AI Agent - arXivDaily 专题

2503.00841 2025-03-04 cs.AI 57%

A Law Reasoning Benchmark for LLM with Tree-Organized Structures including Factum Probandum, Evidence and Experiences

Jiaxin Shen, Jinan Xu, Huiqi Hu, Luyi Lin, Fei Zheng, Guoyang Ma, Fandong Meng, Jie Zhou, Wenjuan Han

机构 * Beijing Jiaotong University（北京交通大学）； Tencent Inc（腾讯公司）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments 20 pages, 13 figures

2405.20986 2025-03-04 cs.LG cs.CV 57%

Predictive Uncertainty Quantification for Bird's Eye View Segmentation: A Benchmark and Novel Loss Function

Linlin Yu, Bowen Yang, Tianhao Wang, Kangshuo Li, Feng Chen

机构 * The University of Texas at Dallas（德克萨斯大学达拉斯分校）； Cypress Woods High School（赛普拉斯伍兹高中）

专题命中 Agent评测：planning(abstract)；分类 cs.LG

Comments ICLR 2025

2502.20635 2025-03-03 cs.HC cs.LG 57%

Can LLM Assist in the Evaluation of the Quality of Machine Learning Explanations?

Bo Wang, Yiqiao Li, Jianlong Zhou, Fang Chen

机构 * Data Science Institute, University of Technology Sydney（悉尼科技大学数据科学学院）

专题命中 Agent评测：workflow(abstract)；分类 cs.LG

2502.19701 2025-02-28 cs.SI cs.AI cs.CY 57%

Extending the Hegselmann-Krause Model of Opinion Dynamics to include AI Oracles

Allen G. Rodrigo

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2407.06423 2025-02-28 cs.AI 57%

InsightBench: Evaluating Business Analytics Agents Through Multi-Step Insight Generation

Gaurav Sahu, Abhay Puri, Juan Rodriguez, Amirhossein Abaskohi, Mohammad Chegini, Alexandre Drouin, Perouz Taslakian, Valentina Zantedeschi, Alexandre Lacoste, David Vazquez, Nicolas Chapados, Christopher Pal, Sai Rajeswar Mudumba, Issam Hadj Laradji

机构 * ServiceNow Research（ServiceNow研究院）； Mila - Quebec AI Institute（米拉-魁北克人工智能研究所）； École de Technologie Supérieure（高等技术学院）； University of British Columbia（不列颠哥伦比亚大学）； University of Waterloo（滑铁卢大学）； University of Victoria（维多利亚大学）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments Accepted to ICLR 2025

2301.06086 2025-02-28 cs.GT cs.AI cs.MA 57%

Constrained Serial Dictatorships can be Fair

Sylvain Bouveret, Hugo Gilbert, Jérôme Lang, Guillaume Méroué

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2502.18754 2025-02-27 cs.IR cs.AI 57%

AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platforms

Yuwei Yan, Yu Shang, Qingbin Zeng, Yu Li, Keyu Zhao, Zhiheng Zheng, Xuefei Ning, Tianji Wu, Shengen Yan, Yu Wang, Fengli Xu, Yong Li

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments 8 pages, 10 figures, in Proceedings of the ACM Web Conference 2025 (WWW '25)

2410.14141 2025-02-26 cs.RO cs.CL 57%

Coherence-Driven Multimodal Safety Dialogue with Active Learning for Embodied Agents

Sabit Hassan, Hye-Young Chung, Xiang Zhi Tan, Malihe Alikhani

机构 * University of Pittsburgh（匹兹堡大学）； Northeastern University（东北大学）

专题命中 Agent评测：agent(abstract)；分类 cs.CL

Comments To appear at AAMAS, 2025

2405.13861 2025-02-26 cs.LG 57%

Transformers Can Learn Temporal Difference Methods for In-Context Reinforcement Learning

Jiuqi Wang, Ethan Blaser, Hadi Daneshmand, Shangtong Zhang

专题命中 Agent评测：agent(abstract)；分类 cs.LG

2405.16581 2025-02-25 cs.LG 57%

On Bits and Bandits: Quantifying the Regret-Information Trade-off

Itai Shufaro, Nadav Merlis, Nir Weinberger, Shie Mannor

机构 * Technion（以色列理工学院）； ENSAE（法国国立统计与经济管理学校）； NVIDIA Research（英伟达研究院）

专题命中 Agent评测：agent(abstract)；分类 cs.LG

2502.16175 2025-02-25 cs.CV cs.AI cs.GR 57%

Mojito: LLM-Aided Motion Instructor with Jitter-Reduced Inertial Tokens

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（德莫斯科技）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments First three authors contribute equally. Project page: https://koyui.github.io/mojito/

2502.14676 2025-02-25 cs.CV cs.AI 57%

BP-SGCN: Behavioral Pseudo-Label Informed Sparse Graph Convolution Network for Pedestrian and Heterogeneous Trajectory Prediction

Ruochen Li, Stamos Katsigiannis, Tae-Kyun Kim, Hubert P. H. Shum

机构 * Durham University（杜伦大学）； KAIST（韩国科学技术院）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2502.14966 2025-02-24 cs.CR cs.AI 57%

CyberSentinel: An Emergent Threat Detection System for AI Security

Krti Tallam

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2502.14892 2025-02-24 cs.CV cs.AI 57%

EgoSpeak: Learning When to Speak for Egocentric Conversational Agents in the Wild

Junhyeok Kim, Min Soo Kim, Jiwan Chung, Jungbin Cho, Jisoo Kim, Sungwoong Kim, Gyeongbo Sim, Youngjae Yu

机构 * Yonsei University（延世大学）； NCSOFT Corporation（NCsoft公司）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments NAACL 2025 Findings. Project page at https://jun297.github.io/EgoSpeak/

2502.13996 2025-02-21 cs.LG 57%

Beyond Single-Value Metrics: Evaluating and Enhancing LLM Unlearning with Cognitive Diagnosis

Yicheng Lang, Kehan Guo, Yue Huang, Yujun Zhou, Haomin Zhuang, Tianyu Yang, Yao Su, Xiangliang Zhang

机构 * University of Notre Dame（圣母大学）； Worcester Polytechnic Institute（伍斯特理工学院）

专题命中 Agent评测：agent(abstract)；分类 cs.LG

2412.18297 2025-02-21 cs.GT cs.LG 57%

Learning to Play Against Unknown Opponents

Eshwar Ram Arunachaleswaran, Natalie Collina, Jon Schneider

机构 * University of Pennsylvania（宾夕法尼亚大学）； Simons Institute for the Theory of Computing（西蒙斯理论计算研究所）； Google Research（谷歌研究院）

专题命中 Agent评测：agent(abstract)；分类 cs.LG

2412.02861 2025-02-21 stat.ML cs.LG 57%

An Information-Theoretic Analysis of Thompson Sampling for Logistic Bandits

Amaury Gouverneur, Borja Rodríguez-Gálvez, Tobias J. Oechtering, Mikael Skoglund

机构 * KTH Royal Institute of Technology（瑞典皇家理工学院）

专题命中 Agent评测：agent(abstract)；分类 cs.LG

Comments 21 pages, under review

2502.13569 2025-02-20 cs.AI 57%

Model Evolution Framework with Genetic Algorithm for Multi-Task Reinforcement Learning

Yan Yu, Wengang Zhou, Yaodong Yang, Wanxuan Lu, Yingyan Hou, Houqiang Li

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2502.13550 2025-02-20 cs.CL 57%

STaR-SQL: Self-Taught Reasoner for Text-to-SQL

Mingqian He, Yongliang Shen, Wenqi Zhang, Qiuying Peng, Jun Wang, Weiming Lu

机构 * Zhejiang University（浙江大学）； OPPO Research Institute（OPPO研究院）

专题命中 Agent评测：agent(abstract)；分类 cs.CL

2402.12812 2025-02-20 cs.LG cs.DC 57%

Scalable Decentralized Algorithms for Online Personalized Mean Estimation

Franco Galante, Giovanni Neglia, Emilio Leonardi

专题命中 Agent评测：agent(abstract)；分类 cs.LG

2502.12926 2025-02-19 cs.AI 57%

Towards more Contextual Agents: An extractor-Generator Optimization Framework

Mourad Aouini, Jinan Loubani

机构 * CGI（CGI公司）； CNRS（法国国家科学研究中心）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2502.12458 2025-02-19 cs.CL 57%

An Empirical Evaluation of Encoder Architectures for Fast Real-Time Long Conversational Understanding

Annamalai Senthilnathan, Kristjan Arumae, Mohammed Khalilia, Zhengzheng Xing, Aaron R. Colak

机构 * Qualtrics Inc.（昆泰睿公司）

专题命中 Agent评测：agent(abstract)；分类 cs.CL

2502.11083 2025-02-18 cs.CL 57%

Streamlining the Collaborative Chain of Models into A Single Forward Pass in Generation-Based Tasks

Yuanjie Lyu, Chao Zhang, Yuhao Chen, Yong Chen, Tong Xu

机构 * University of Science and Technology of China（中国科学技术大学）

专题命中 Agent评测：agent(abstract)；分类 cs.CL

2412.16516 2025-02-18 cs.CL cs.HC 57%

HammerBench: Fine-Grained Function-Calling Evaluation in Real Mobile Device Scenarios

Jun Wang, Jiamu Zhou, Muning Wen, Xiaoyun Mo, Haoyu Zhang, Qiqiang Lin, Cheng Jin, Xihuai Wang, Weinan Zhang, Qiuying Peng, Jun Wang

机构 * OPPO

专题命中 Agent评测：agent(abstract)；分类 cs.CL

2408.04203 2025-02-18 cs.AI 57%

MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents

Yanqi Dai, Huanran Hu, Lei Wang, Shengjie Jin, Xu Chen, Zhiwu Lu

机构 * Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）； College of Information and Electrical Engineering, China Agricultural University（中国农业大学信息与电气工程学院）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments Accepted for the 13th International Conference on Learning Representations (ICLR 2025)

2408.05146 2025-02-18 cs.LG cs.GT cs.MA 57%

Performative Prediction on Games and Mechanism Design

António Góis, Mehrnaz Mofakhami, Fernando P. Santos, Gauthier Gidel, Simon Lacoste-Julien

机构 * Mila（米拉研究所）； Université de Montréal（蒙特利尔大学）； University of Amsterdam（阿姆斯特丹大学）

专题命中 Agent评测：agent(abstract)；分类 cs.LG

Comments Accepted to AISTATS 2025; code available at https://github.com/antoniogois/performative-games

2502.08966 2025-02-17 cs.CR cs.AI 57%

RTBAS: Defending LLM Agents Against Prompt Injection and Privacy Leakage

Peter Yong Zhong, Siyuan Chen, Ruiqi Wang, McKenna McCall, Ben L. Titzer, Heather Miller, Phillip B. Gibbons

机构 * Carnegie Mellon University（卡内基梅隆大学）； Two Sigma Investments（Two Sigma投资公司）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

2502.09060 2025-02-14 cs.SE cs.CY cs.SI 57%

Anchor Sponsor Firms in Open Source Software Ecosystems

Brigitta Németh, Johannes Wachs

专题命中 Agent评测：agent(abstract)；分类 cs.SE

2502.08856 2025-02-14 cs.LG 57%

A Systematic Evaluation of Generative Models on Tabular Transportation Data

Chengen Wang, Alvaro Cardenas, Gurcan Comert, Murat Kantarcioglu

机构 * University of Texas at Dallas（德克萨斯大学达拉斯分校）； University of California, Santa Cruz（加利福尼亚大学圣克鲁兹分校）； Benedict College（本尼迪克特学院）

专题命中 Agent评测：planning(abstract)；分类 cs.LG

2410.21280 2025-02-12 q-fin.TR cs.AI 57%

TraderTalk: An LLM Behavioural ABM applied to Simulating Human Bilateral Trading Interactions

Alicia Vidler, Toby Walsh

机构 * UNSW（新南威尔士大学）

专题命中 Agent评测：agent(abstract)；分类 cs.AI

Comments 4 pages

1. Agent评测 15866 篇

A Law Reasoning Benchmark for LLM with Tree-Organized Structures including Factum Probandum, Evidence and Experiences

Predictive Uncertainty Quantification for Bird's Eye View Segmentation: A Benchmark and Novel Loss Function

Can LLM Assist in the Evaluation of the Quality of Machine Learning Explanations?

Extending the Hegselmann-Krause Model of Opinion Dynamics to include AI Oracles

InsightBench: Evaluating Business Analytics Agents Through Multi-Step Insight Generation

Constrained Serial Dictatorships can be Fair

AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platforms

Coherence-Driven Multimodal Safety Dialogue with Active Learning for Embodied Agents

Transformers Can Learn Temporal Difference Methods for In-Context Reinforcement Learning

On Bits and Bandits: Quantifying the Regret-Information Trade-off

Mojito: LLM-Aided Motion Instructor with Jitter-Reduced Inertial Tokens

BP-SGCN: Behavioral Pseudo-Label Informed Sparse Graph Convolution Network for Pedestrian and Heterogeneous Trajectory Prediction

CyberSentinel: An Emergent Threat Detection System for AI Security

EgoSpeak: Learning When to Speak for Egocentric Conversational Agents in the Wild

Beyond Single-Value Metrics: Evaluating and Enhancing LLM Unlearning with Cognitive Diagnosis

Learning to Play Against Unknown Opponents

An Information-Theoretic Analysis of Thompson Sampling for Logistic Bandits

Model Evolution Framework with Genetic Algorithm for Multi-Task Reinforcement Learning

STaR-SQL: Self-Taught Reasoner for Text-to-SQL

Scalable Decentralized Algorithms for Online Personalized Mean Estimation

Towards more Contextual Agents: An extractor-Generator Optimization Framework

An Empirical Evaluation of Encoder Architectures for Fast Real-Time Long Conversational Understanding

Streamlining the Collaborative Chain of Models into A Single Forward Pass in Generation-Based Tasks

HammerBench: Fine-Grained Function-Calling Evaluation in Real Mobile Device Scenarios

MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents

Performative Prediction on Games and Mechanism Design

RTBAS: Defending LLM Agents Against Prompt Injection and Privacy Leakage

Anchor Sponsor Firms in Open Source Software Ecosystems

A Systematic Evaluation of Generative Models on Tabular Transportation Data

TraderTalk: An LLM Behavioural ABM applied to Simulating Human Bilateral Trading Interactions