartik Nagpal,Dayi Dong,Jean-Baptiste Bouvier,Negar Mehr
类目: Multiagent Systems (cs.MA); Machine Learning (cs.LG); Robotics (cs.RO)
*备注: 8 pages+Appendix, 6 Figures, AAMAS 2025

点击查看摘要

[LG-33] Posterior Inference with Diffusion Models for High-dimensional Black-box Optimization

链接: https://arxiv.org/abs/2502.16824
作者: Taeyoung Yun,Kiyoung Om,Jaewoo Lee,Sujin Yun,Jinkyoo Park
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 21 pages, 12 figures, 5 tables

点击查看摘要

[LG-34] Fast Accurate Manifold Denoising by Tunneling Riemannian Optimization

链接: https://arxiv.org/abs/2502.16819
作者: Shiyu Wang,Mariam Avagyan,Yihan Shen,Arnaud Lamy,Tingran Wang,Szabolcs Márka,Zsuzsa Márka,John Wright
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:

点击查看摘要

[LG-35] Forecasting Rare Language Model Behaviors

链接: https://arxiv.org/abs/2502.16797
作者: Erik Jones,Meg Tong,Jesse Mu,Mohammed Mahfoud,Jan Leike,Roger Grosse,Jared Kaplan,William Fithian,Ethan Perez,Mrinank Sharma
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-36] CipherPrune: Efficient and Scalable Private Transformer Inference ICLR2025

链接: https://arxiv.org/abs/2502.16782
作者: Yancheng Zhang,Jiaqi Xue,Mengxin Zheng,Mimi Xie,Mingzhe Zhang,Lei Jiang,Qian Lou
类目: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
*备注: Accepted by ICLR 2025

点击查看摘要

[LG-37] Model-Based Exploration in Monitored Markov Decision Processes

链接: https://arxiv.org/abs/2502.16772
作者: Alireza Kazemipour,Simone Parisi,Matthew E. Taylor,Michael Bowling
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-38] Exact Learning of Permutations for Nonzero Binary Inputs with Logarithmic Training Size and Quadratic Ensemble Complexity

链接: https://arxiv.org/abs/2502.16763
作者: George Giapitzakis,Artur Back de Luca,Kimon Fountoulakis
类目: Machine Learning (cs.LG)
*备注: 21 pages, 1 figure

点击查看摘要

[LG-39] Keeping up with dynamic attackers: Certifying robustness to adaptive online data poisoning AISTATS

链接: https://arxiv.org/abs/2502.16737
作者: Avinandan Bose,Laurent Lessard,Maryam Fazel,Krishnamurthy Dj Dvijotham
类目: Machine Learning (cs.LG)
*备注: Proceedings of the 28th International Conference on Artificial Intelligence and Statistics (AISTATS) 2025, Mai Khao, Thailand. PMLR: Volume 258

点击查看摘要

Abstract:The rise of foundation models fine-tuned on human feedback from potentially untrusted users has increased the risk of adversarial data poisoning, necessitating the study of robustness of learning algorithms against such attacks. Existing research on provable certified robustness against data poisoning attacks primarily focuses on certifying robustness for static adversaries who modify a fraction of the dataset used to train the model before the training algorithm is applied. In practice, particularly when learning from human feedback in an online sense, adversaries can observe and react to the learning process and inject poisoned samples that optimize adversarial objectives better than when they are restricted to poisoning a static dataset once, before the learning algorithm is applied. Indeed, it has been shown in prior work that online dynamic adversaries can be significantly more powerful than static ones. We present a novel framework for computing certified bounds on the impact of dynamic poisoning, and use these certificates to design robust learning algorithms. We give an illustration of the framework for the mean estimation and binary classification problems and outline directions for extending this in further work. The code to implement our certificates and replicate our results is available at this https URL.

[LG-40] owards Optimal Adversarial Robust Reinforcement Learning with Infinity Measurement Error

链接: https://arxiv.org/abs/2502.16734
作者: Haoran Li,Zicheng Zhang,Wang Luo,Congying Han,Jiayu Lv,Tiande Guo,Yudong Hu
类目: Machine Learning (cs.LG)
*备注: arXiv admin note: substantial text overlap with arXiv:2402.02165

点击查看摘要

[LG-41] Model-agnostic Coreset Selection via LLM -based Concept Bottlenecks

链接: https://arxiv.org/abs/2502.16733
作者: Akshay Mehra,Trisha Mittal,Subhadra Gopalakrishnan,Joshua Kimball
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-42] o Share or Not to Share: Investigating Weight Sharing in Variational Graph Autoencoders WWW2025

链接: https://arxiv.org/abs/2502.16724
作者: Guillaume Salha-Galvan,Jiaying Xu
类目: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
*备注: 2025 ACM Web Conference (WWW 2025)

点击查看摘要

[LG-43] Subsampling Graphs with GNN Performance Guarantees

链接: https://arxiv.org/abs/2502.16703
作者: Mika Sarkin Jain,Stefanie Jegelka,Ishani Karmarkar,Luana Ruiz,Ellen Vitercik
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-44] Analyzing Factors Influencing Driver Willingness to Accept Advanced Driver Assistance Systems

链接: https://arxiv.org/abs/2502.16688
作者: Hannah Musau,Nana Kankam Gyimah,Judith Mwakalonge,Gurcan Comert,Saidi Siuhi
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-45] MetaSym: A Symplectic Meta-learning Framework for Physical Intelligence

链接: https://arxiv.org/abs/2502.16667
作者: Pranav Vaidhyanathan,Aristotelis Papatheodorou,Mark T. Mitchison,Natalia Ares,Ioannis Havoutis
类目: Machine Learning (cs.LG); Robotics (cs.RO); Computational Physics (physics.comp-ph); Quantum Physics (quant-ph)
*备注: 8+10 pages, 5 figures, 4 tables

点击查看摘要

[LG-46] Geometric Kolmogorov-Arnold Superposition Theorem

链接: https://arxiv.org/abs/2502.16664
作者: Francesco Alesiani,Takashi Maruyama,Henrik Christiansen,Viktor Zaverkin
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-47] Volume Optimality in Conformal Prediction with Structured Prediction Sets

链接: https://arxiv.org/abs/2502.16658
作者: Chao Gao,Liren Shan,Vaidehi Srinivas,Aravindan Vijayaraghavan
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 41 pages, 19 figures, 2 tables

点击查看摘要

[LG-48] Optimal Kernel Learning for Gaussian Process Models with High-Dimensional Input

链接: https://arxiv.org/abs/2502.16617
作者: Lulu Kang,Minshen Xu
类目: Machine Learning (cs.LG); Methodology (stat.ME); Machine Learning (stat.ML)
*备注: 30 pages, 7 tables, 8 figures

点击查看摘要

[LG-49] Guarding the Privacy of Label-Only Access to Neural Network Classifiers via iDP Verification

链接: https://arxiv.org/abs/2502.16519
作者: Anan Kabaha,Dana Drachsler-Cohen
类目: Machine Learning (cs.LG); Programming Languages (cs.PL)
*备注:

点击查看摘要

[LG-50] Improved Margin Generalization Bounds for Voting Classifiers

链接: https://arxiv.org/abs/2502.16462
作者: Mikael Møller Høgsgaard,Kasper Green Larsen
类目: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS); Statistics Theory (math.ST); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-51] MAPN: Enhancing Heterogeneous Sparse Graph Representation by Mamba-based Asynchronous Aggregation

链接: https://arxiv.org/abs/2502.16454
作者: Xuqi Mao,Zhenying He,X. Sean Wang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-52] Compression Scaling Laws:Unifying Sparsity and Quantization

链接: https://arxiv.org/abs/2502.16440
作者: Elias Frantar,Utku Evci,Wonpyo Park,Neil Houlsby,Dan Alistarh
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-53] Automated Flow Pattern Classification in Multi-phase Systems Using AI and Capacitance Sensing Techniques

链接: https://arxiv.org/abs/2502.16432
作者: Nian Ran,Fayez M. Al-Alweet,Richard Allmendinger,Ahmad Almakhlafi
类目: Machine Learning (cs.LG)
*备注: 29 pages, Applied Soft Computing under reviewed

点击查看摘要

[LG-54] UniDyG: A Unified and Effective Representation Learning Approach for Large Dynamic Graphs

链接: https://arxiv.org/abs/2502.16431
作者: Yuanyuan Xu,Wenjie Zhang,Xuemin Lin,Ying Zhang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

Abstract:Dynamic graphs are formulated in continuous-time or discrete-time dynamic graphs. They differ in temporal granularity: Continuous-Time Dynamic Graphs (CTDGs) exhibit rapid, localized changes, while Discrete-Time Dynamic Graphs (DTDGs) show gradual, global updates. This difference leads to isolated developments in representation learning for each type. To advance representation learning, recent research attempts to design a unified model capable of handling both CTDGs and DTDGs. However, it typically focuses on local dynamic propagation for temporal structure learning in the time domain, failing to accurately capture the structural evolution associated with each temporal granularity. In addition, existing works-whether specific or unified-often overlook the issue of temporal noise, compromising the model robustness and effectiveness. To better model both types of dynamic graphs, we propose UniDyG, a unified and effective representation learning approach, which scales to large dynamic graphs. We first propose a novel Fourier Graph Attention (FGAT) mechanism that can model local and global structural correlations based on recent neighbors and complex-number selective aggregation, while theoretically ensuring consistent representations of dynamic graphs over time. Based on approximation theory, we demonstrate that FGAT is well-suited to capture the underlying structures in CTDGs and DTDGs. We further enhance FGAT to resist temporal noise by designing an energy-gated unit, which adaptively filters out high-frequency noise according to the energy. Last, we leverage our FGAT mechanisms for temporal structure learning and employ the frequency-enhanced linear function for node-level dynamic updates, facilitating the generation of high-quality temporal embeddings. Extensive experiments show that our UniDyG achieves an average improvement of 14.4% over sixteen baselines across nine dynamic graphs.

[LG-55] Network Tomography with Path-Centric Graph Neural Network

链接: https://arxiv.org/abs/2502.16430
作者: Yuntong Hu,Junxiang Wang,Liang Zhao
类目: Machine Learning (cs.LG)
*备注: 13 pages, 6 figures

点击查看摘要

[LG-56] Active Learning Classification from a Signal Separation Perspective

链接: https://arxiv.org/abs/2502.16425
作者: Hrushikesh Mhaskar,Ryan O’Dowd,Efstratios Tsoukanis
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:

点击查看摘要

[LG-57] Worst-case Error Bounds for Online Learning of Smooth Functions

链接: https://arxiv.org/abs/2502.16388
作者: Weian Xie
类目: Machine Learning (cs.LG); Discrete Mathematics (cs.DM); Data Structures and Algorithms (cs.DS); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-58] Simultaneous Swap Regret Minimization via KL-Calibration

链接: https://arxiv.org/abs/2502.16387
作者: Haipeng Luo,Spandan Senapati,Vatsal Sharan
类目: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS); Computer Science and Game Theory (cs.GT)
*备注:

点击查看摘要

[LG-59] Verifying Classification with Limited Disclosure

链接: https://arxiv.org/abs/2502.16352
作者: Siddharth Bhandari,Liren Shan
类目: Machine Learning (cs.LG); Computers and Society (cs.CY); Data Structures and Algorithms (cs.DS)
*备注: 18 pages, 0 figures

点击查看摘要

[LG-60] Machine Learning-Based Cloud Computing Compliance Process Automation

链接: https://arxiv.org/abs/2502.16344
作者: Yuqing Wang,Xiao Yang
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注:

点击查看摘要

[LG-61] Generalization is not a universal guarantee: Estimating similarity to training data with an ensemble out-of-distribution metric

链接: https://arxiv.org/abs/2502.16329
作者: W. Max Schreyer(1, 2 and 3),Christopher Anderson(3),Reid F. Thompson(1, 2, 3, 4, 5 and 6) ((1) Computational Biology Program, Oregon Health amp; Science University, Portland, USA, (2) Department of Biomedical Engineering, Oregon Health amp; Science University, Portland, USA, (3) Portland VA Research Foundation, Portland, USA, (4) Department of Radiation Medicine, Oregon Health amp; Science University, Portland, USA, (5) Department of Medical Informatics and Clinical Epidemiology, Oregon Health amp; Science University, Portland, USA, (6) Division of Hospital and Specialty Medicine, VA Portland Healthcare System, Portland, USA)
类目: Machine Learning (cs.LG)
*备注: 10 pages, 5 figures

点击查看摘要

[LG-62] Risk-Averse Reinforcement Learning: An Optimal Transport Perspective on Temporal Difference Learning

链接: https://arxiv.org/abs/2502.16328
作者: Zahra Shahrooei,Ali Baheri
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-63] HetFS: A Method for Fast Similarity Search with Ad-hoc Meta-paths on Heterogeneous Information Networks

链接: https://arxiv.org/abs/2502.16288
作者: Xuqi Mao,Zhenyi Chen,Zhenying He,Yinan Jing,Kai Zhang,X. Sean Wang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-64] FHGE: A Fast Heterogeneous Graph Embedding with Ad-hoc Meta-paths

链接: https://arxiv.org/abs/2502.16281
作者: Xuqi Mao,Zhenying He,X. Sean Wang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-65] PLS-based approach for fair representation learning

链接: https://arxiv.org/abs/2502.16263
作者: Elena M. De-Diego,Adrián Perez-Suay,Paula Gordaliza,Jean-Michel Loubes
类目: Machine Learning (cs.LG); Computers and Society (cs.CY); Statistics Theory (math.ST); Machine Learning (stat.ML)
*备注:

点击查看摘要

Abstract:We revisit the problem of fair representation learning by proposing Fair Partial Least Squares (PLS) components. PLS is widely used in statistics to efficiently reduce the dimension of the data by providing representation tailored for the prediction. We propose a novel method to incorporate fairness constraints in the construction of PLS components. This new algorithm provides a feasible way to construct such features both in the linear and the non linear case using kernel embeddings. The efficiency of our method is evaluated on different datasets, and we prove its superiority with respect to standard fair PCA method.

[LG-66] Verifying Quantized Graph Neural Networks is PSPACE-complete

链接: https://arxiv.org/abs/2502.16244
作者: Marco Sälzer,François Schwarzentruber,Nicolas Troquard
类目: Logic in Computer Science (cs.LO); Computational Complexity (cs.CC); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-67] Separated Contrastive Learning for Matching in Cross-domain Recommendation with Curriculum Scheduling

链接: https://arxiv.org/abs/2502.16239
作者: Heng Chang,Liang Gu,Cheng Hu,Zhinan Zhang,Hong Zhu,Yuhui Xu,Yuan Fang,Zhen Chen
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
*备注: Accepted by TheWebConf 2025

点击查看摘要

[LG-68] Flow-based linear embedding for Bayesian filtering of nonlinear stochastic dynamical systems

链接: https://arxiv.org/abs/2502.16232
作者: Xintong Wang,Xiaofei Guan,Ling Guo,Hao Wu
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-69] Learning Humanoid Locomotion with World Model Reconstruction

链接: https://arxiv.org/abs/2502.16230
作者: Wandong Sun,Long Chen,Yongbo Su,Baoshi Cao,Yang Liu,Zongwu Xie
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-70] Interpreting core forms of urban morphology linked to urban functions with explainable graph neural network

链接: https://arxiv.org/abs/2502.16210
作者: Dongsheng Chen,Yu Feng,Xun Li,Mingya Qu,Peng Luo,Liqiu Meng
类目: Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-71] Co-evolution-based Metal-binding Residue Prediction with Graph Neural Networks

链接: https://arxiv.org/abs/2502.16189
作者: Sayedmohammadreza Rastegari,Sina Tabakhi,Xianyuan Liu,Wei Sang,Haiping Lu
类目: Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci); Biomolecules (q-bio.BM); Quantitative Methods (q-bio.QM)
*备注: 7 pages, 3 figures

点击查看摘要

[LG-72] Practical programming research of Linear DML model based on the simplest Python code: From the standpoint of novice researchers

链接: https://arxiv.org/abs/2502.16172
作者: Shunxin Yao
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注: 12 pages, 4 tables, 3 figures

点击查看摘要

[LG-73] Advanced Text Analytics – Graph Neural Network for Fake News Detection in Social Media

链接: https://arxiv.org/abs/2502.16157
作者: Anantram Patel,Vijay Kumar Sutrakar
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-74] DUPRE: Data Utility Prediction for Efficient Data Valuation AAMAS2025

链接: https://arxiv.org/abs/2502.16152
作者: Kieu Thao Nguyen Pham,Rachael Hwee Ling Sim,Quoc Phong Nguyen,See Kiong Ng,Bryan Kian Hsiang Low
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
*备注: 16 pages, 7 figures, the paper got accepted AAMAS 2025

点击查看摘要

[LG-75] An Improved Deep Learning Model for Word Embeddings Based Clustering for Large Text Datasets

链接: https://arxiv.org/abs/2502.16139
作者: Vijay Kumar Sutrakar,Nikhil Mogre
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-76] FedOC: Optimizing Global Prototypes with Orthogonality Constraints for Enhancing Embeddings Separation in Heterogeneous Federated Learning

链接: https://arxiv.org/abs/2502.16119
作者: Fucheng Guo,Zeyu Luan,Qing Li,Dan Zhao,Yong Jiang
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注:

点击查看摘要

[LG-77] Integrating Weather Station Data and Radar for Precipitation Nowcasting: SmaAt-fUsion and SmaAt-Krige-GNet

链接: https://arxiv.org/abs/2502.16116
作者: Aleksej Cornelissen,Jie Shi,Siamak Mehrkanoon
类目: Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注: 11 pages, 7 figures

点击查看摘要

[LG-78] Category-free Out-of-Distribution Node Detection with Feature Resonance

链接: https://arxiv.org/abs/2502.16076
作者: Shenzhi Yang,Junbo Zhao,Shouqing Yang,Yixuan Li,Dingyu Yang,Xiaofang Zhang,Haobo Wang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-79] Implicit Bias of Gradient Descent for Non-Homogeneous Deep Networks

链接: https://arxiv.org/abs/2502.16075
作者: Yuhang Cai,Kangjie Zhou,Jingfeng Wu,Song Mei,Michael Lindsey,Peter L. Bartlett
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注: 96 pages

点击查看摘要

[LG-80] Single Domain Generalization with Model-aware Parametric Batch-wise Mixup

链接: https://arxiv.org/abs/2502.16064
作者: Marzi Heidari,Yuhong Guo
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-81] Since Faithfulness Fails: The Performance Limits of Neural Causal Discovery

链接: https://arxiv.org/abs/2502.16056
作者: Mateusz Olko,Mateusz Gajewski,Joanna Wojciechowska,Mikołaj Morzy,Piotr Sankowski,Piotr Miłoś
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 19 pages, 12 figures

点击查看摘要

[LG-82] Quasi Zigzag Persistence: A Topological Framework for Analyzing Time-Varying Data

链接: https://arxiv.org/abs/2502.16049
作者: Tamal K. Dey,Shreyas N. Samaga
类目: Machine Learning (cs.LG); Algebraic Topology (math.AT)
*备注:

点击查看摘要

[LG-83] Learning Neural Networks with Distribution Shift: Efficiently Certifiable Guarantees ICLR2025

链接: https://arxiv.org/abs/2502.16021
作者: Gautam Chandrasekaran,Adam R. Klivans,Lin Lin Lee,Konstantinos Stavropoulos
类目: Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
*备注: To appear in The Thirteenth International Conference on Learning Representations (ICLR 2025) 38 pages

点击查看摘要

[LG-84] News Sentiment as a Predictor for American Domestic Migration

链接: https://arxiv.org/abs/2502.15998
作者: Benjamin Lane,Simeon Sayer
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注: 8 pages, 3 figures

点击查看摘要

[LG-85] Human Guided Learning of Transparent Regression Models

链接: https://arxiv.org/abs/2502.15992
作者: Lukas Pensel,Stefan Kramer
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-86] Mean-Shift Distillation for Diffusion Mode Seeking

链接: https://arxiv.org/abs/2502.15989
作者: Vikas Thamizharasan,Nikitas Chatzis,Iliyan Georgiev,Matthew Fisher,Difan Liu,Nanxuan Zhao,Evangelos Kalogerakis,Michal Lukac
类目: Machine Learning (cs.LG); Graphics (cs.GR)
*备注: 12 pages, 8 figures

点击查看摘要

[LG-87] Near Optimal Decision Trees in a SPLIT Second

链接: https://arxiv.org/abs/2502.15988
作者: Varun Babbar,Hayden McTavish,Cynthia Rudin,Margo Seltzer
类目: Machine Learning (cs.LG)
*备注: Currently under review

点击查看摘要

[LG-88] CoRe: Coherency Regularization for Hierarchical Time Series

链接: https://arxiv.org/abs/2502.15983
作者: Rares Cristian,Pavithra Harhsa,Georgia Perakis,Brian Quanz
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-89] Enhancing PPO with Trajectory-Aware Hybrid Policies

链接: https://arxiv.org/abs/2502.15968
作者: Qisai Liu,Zhanhong Jiang,Hsin-Jung Yang,Mahsa Khosravi,Joshua R. Waite,Soumik Sarkar
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-90] owards Efficient Contrastive PAC Learning

链接: https://arxiv.org/abs/2502.15962
作者: Jie Shen
类目: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-91] owards Understanding Gradient Flow Dynamics of Homogeneous Neural Networks Beyond the Origin

链接: https://arxiv.org/abs/2502.15952
作者: Akshay Kumar,Jarvis Haupt
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注:

点击查看摘要

[LG-92] Orthogonal Calibration for Asynchronous Federated Learning

链接: https://arxiv.org/abs/2502.15940
作者: Jiayun Zhang,Shuheng Li,Haiyu Huang,Xiaofan Yu,Rajesh K. Gupta,Jingbo Shang
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-93] On the Design of Safe Continual RL Methods for Control of Nonlinear Systems

链接: https://arxiv.org/abs/2502.15922
作者: Austin Coursey,Marcos Quinones-Grueiro,Gautam Biswas
类目: Machine Learning (cs.LG); Robotics (cs.RO)
*备注:

点击查看摘要

[LG-94] Connecting the geometry and dynamics of many-body complex systems with message passing neural operators

链接: https://arxiv.org/abs/2502.15913
作者: Nicholas A. Gabriel,Neil F. Johnson,George Em Karniadakis
类目: Machine Learning (cs.LG); Physics and Society (physics.soc-ph)
*备注:

点击查看摘要

[LG-95] S-OOD: Evaluating Time-Series Out-of-Distribution Detection and Prospective Directions for Progress AAAI-25

链接: https://arxiv.org/abs/2502.15901
作者: Onat Gungor,Amanda Sofie Rios,Nilesh Ahuja,Tajana Rosing
类目: Machine Learning (cs.LG)
*备注: Accepted for an oral presentation at AAAI-25 AI4TS

点击查看摘要

[LG-96] Explaining the Success of Nearest Neighbor Methods in Prediction

链接: https://arxiv.org/abs/2502.15900
作者: George H. Chen,Devavrat Shah
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: Originally published on May 31, 2018 in Foundations and Trends in Machine Learning; this revised version fixes some proof details for k-NN and fixed-radius NN regression and classification

点击查看摘要

[LG-97] Efficient Estimation of Shortest-Path Distance Distributions to Samples in Graphs

链接: https://arxiv.org/abs/2502.15890
作者: Alan Zhu,Jiaqi Ma,Qiaozhu Mei
类目: ocial and Information Networks (cs.SI); Machine Learning (cs.LG)
*备注:

点击查看摘要

Abstract:As large graph datasets become increasingly common across many fields, sampling is often needed to reduce the graphs into manageable sizes. This procedure raises critical questions about representativeness as no sample can capture the properties of the original graph perfectly, and different parts of the graph are not evenly affected by the loss. Recent work has shown that the distances from the non-sampled nodes to the sampled nodes can be a quantitative indicator of bias and fairness in graph machine learning. However, to our knowledge, there is no method for evaluating how a sampling method affects the distribution of shortest-path distances without actually performing the sampling and shortest-path calculation. In this paper, we present an accurate and efficient framework for estimating the distribution of shortest-path distances to the sample, applicable to a wide range of sampling methods and graph structures. Our framework is faster than empirical methods and only requires the specification of degree distributions. We also extend our framework to handle graphs with community structures. While this introduces a decrease in accuracy, we demonstrate that our framework remains highly accurate on downstream comparison-based tasks. Code is publicly available at this https URL. Subjects: Social and Information Networks (cs.SI); Machine Learning (cs.LG) Cite as: arXiv:2502.15890 [cs.SI] (or arXiv:2502.15890v1 [cs.SI] for this version) https://doi.org/10.48550/arXiv.2502.15890 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

[LG-98] Deep Reinforcement Learning-Based Bidding Strategies for Prosumers Trading in Double Auction-Based Transactive Energy Market

链接: https://arxiv.org/abs/2502.15774
作者: Jun Jiang,Yuanliang Li,Luyang Hou,Mohsen Ghafouri,Peng Zhang,Jun Yan,Yuhong Liu
类目: ystems and Control (eess.SY); Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-99] JExplore: Design Space Exploration Tool for Nvidia Jetson Boards

链接: https://arxiv.org/abs/2502.15773
作者: Basar Kutukcu,Sinan Xie,Sabur Baidya,Sujit Dey
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: 4 pages, 4 figures

点击查看摘要

[LG-100] Asymptotic evaluation of the information processing capacity in reservoir computing

链接: https://arxiv.org/abs/2502.15769
作者: Yohei Saito
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-101] Digi-Q: Learning Q-Value Functions for Training Device-Control Agents ICLR2025

链接: https://arxiv.org/abs/2502.15760
作者: Hao Bai,Yifei Zhou,Li Erran Li,Sergey Levine,Aviral Kumar
类目: Machine Learning (cs.LG)
*备注: Accepted to ICLR 2025

点击查看摘要

[LG-102] RKM: Twin Restricted Kernel Machines for Classification and Regression

链接: https://arxiv.org/abs/2502.15759
作者: A. Quadir,M. Tanveer
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-103] Maturity Framework for Enhancing Machine Learning Quality

链接: https://arxiv.org/abs/2502.15758
作者: Angelantonio Castelli,Georgios Christos Chouliaras,Dmitri Goldenberg
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注:

点击查看摘要

[LG-104] Causal Covariate Shift Correction using Fisher information penalty

链接: https://arxiv.org/abs/2502.15756
作者: Behraj Khan,Behroz Mirza,Tahir Syed
类目: Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-105] CacheMamba: Popularity Prediction for Mobile Edge Caching Networks via Selective State Spaces

链接: https://arxiv.org/abs/2502.15746
作者: Ghazaleh Kianfar,Zohreh Hajiakhondi-Meybodi,Arash Mohammadi
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注:

点击查看摘要

[LG-106] Channel Gain Map Construction based on Subregional Learning and Prediction

链接: https://arxiv.org/abs/2502.15733
作者: Jiayi Chen,Ruifeng Gao,Jue Wang,Shu Sun,Yi Wu
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-107] Making Sense of Data in the Wild: Data Analysis Automation at Scale

链接: https://arxiv.org/abs/2502.15718
作者: Mara Graziani,Malina Molnar,Irina Espejo Morales,Joris Cadow-Gossweiler,Teodoro Laino
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-108] A Statistical Learning Approach for Feature-Aware Task-to-Core Allocation in Heterogeneous Platforms

链接: https://arxiv.org/abs/2502.15716
作者: Mohammad Pivezhandi,Abusayeed Saifullah,Prashant Modekurthy
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: 12 pages, 7 figures

点击查看摘要

[LG-109] On Quantile Regression Forests for Modelling Mixed-Frequency and Longitudinal Data

链接: https://arxiv.org/abs/2502.17137
作者: Mila Andreani
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
*备注: PhD Thesis

点击查看摘要

[LG-110] Gabor-Enhanced Physics-Informed Neural Networks for Fast Simulations of Acoustic Wavefields

链接: https://arxiv.org/abs/2502.17134
作者: Mohammad Mahdi Abedi,David Pardo,Tariq Alkhalifah
类目: Geophysics (physics.geo-ph); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-111] Random Projections and Natural Sparsity in Time-Series Classification: A Theoretical Analysis

链接: https://arxiv.org/abs/2502.17061
作者: Jorge Marco-Blanco,Rubén Cuevas
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-112] Your Assumed DAG is Wrong and Heres How To Deal With It

链接: https://arxiv.org/abs/2502.17030
作者: Kirtan Padh,Zhufeng Li,Cecilia Casolo,Niki Kilbertus
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-113] Active Learning for Conditional Inverse Design with Crystal Generation and Foundation Atomic Models

链接: https://arxiv.org/abs/2502.16984
作者: Zhuoyuan Li,Siyu Liu,Beilin Ye,David J. Srolovitz,Tongqi Wen
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-114] Provable Benefits of Unsupervised Pre-training and Transfer Learning via Single-Index Models

链接: https://arxiv.org/abs/2502.16849
作者: Taj Jones-McCormick,Aukosh Jagannath,Subhabrata Sen
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-115] Finite-Sample Analysis of Policy Evaluation for Robust Averag e Reward Reinforcement Learning

链接: https://arxiv.org/abs/2502.16816
作者: Yang Xu,Washim Uddin Mondal,Vaneet Aggarwal
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-116] A new framework for X-ray absorption spectroscopy data analysis based on machine learning: XASDAML

链接: https://arxiv.org/abs/2502.16665
作者: Xue Han,Haodong Yao,Fei Zhan,Xueqi Song,Junfang Zhao,Haifeng Zhao
类目: Computational Physics (physics.comp-ph); Machine Learning (cs.LG)
*备注:

点击查看摘要

Abstract:X-ray absorption spectroscopy (XAS) is a powerful technique to probe the electronic and structural properties of materials. With the rapid growth in both the volume and complexity of XAS datasets driven by advancements in synchrotron radiation facilities, there is an increasing demand for advanced computational tools capable of efficiently analyzing large-scale data. To address these needs, we introduce XASDAML,a flexible, machine learning based framework that integrates the entire data-processing workflow-including dataset construction for spectra and structural descriptors, data filtering, ML modeling, prediction, and model evaluation-into a unified platform. Additionally, it supports comprehensive statistical analysis, leveraging methods such as principal component analysis and clustering to reveal potential patterns and relationships within large datasets. Each module operates independently, allowing users to modify or upgrade modules in response to evolving research needs or technological advances. Moreover, the platform provides a user-friendly interface via Jupyter Notebook, making it accessible to researchers at varying levels of expertise. The versatility and effectiveness of XASDAML are exemplified by its application to a copper dataset, where it efficiently manages large and complex data, supports both supervised and unsupervised machine learning models, provides comprehensive statistics for structural descriptors, generates spectral plots, and accurately predicts coordination numbers and bond lengths. Furthermore, the platform streamlining the integration of XAS with machine learning and lowering the barriers to entry for new users.

[LG-117] ransformations of predictions and realizations in consistent scoring functions

链接: https://arxiv.org/abs/2502.16542
作者: Hristos Tyralis,Georgia Papacharalampous
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 28 pages

点击查看摘要

[LG-118] Update hydrological states or meteorological forcings? Comparing data assimilation methods for differentiable hydrologic models

链接: https://arxiv.org/abs/2502.16444
作者: Amirmoez Jamaat,Yalan Song,Farshid Rahmani,Jiangtao Liu,Kathryn Lawson,Chaopeng Shen
类目: Fluid Dynamics (physics.flu-dyn); Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注:

点击查看摘要

[LG-119] Subspace Recovery in Winsorized PCA: Insights into Accuracy and Robustness

链接: https://arxiv.org/abs/2502.16391
作者: Sangil Han,Kyoowon Kim,Sungkyu Jung
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Applications (stat.AP); Methodology (stat.ME)
*备注:

点击查看摘要

[LG-120] Software defined demodulation of multiple frequency shift keying with dense neural network for weak signal communications

链接: https://arxiv.org/abs/2502.16371
作者: Mykola Kozlenko,Vira Vialkova
类目: ignal Processing (eess.SP); Machine Learning (cs.LG)
*备注: 6 pages, 9 figures

点击查看摘要

[LG-121] Monotonicity Testing of High-Dimensional Distributions with Subcube Conditioning

链接: https://arxiv.org/abs/2502.16355
作者: Deeparnab Chakrabarty,Xi Chen,Simeon Ristic,C. Seshadhri,Erik Waingarten
类目: atistics Theory (math.ST); Computational Complexity (cs.CC); Discrete Mathematics (cs.DM); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-122] Rectifying Conformity Scores for Better Conditional Coverag e

链接: https://arxiv.org/abs/2502.16336
作者: Vincent Plassier,Alexander Fishkov,Victor Dheur,Mohsen Guizani,Souhaib Ben Taieb,Maxim Panov,Eric Moulines
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注:

点击查看摘要

[LG-123] Sampling through Algorithmic Diffusion in non-convex Perceptron problems

链接: https://arxiv.org/abs/2502.16292
作者: Elizaveta Demyanenko,Davide Straziota,Carlo Baldassi,Carlo Lucibello
类目: Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-124] Brain-Model Evaluations Need the NeuroAI Turing Test

链接: https://arxiv.org/abs/2502.16238
作者: Jenelle Feather,Meenakshi Khosla,N. Apurva Ratan Murty,Aran Nayebi
类目: Neurons and Cognition (q-bio.NC); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 9 pages, 4 figures, 2 tables

点击查看摘要

[LG-125] Statistical Inference in Reinforcement Learning: A Selective Survey

链接: https://arxiv.org/abs/2502.16195
作者: Chengchun Shi
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-126] A Review of Causal Decision Making

链接: https://arxiv.org/abs/2502.16156
作者: Lin Ge,Hengrui Cai,Runzhe Wan,Yang Xu,Rui Song
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:

点击查看摘要

[LG-127] Contrastive Similarity Learning for Market Forecasting: The ContraSim Framework

链接: https://arxiv.org/abs/2502.16023
作者: Nicholas Vinden,Raeid Saqur,Zining Zhu,Frank Rudzicz
类目: atistical Finance (q-fin.ST); Machine Learning (cs.LG)
*备注: 8 pages, 3 appendices

点击查看摘要

[LG-128] Exact Recovery of Sparse Binary Vectors from Generalized Linear Measurements

链接: https://arxiv.org/abs/2502.16008
作者: Arya Mazumdar,Neha Sangwan
类目: Machine Learning (stat.ML); Information Theory (cs.IT); Machine Learning (cs.LG)
*备注:

点击查看摘要

信息检索

[IR-0] FilterLLM : Text-To-Distribution LLM for Billion-Scale Cold-Start Recommendation

链接: https://arxiv.org/abs/2502.16924
作者: Ruochen Liu,Hao Chen,Yuanchen Bei,Zheyu Zhou,Lijia Chen,Qijie Shen,Feiran Huang,Fakhri Karray,Senzhang Wang
类目: Information Retrieval (cs.IR)
*备注: 12 pages

点击查看摘要

[IR-1] Multimodal Search in Chemical Documents and Reactions SIGIR2025

链接: https://arxiv.org/abs/2502.16865
作者: Ayush Kumar Shah,Abhisek Dey,Leo Luo,Bryan Amador,Patrick Philippy,Ming Zhong,Siru Ouyang,David Mark Friday,David Bianchi,Nick Jackson,Richard Zanibbi,Jiawei Han
类目: Information Retrieval (cs.IR)
*备注: 4 pages, 2 figures, SIGIR 2025 Demonstration Submission

点击查看摘要

[IR-2] he Blessing of Reasoning : LLM -Based Contrastive Explanations in Black-Box Recommender Systems

链接: https://arxiv.org/abs/2502.16759
作者: Yuyan Wang,Pan Li,Minmin Chen
类目: Information Retrieval (cs.IR)
*备注:

点击查看摘要

[IR-3] Design and Implementation of a Scalable Clinical Data Warehouse for Resource-Constrained Healthcare Systems

链接: https://arxiv.org/abs/2502.16674
作者: Shovito Barua Soumma,Fahim Shahriar,Umme Niraj Mahi,Md Hasin Abrar,Md Abdur Rahman Fahad,Abu Sayed Md. Latiful Hoque
类目: Information Retrieval (cs.IR)
*备注:

点击查看摘要

[IR-4] Unified Semantic and ID Representation Learning for Deep Recommenders

链接: https://arxiv.org/abs/2502.16474
作者: Guanyu Lin,Zhigang Hua,Tao Feng,Shuang Yang,Bo Long,Jiaxuan You
类目: Information Retrieval (cs.IR)
*备注:

点击查看摘要

[IR-5] Exploiting Epistemic Uncertainty in Cold-Start Recommendation Systems

链接: https://arxiv.org/abs/2502.16256
作者: Yang Xiang,Li Fan,Chenke Yin,Menglin Kong,Chengtao Ji
类目: Information Retrieval (cs.IR)
*备注:

点击查看摘要

[IR-6] ag-Pag: A Dedicated Tool for Systematic Web Page Annotations

链接: https://arxiv.org/abs/2502.16150
作者: Anton Pogrebnjak,Julian Schelb,Andreas Spitz,Celina Kacperski,Roberto Ulloa
类目: Information Retrieval (cs.IR)
*备注:

点击查看摘要

[IR-7] Semantic Gaussian Mixture Variational Autoencoder for Sequential Recommendation DASFAA2025

链接: https://arxiv.org/abs/2502.16140
作者: Beibei Li,Tao Xiang,Beihong Jin,Yiyuan Zheng,Rui Zhao
类目: Information Retrieval (cs.IR)
*备注: Accepted by DASFAA 2025

点击查看摘要

[IR-8] ESANS: Effective and Semantic-Aware Negative Sampling for Large-Scale Retrieval Systems

链接: https://arxiv.org/abs/2502.16077
作者: Haibo Xing,Kanefumi Matsuyama,Hao Deng,Jinxin Hu,Yu Zhang,Xiaoyi Zeng
类目: Information Retrieval (cs.IR)
*备注: 10 pages, 6 figures, Proceedings of the ACM Web Conference 2025

点击查看摘要

[IR-9] Joint Similarity Item Exploration and Overlapped User Guidance for Multi-Modal Cross-Domain Recommendation

链接: https://arxiv.org/abs/2502.16068
作者: Weiming Liu,Chaochao Chen,Jiahe Xu,Xinting Liao,Fan Wang,Xiaolin Zheng,Zhihui Fu,Ruiguang Pei,Jun Wang
类目: Information Retrieval (cs.IR)
*备注:

点击查看摘要

[IR-10] Open-Source Retrieval Augmented Generation Framework for Retrieving Accurate Medication Insights from Formularies for African Healthcare Workers

链接: https://arxiv.org/abs/2502.15722
作者: Axum AI:J. Owoyemi,S. Abubakar,A. Owoyemi,T.O. Togunwa,F.C. Madubuko,S. Oyatoye,Z. Oyetolu,K. Akyea,A.O. Mohammed,A. Adebakin
类目: Information Retrieval (cs.IR)
*备注: 4 pages, 2 tables and 3 figures

点击查看摘要

[IR-11] A Survey on Multimodal Recommender Systems: Recent Advances and Future Directions

链接: https://arxiv.org/abs/2502.15711
作者: Jinfeng Xu,Zheyu Chen,Shuo Yang,Jinze Li,Wei Wang,Xiping Hu,Steven Hoi,Edith Ngai
类目: Information Retrieval (cs.IR); Multimedia (cs.MM)
*备注:

点击查看摘要

附件下载

点击下载今日全部论文列表