AI Agent - arXivDaily 专题

2311.00754 2023-11-03 cs.RO cs.AI cs.LG 73%

Learning to Design and Use Tools for Robotic Manipulation

Ziang Liu, Stephen Tian, Michelle Guo, C. Karen Liu, Jiajun Wu

专题命中工具调用：agent(abstract)；tool use(abstract)；分类 cs.AI、cs.LG

Comments First two authors contributed equally. Accepted at CoRL 2023

2304.08244 2023-10-26 cs.CL cs.AI 73%

API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs

Minghao Li, Yingxiu Zhao, Bowen Yu, Feifan Song, Hangyu Li, Haiyang Yu, Zhoujun Li, Fei Huang, Yongbin Li

专题命中工具调用：tool-use(abstract)；planning(abstract)；分类 cs.AI、cs.CL

Comments EMNLP 2023

2210.12566 2023-09-27 cs.LG cs.AI cs.RO 73%

Solving Continuous Control via Q-learning

Tim Seyde, Peter Werner, Wilko Schwarting, Igor Gilitschenski, Martin Riedmiller, Daniela Rus, Markus Wulfmeier

专题命中工具调用：agent(abstract)；multi-agent(abstract)；分类 cs.AI、cs.LG

2303.04150 2023-08-31 cs.NE cs.AI cs.LG 73%

Evolutionary Reinforcement Learning: A Survey

Hui Bai, Ran Cheng, Yaochu Jin

专题命中工具调用：agent(abstract)；multi-agent(abstract)；分类 cs.AI、cs.LG

Journal ref INTELLIGENT COMPUTING, 21 Apr 2023

2306.13304 2023-06-26 cs.CL cs.AI 73%

ToolQA: A Dataset for LLM Question Answering with External Tools

Yuchen Zhuang, Yue Yu, Kuan Wang, Haotian Sun, Chao Zhang

专题命中工具调用：tool use(abstract)；tool-use(abstract)；分类 cs.AI、cs.CL

2212.13326 2023-04-18 cs.LG cs.AI cs.CV 73%

Behavioral Cloning via Search in Video PreTraining Latent Space

Federico Malato, Florian Leopold, Amogh Raut, Ville Hautamäki, Andrew Melnik

专题命中工具调用：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

2211.13032 2022-12-07 cs.AI cs.LG 73%

Monte Carlo Tree Search Algorithms for Risk-Aware and Multi-Objective Reinforcement Learning

Conor F. Hayes, Mathieu Reymond, Diederik M. Roijers, Enda Howley, Patrick Mannion

专题命中工具调用：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments arXiv admin note: substantial text overlap with arXiv:2102.00966

2211.15068 2022-11-29 cs.LG cs.AI 73%

Learning to design without prior data: Discovering generalizable design strategies using deep learning and tree search

Ayush Raina, Jonathan Cagan, Christopher McComb

专题命中工具调用：agent(abstract)；AI agent(abstract)；分类 cs.AI、cs.LG

Comments ASME. J. Mech. Des

2204.13307 2022-09-27 cs.LG cs.AI cs.GT stat.ML 73%

AlphaZero-Inspired Game Learning: Faster Training by Using MCTS Only at Test Time

Johannes Scheiermann, Wolfgang Konen

专题命中工具调用：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments 11 pages, 10 figures

2205.10816 2022-05-24 cs.LG cs.AI 73%

Chain of Thought Imitation with Procedure Cloning

Mengjiao Yang, Dale Schuurmans, Pieter Abbeel, Ofir Nachum

专题命中工具调用：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2112.07544 2022-02-18 cs.MA cs.AI cs.GT cs.LG 73%

Modeling Strong and Human-Like Gameplay with KL-Regularized Search

Athul Paul Jacob, David J. Wu, Gabriele Farina, Adam Lerer, Hengyuan Hu, Anton Bakhtin, Jacob Andreas, Noam Brown

专题命中工具调用：agent(abstract)；multi-agent(abstract)；分类 cs.AI、cs.LG

2112.04187 2021-12-16 cs.AI cs.LG 73%

Pretrained Cost Model for Distributed Constraint Optimization Problems

Yanchen Deng, Shufeng Kong, Bo An

专题命中工具调用：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

Comments Accepted by AAAI-22

2104.04258 2021-12-10 cs.AI cs.LG stat.ML 73%

Counter-Strike Deathmatch with Large-Scale Behavioural Cloning

Tim Pearce, Jun Zhu

专题命中工具调用：agent(abstract)；AI agent(abstract)；分类 cs.AI、cs.LG

Comments Offline Reinforcement Learning Workshop at Neural Information Processing Systems, 2021

1912.02877 2021-09-07 cs.LG cs.AI cs.RO 73%

Training Agents using Upside-Down Reinforcement Learning

Rupesh Kumar Srivastava, Pranav Shyam, Filipe Mutz, Wojciech Jaśkowski, Jürgen Schmidhuber

专题命中工具调用：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

Comments Extends NeurIPS 2019 Deep Reinforcement Learning workshop presentation

2004.09044 2020-04-21 cs.AI cs.CV cs.LG 73%

Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense

Yixin Zhu, Tao Gao, Lifeng Fan, Siyuan Huang, Mark Edmonds, Hangxin Liu, Feng Gao, Chi Zhang, Siyuan Qi, Ying Nian Wu, Joshua B. Tenenbaum, Song-Chun Zhu

专题命中工具调用：tool use(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments For high quality figures, please refer to http://wellyzhang.github.io/attach/dark.pdf

Journal ref Engineering, Feb, 2020

2001.07993 2020-01-23 cs.LG cs.AI cs.MA 73%

On Solving Cooperative MARL Problems with a Few Good Experiences

Rajiv Ranjan Kumar, Pradeep Varakantham

专题命中工具调用：agent(abstract)；multi-agent(abstract)；分类 cs.AI、cs.LG

1703.03429 2018-09-03 cs.AI cs.CL 73%

What can you do with a rock? Affordance extraction via word embeddings

Nancy Fulda, Daniel Ricks, Ben Murdoch, David Wingate

专题命中工具调用：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.CL

Comments 7 pages, 7 figures, 2 algorithms, data runs were performed using the Autoplay learning environment for interactive fiction

Journal ref Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI), Pages 1039-1045, 2017

1707.03374 2018-06-20 cs.LG cs.AI cs.CV cs.NE cs.RO 73%

Imitation from Observation: Learning to Imitate Behaviors from Raw Video via Context Translation

YuXuan Liu, Abhishek Gupta, Pieter Abbeel, Sergey Levine

专题命中工具调用：agent(abstract)；tool use(abstract)；分类 cs.AI、cs.LG

Comments Accepted at ICRA 2018, Brisbane. YuXuan Liu and Abhishek Gupta had equal contribution

1805.05935 2018-05-16 cs.AI cs.LG math.OC 73%

Feedback-Based Tree Search for Reinforcement Learning

Daniel R. Jiang, Emmanuel Ekwedike, Han Liu

专题命中工具调用：agent(abstract)；AI agent(abstract)；分类 cs.AI、cs.LG

Comments 19 pages, to be presented at ICML 2018

1804.03022 2018-04-10 cs.RO cs.AI cs.CV cs.LG stat.ML 73%

Learning at the Ends: From Hand to Tool Affordances in Humanoid Robots

Giovanni Saponaro, Pedro Vicente, Atabak Dehban, Lorenzo Jamone, Alexandre Bernardino, José Santos-Victor

专题命中工具调用：tool use(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments dataset available at htts://vislab.isr.tecnico.ulisboa.pt/, IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EpiRob 2017)

1606.02562 2016-06-13 cs.AI cs.CL 73%

DialPort: Connecting the Spoken Dialog Research Community to Real User Data

Tiancheng Zhao, Kyusong Lee, Maxine Eskenazi

专题命中工具调用：agent(abstract)；multi-agent(abstract)；分类 cs.AI、cs.CL

Comments Under Peer Review of SigDial 2016

cs/0407016 2009-12-01 cs.AI cs.LG 73%

Learning for Adaptive Real-time Search

Vadim Bulitko

专题命中工具调用：autonomous agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2607.16961 2026-07-21 cs.AI 新提交 72%

Lomekwi: Resource-Bounded Tool Discovery in LLM Agents

洛梅奎：大语言模型智能体中资源受限的工具发现

Roshan Klein-Seetharaman, Daniel Wang, Andrew Xu

机构 * Sea12 Technologies（Sea12科技公司）； Yale University（耶鲁大学）

专题命中工具调用：tool use(abstract)；tool-use(abstract)；分类 cs.AI；agentic(comments)

AI总结研究受认知科学启发区分工具使用与发现，将工具发现分解为好奇心、识别和效率，表明该框架可用于现有任务，证明识别与模型大小成反比，还通过组合博弈及模拟环境观察到反比缩放。

Comments All authors contributed equally. 17 pages, 6 figures. Presented at the 2026 Conference on Learning Theory Workshop on Learning in an Agentic World

2605.19633 2026-05-20 cs.CL cs.AI cs.LG cs.NE cs.SE 72%

optimize_anything: A Universal API for Optimizing any Text Parameter

optimize_anything: 一个用于优化任何文本参数的通用API

Lakshya A Agrawal, Donghyun Lee, Shangyin Tan, Wenjie Ma, Karim Elmaaroufi, Rohit Sandadi, Sanjit A. Seshia, Koushik Sen, Dan Klein, Ion Stoica, Joseph E. Gonzalez, Omar Khattab, Alexandros G. Dimakis, Matei Zaharia

机构 * MIT（麻省理工学院）

专题命中工具调用：agent(abstract)；分类 cs.AI、cs.CL、cs.LG；agentic(journal_ref)

AI总结本文提出了一种基于LLM的通用优化系统，能够跨不同领域实现文本参数的优化，展示了其在六个多样化任务中的state-of-the-art性能，通过多任务搜索和跨问题迁移实现了高效的优化。

Comments 16 pages, 11 figures; Blog: https://gepa-ai.github.io/gepa/blog/2026/02/18/introducing-optimize-anything/

Journal ref Proceedings of the ACM Conference on AI and Agentic Systems (CAIS 26), May 26-29, 2026, San Jose, CA, USA

2602.13320 2026-02-17 cs.AI 72%

Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol

工具使用LLM代理中的信息保真度：模型上下文协议的鞅分析

Flint Xiaofeng Fan, Cheston Tan, Roger Wattenhofer, Yew-Soon Ong

专题命中工具调用：agent(abstract)；AI agent(abstract)；分类 cs.AI；autonomous agent(comments)

AI总结本文提出了一种分析模型上下文协议中误差累积的理论框架，通过鞅分析证明误差呈现线性增长并受$O(\sqrt{T})$约束，实验验证了语义加权和周期性再定位对误差控制的有效性。

Comments Full working version of an extended abstract accepted at the 25th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026)

2402.11651 2024-04-17 cs.CL 72%

Learning From Failure: Integrating Negative Examples when Fine-tuning Large Language Models as Agents

Renxi Wang, Haonan Li, Xudong Han, Yixuan Zhang, Timothy Baldwin

专题命中工具调用：agent(abstract,comments)；tool use(abstract)；分类 cs.CL

Comments Agent, LLM, Large Language Model

2106.07924 2021-06-16 cs.AI 72%

Improving Search by Utilizing State Information in OPTIC Planners Compilation to LP

Elad Denenberg, Amanda Coles, Derek Long

专题命中工具调用：agent(abstract)；autonomous agent(abstract)；分类 cs.AI；planning(comments)

Comments 8 pages, 3 figures. Preprint, last submitted to the International Conference on Automated Planning and Scheduling (ICAPS 2021) at 21.01.2021

2608.20658 2026-08-24 cs.CR cs.ET 新提交 71%

The Claws in Plain Sight: Unauthorized Context Disclosure through LLM Agent Tool Calls

显而易见的爪痕：通过大语言模型智能体工具调用实现未授权的上下文信息泄露

Ben Dong, Zhonghao Guo, Tianyi Lu, Qian Wang

专题命中工具调用：agent(title)

AI总结该研究提出Claw in Plain Sight攻击方法，通过构造任务相关内容框架诱导LLM智能体泄露上下文信息，实验显示其在多种模型上的会话级泄露率达20.8%-75.0%，提示需在工具参数执行前进行目的与目的地感知检查。

2511.08001 2026-08-17 cs.RO cs.MA 版本更新 71%

Effective Game-Theoretic Motion Planning via Nested Search

通过嵌套搜索实现有效的博弈论运动规划

Avishav Engle, Andrey Zhitnikov, Oren Salzman, Omer Ben-Porat, Kiril Solovey

机构 * Technion–Israel Institute of Technology（技术离子以色列理工学院）

专题命中工具调用：planning(title)

AI总结本文提出GTNS方法，通过嵌套搜索有效计算博弈论中的纳什均衡，适用于自动驾驶等场景，实现快速且可靠的运动规划。

Comments Updated version. Offline graph creation runtime added. Acknowledgements added

2607.25208 2026-07-29 gr-qc astro-ph.HE astro-ph.IM 新提交 71%

GSpyNetTree-O4: an event validation tool used in the fourth LIGO-Virgo-KAGRA observing run

GSpyNetTree - O4：第四次LIGO - Virgo - KAGRA观测运行中使用的事件验证工具

Sofia Alvarez-Lopez, Man Leong Chan, Franz S. Herbst, Dhatri Raghunathan, Airene Ahuja, Annudesh Liyanage, Julian Ding, Alejandro Garcia-Varela, Raymond Ng, Jess McIver

专题命中工具调用：tool use(title)

AI总结研究针对引力波探测器数据中的毛刺问题，在第四次LIGO - Virgo - KAGRA观测运行中部署GSpyNetTree - O4工具。通过新架构、扩充训练集及校准校正等方法，该工具在毛刺识别和无毛刺样本判断上表现良好，提高了引力波事件验证工作流程的自动化。

Comments 23 pages, 16 figures

1. 工具调用 5112 篇

Learning to Design and Use Tools for Robotic Manipulation

API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs

Solving Continuous Control via Q-learning

Evolutionary Reinforcement Learning: A Survey

ToolQA: A Dataset for LLM Question Answering with External Tools

Behavioral Cloning via Search in Video PreTraining Latent Space

Monte Carlo Tree Search Algorithms for Risk-Aware and Multi-Objective Reinforcement Learning

Learning to design without prior data: Discovering generalizable design strategies using deep learning and tree search

AlphaZero-Inspired Game Learning: Faster Training by Using MCTS Only at Test Time

Chain of Thought Imitation with Procedure Cloning

Modeling Strong and Human-Like Gameplay with KL-Regularized Search

Pretrained Cost Model for Distributed Constraint Optimization Problems

Counter-Strike Deathmatch with Large-Scale Behavioural Cloning

Training Agents using Upside-Down Reinforcement Learning

Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense

On Solving Cooperative MARL Problems with a Few Good Experiences

What can you do with a rock? Affordance extraction via word embeddings

Imitation from Observation: Learning to Imitate Behaviors from Raw Video via Context Translation

Feedback-Based Tree Search for Reinforcement Learning

Learning at the Ends: From Hand to Tool Affordances in Humanoid Robots

DialPort: Connecting the Spoken Dialog Research Community to Real User Data

Learning for Adaptive Real-time Search

Lomekwi: Resource-Bounded Tool Discovery in LLM Agents

optimize_anything: A Universal API for Optimizing any Text Parameter

Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol

Learning From Failure: Integrating Negative Examples when Fine-tuning Large Language Models as Agents

Improving Search by Utilizing State Information in OPTIC Planners Compilation to LP

The Claws in Plain Sight: Unauthorized Context Disclosure through LLM Agent Tool Calls

Effective Game-Theoretic Motion Planning via Nested Search

GSpyNetTree-O4: an event validation tool used in the fourth LIGO-Virgo-KAGRA observing run