AI Agent - arXivDaily 专题

2510.24259 2025-10-29 cs.CL cs.RO 79%

Can LLMs Translate Human Instructions into a Reinforcement Learning Agent's Internal Emergent Symbolic Representation?

Ziqi Ma, Sao Mai Nguyen, Philippe Xu

机构 * U2IS, ENSTA, IP-Paris（U2IS、ENSTA、IP-巴黎）

专题命中 Agent评测：agent(title,abstract)；分类 cs.CL

2510.22559 2025-10-28 cs.CL 79%

A Closed-Loop Personalized Learning Agent Integrating Neural Cognitive Diagnosis, Bounded-Ability Adaptive Testing, and LLM-Driven Feedback

Zhifeng Wang, Xinyue Zheng, Chunyan Zeng

机构 * Faculty of Artificial Intelligence in Education（人工智能教育学院）； Central China Normal University（中国中部师范大学）； School of Electrical and Electronic Engineering（电气与电子工程学院）； Hubei University of Technology（湖北工业大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.CL

Comments 8 pages, 6 figures

2510.22471 2025-10-28 cs.GT cs.LG 79%

Learning Local Stackelberg Equilibria from Repeated Interactions with a Learning Agent

Nivasini Ananthakrishnan, Yuval Dagan, Kunhe Yang

机构 * UC Berkeley（伯克利大学）； Tel Aviv University（特拉维夫大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.LG

2510.21177 2025-10-27 cs.LG 79%

Scalable Principal-Agent Contract Design via Gradient-Based Optimization

Tomer Galanti, Aarya Bookseller, Korok Ray

专题命中 Agent评测：agent(title,abstract)；分类 cs.LG

2507.13937 2025-10-27 cs.CL 79%

Marcel: A Lightweight and Open-Source Conversational Agent for University Student Support

Jan Trienes, Anastasiia Derzhanskaia, Roland Schwarzkopf, Markus Mühling, Jörg Schlötterer, Christin Seifert

机构 * Marburg University（马尔堡大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.CL

Comments Accepted at EMNLP 2025 (System Demonstrations)

2510.18596 2025-10-22 cs.SE cs.CV 79%

CUARewardBench: A Benchmark for Evaluating Reward Models on Computer-using Agent

Haojia Lin, Xiaoyu Tan, Yulei Qin, Zihan Xu, Yuchen Shi, Zongyi Li, Gang Li, Shaofei Cai, Siqi Cai, Chaoyou Fu, Ke Li, Xing Sun

机构 * Youtu-Agent Team（YouTu-Agent团队）

专题命中 Agent评测：agent(title,abstract)；分类 cs.SE

Comments 24 pages, 6 figures

2510.18483 2025-10-22 cs.AI 79%

StarBench: A Turn-Based RPG Benchmark for Agentic Multimodal Decision-Making and Information Seeking

Haoran Zhang, Chenhao Zhu, Sicong Guo, Hanzhe Guo, Haiming Li, Donglin Yu

机构 * University of Michigan（密歇根大学）； Stanford University（斯坦福大学）； University of Illinois Urbana–Champaign（伊利诺伊大学厄巴纳-香槟分校）

专题命中 Agent评测：agentic(title,abstract)；分类 cs.AI

2510.16085 2025-10-21 cs.CY cs.AI 79%

MoPHES:Leveraging on-device LLMs as Agent for Mobile Psychological Health Evaluation and Support

Xun Wei, Pukai Zhou, Zeyu Wang

机构 * School of Software Engineering, Jiangxi University of Science and Technology（江西科技理工大学软件工程学院）； School of Computer Science, Shenzhen University（深圳大学计算机科学学院）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments This work has been submitted to the IEEE for possible publication

2508.16508 2025-10-17 cs.MA cs.SE 79%

ABMax: A JAX-based Agent-based Modeling Framework

Siddharth Chaturvedi, Ahmed El-Gazzar, Marcel van Gerven

专题命中 Agent评测：agent(title,abstract)；分类 cs.SE

Comments 8 pages, 7 figures, 4 tables, 2 algorithms

2510.13371 2025-10-16 cs.IR cs.AI 79%

MADREC: A Multi-Aspect Driven LLM Agent for Explainable and Adaptive Recommendation

Jiin Park, Misuk Kim

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments 18 pages

2505.16700 2025-10-14 cs.AI 79%

MCP-RADAR: A Multi-Dimensional Benchmark for Evaluating Tool Use Capabilities in Large Language Models

Xuanqi Gao, Siyi Xie, Juan Zhai, Shiqing Ma, Chao Shen

机构 * Xi’an Jiaotong University（西安交通大学）； University of Massachusetts at Amherst（马萨诸塞大学阿默斯特分校）

专题命中 Agent评测：tool use(title)；agent(abstract)；分类 cs.AI

2502.12575 2025-10-14 cs.CR cs.AI 79%

DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent

Pengyu Zhu, Zhenhong Zhou, Yuanhe Zhang, Shilinlu Yan, Kun Wang, Sen Su

机构 * Beijing University of Posts and Telecommunications（北京邮电大学）； Nanyang Technological University（南洋理工大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

2510.06674 2025-10-10 cs.AI 79%

Agent-in-the-Loop: A Data Flywheel for Continuous Improvement in LLM-based Customer Support

Cen Mia Zhao, Tiantian Zhang, Hanchen Su, Yufeng Wayne Zhang, Shaowei Su, Mingzhi Xu, Yu Elaine Liu, Wei Han, Jeremy Werner, Claire Na Cheng, Yashar Mehdad

机构 * Airbnb, Inc.（Airbnb公司）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments EMNLP 2025 Industry Track submission (Paper #305). Preprint. Main text within the 7-page industry limit (references/appendices excluded). Contains multiple figures and tables

2510.05865 2025-10-08 cs.AI cs.CV cs.RO 79%

The Safety Challenge of World Models for Embodied AI Agents: A Review

Lorenzo Baraldi, Zifan Zeng, Chongzhe Zhang, Aradhana Nayak, Hongbo Zhu, Feng Liu, Qunli Zhang, Peng Wang, Shiming Liu, Zheng Hu, Angelo Cangelosi, Lorenzo Baraldi

机构 * University of Pisa（比萨大学）； Huawei RAMS Lab（华为RAMS实验室）； Technical University of Munich（慕尼黑技术大学）； Technical University of Berlin（柏林技术大学）； University of Manchester（曼彻斯特大学）； University of Modena and Reggio Emilia（莫德纳和雷吉奥艾米利亚大学）

专题命中 Agent评测：AI agent(title)；agent(abstract)；分类 cs.AI

2510.04201 2025-10-07 cs.CV cs.AI 79%

World-To-Image: Grounding Text-to-Image Generation with Agent-Driven World Knowledge

Moo Hyun Son, Jintaek Oh, Sun Bin Mun, Jaechul Roh, Sehyun Choi

机构 * The Hong Kong University of Science and Technology（香港科学与技术大学）； Georgia Institute of Technology（佐治亚理工学院）； University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）； TwelveLabs

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

2510.00549 2025-10-03 cs.DB cs.AI 79%

EMR-AGENT: Automating Cohort and Feature Extraction from EMR Databases

Kwanhyung Lee, Sungsoo Hong, Joonhyung Park, Jeonghyeop Lim, Juhwan Choi, Donghwee Yoon, Eunho Yang

机构 * AITRICS ； Korea Advanced Institute of Science and Technology（韩国科学技术院）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments currently under submission to ICLR 2026

2509.23006 2025-09-30 cs.AI 79%

Creative Adversarial Testing (CAT): A Novel Framework for Evaluating Goal-Oriented Agentic AI Systems

Hassen Dhrif

机构 * Amazon（亚马逊）

专题命中 Agent评测：agentic(title,abstract)；分类 cs.AI

2509.21291 2025-09-26 cs.AI cs.CV 79%

VC-Agent: An Interactive Agent for Customized Video Dataset Collection

Yidan Zhang, Mutian Xu, Yiming Hao, Kun Zhou, Jiahao Chang, Xiaoqiang Liu, Pengfei Wan, Hongbo Fu, Xiaoguang Han

机构 * SSE, The Chinese University of Hong Kong, Shenzhen（深圳中文大学香港科技大学 SSE）； The Hong Kong University of Science（香港科学大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments Project page: https://allenyidan.github.io/vcagent_page/

2509.17259 2025-09-23 cs.AI 79%

Mind the Gap: Comparing Model- vs Agentic-Level Red Teaming with Action-Graph Observability on GPT-OSS-20B

Ilham Wicaksono, Zekun Wu, Rahul Patel, Theo King, Adriano Koshiyama, Philip Treleaven

机构 * University College London（伦敦大学学院）； Holistic AI（整体AI）

专题命中 Agent评测：agentic(title,abstract)；分类 cs.AI

Comments Winner of the OpenAI GPT-OSS-20B Red Teaming Challenge (Kaggle, 2025)

2507.11049 2025-09-23 cs.CL 79%

Journalism-Guided Agentic In-Context Learning for News Stance Detection

Dahyun Lee, Jonghyeon Choi, Jiyoung Han, Kunwoo Park

专题命中 Agent评测：agentic(title)；agent(abstract)；分类 cs.CL

Comments EMNLP 2025 (24 pages)

2509.14477 2025-09-19 cs.CL 79%

Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation

Thales Sales Almeida, João Guilherme Alves Santos, Thiago Laitz, Giovana Kerche Bonás

机构 * Institute of Computing (IC) State University of Campinas（计算学院（IC）坎皮纳斯州立大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.CL

2509.08809 2025-09-11 cs.CL 79%

Evaluating LLMs Without Oracle Feedback: Agentic Annotation Evaluation Through Unsupervised Consistency Signals

Cheng Chen, Haiyan Yin, Ivor Tsang

机构 * Australian Artificial Intelligence Institute (AAII), University of Technology Sydney, Australia（澳大利亚人工智能研究所（AAII）、悉尼技术大学）； Centre for Frontier AI Research, Institute of High Performance Computing（前沿人工智能研究中心、高性能计算研究所）； Agency for Science, Technology and Research, Singapore（科技研究局（新加坡））； College of Computing and Data Science, Nanyang Technological University, Singapore（计算与数据科学学院、南洋理工大学）

专题命中 Agent评测：agentic(title,abstract)；分类 cs.CL

Comments 11 pages, 10 figures

Journal ref Published ICLR 2025 Workshop on Scaling Self-Improving Foundation Models without Human Supervision

2509.06341 2025-09-09 cs.AI 79%

Evaluating Multi-Turn Bargain Skills in LLM-Based Seller Agent

Issue Yishu Wang, Kakam Chong, Xiaofeng Wang, Xu Yan, DeXin Kong, Chen Ju, Ming Chen, Shuai Xiao, Shuguang Han, jufeng chen

机构 * Johns Hopkins University（约翰霍普金斯大学）； Tsinghua University（清华大学）； Shanghai Jiao Tong University（上海交通大学）； Alibaba Group（阿里巴巴集团）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

2508.16172 2025-09-08 cs.AI 79%

Graph RAG as Human Choice Model: Building a Data-Driven Mobility Agent with Preference Chain

Kai Hu, Parfait Atchade-Adelomou, Carlo Adornetto, Adrian Mora-Carrero, Luis Alonso-Pastor, Ariel Noyman, Yubo Liu, Kent Larson

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

2509.02924 2025-09-04 cs.MM cs.AI cs.HC 79%

Simulacra Naturae: Generative Ecosystem driven by Agent-Based Simulations and Brain Organoid Collective Intelligence

Nefeli Manoudaki, Mert Toka, Iason Paterakis, Diarmid Flatley

机构 * Media Arts & Technology UC Santa Barbara（媒体艺术与技术大学圣芭芭拉分校）

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments to be published in IEEE VISAP 2025

2508.21246 2025-09-01 quant-ph cs.AI 79%

HCQA: Hybrid Classical-Quantum Agent for Generating Optimal Quantum Sensor Circuits

Ahmad Alomari, Sathish A. P. Kumar

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments 9 pages, 9 figures

2507.23344 2025-09-01 cs.LG cs.MA 79%

Designing Dynamic Pricing for Bike-sharing Systems via Differentiable Agent-based Simulation

Tatsuya Mitomi, Fumiyasu Makinoshima, Fumiya Makihara, Eigo Segawa

机构 * Fujitsu Limited（富士通株式会社）

专题命中 Agent评测：agent(title,abstract)；分类 cs.LG

Comments The typo in the author's name has been corrected

2508.15526 2025-08-22 cs.CL 79%

SafetyFlow: An Agent-Flow System for Automated LLM Safety Benchmarking

Xiangyang Zhu, Yuan Tian, Chunyi Li, Kaiwei Zhang, Wei Sun, Guangtao Zhai

机构 * Shanghai AI Lab（上海人工智能实验室）； East China Normal University（东华大学）

专题命中 Agent评测：agent(title,abstract)；分类 cs.CL

Comments Code and dataset are available at https://github.com/yangyangyang127/SafetyFlow

2508.12896 2025-08-19 cs.AI cs.HC stat.ME 79%

Reliability, Embeddedness, and Agency: A Utility-Driven Mathematical Framework for Agent-Centric AI Adoption

Faruk Alpay, Taylan Alpay

专题命中 Agent评测：agent(title,abstract)；分类 cs.AI

Comments 17 pages, 7 figures, 4 tables

2508.12257 2025-08-19 cs.CL 79%

Structuring the Unstructured: A Systematic Review of Text-to-Structure Generation for Agentic AI with a Universal Evaluation Framework

Zheye Deng, Chunkit Chan, Tianshi Zheng, Wei Fan, Weiqi Wang, Yangqiu Song

机构 * Department of Computer Science and Engineering, HKUST, Hong Kong SAR, China（计算机科学与工程系，香港科技大学，香港特别行政区，中国）

专题命中 Agent评测：agentic(title,abstract)；分类 cs.CL

Comments Under Review

1. Agent评测 15686 篇

Can LLMs Translate Human Instructions into a Reinforcement Learning Agent's Internal Emergent Symbolic Representation?

A Closed-Loop Personalized Learning Agent Integrating Neural Cognitive Diagnosis, Bounded-Ability Adaptive Testing, and LLM-Driven Feedback

Learning Local Stackelberg Equilibria from Repeated Interactions with a Learning Agent

Scalable Principal-Agent Contract Design via Gradient-Based Optimization

Marcel: A Lightweight and Open-Source Conversational Agent for University Student Support

CUARewardBench: A Benchmark for Evaluating Reward Models on Computer-using Agent

StarBench: A Turn-Based RPG Benchmark for Agentic Multimodal Decision-Making and Information Seeking

MoPHES:Leveraging on-device LLMs as Agent for Mobile Psychological Health Evaluation and Support

ABMax: A JAX-based Agent-based Modeling Framework

MADREC: A Multi-Aspect Driven LLM Agent for Explainable and Adaptive Recommendation

MCP-RADAR: A Multi-Dimensional Benchmark for Evaluating Tool Use Capabilities in Large Language Models

DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent

Agent-in-the-Loop: A Data Flywheel for Continuous Improvement in LLM-based Customer Support

The Safety Challenge of World Models for Embodied AI Agents: A Review

World-To-Image: Grounding Text-to-Image Generation with Agent-Driven World Knowledge

EMR-AGENT: Automating Cohort and Feature Extraction from EMR Databases

Creative Adversarial Testing (CAT): A Novel Framework for Evaluating Goal-Oriented Agentic AI Systems

VC-Agent: An Interactive Agent for Customized Video Dataset Collection

Mind the Gap: Comparing Model- vs Agentic-Level Red Teaming with Action-Graph Observability on GPT-OSS-20B

Journalism-Guided Agentic In-Context Learning for News Stance Detection

Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation

Evaluating LLMs Without Oracle Feedback: Agentic Annotation Evaluation Through Unsupervised Consistency Signals

Evaluating Multi-Turn Bargain Skills in LLM-Based Seller Agent

Graph RAG as Human Choice Model: Building a Data-Driven Mobility Agent with Preference Chain

Simulacra Naturae: Generative Ecosystem driven by Agent-Based Simulations and Brain Organoid Collective Intelligence

HCQA: Hybrid Classical-Quantum Agent for Generating Optimal Quantum Sensor Circuits

Designing Dynamic Pricing for Bike-sharing Systems via Differentiable Agent-based Simulation

SafetyFlow: An Agent-Flow System for Automated LLM Safety Benchmarking

Reliability, Embeddedness, and Agency: A Utility-Driven Mathematical Framework for Agent-Centric AI Adoption

Structuring the Unstructured: A Systematic Review of Text-to-Structure Generation for Agentic AI with a Universal Evaluation Framework