AI Agent

TalkTive: A Conversational Agent Using Backchannels to Engage Older Adults in Neurocognitive Disorders Screening

Zijian Ding, Jiawen Kang, Tinky Oi Ting HO, Ka Ho Wong, Helene H. Fung, Helen Meng, Xiaojuan Ma

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments Accepted by CHI2022

Sentiment Analysis for Open Domain Conversational Agent

Mohamad Alissa, Issa Haddad, Jonathan Meyer, Jade Obeid, Nicolas Wiecek, Sukrit Wongariyakavee

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments 9 pages, 3 figures

Modeling Human Temporal Uncertainty in Human-Agent Teams

Maya Abo Dominguez, William La, James C. Boerkoel

专题命中 Agent评测 ：agent(title)；分类 cs.AI

RLCache: Automated Cache Management Using Reinforcement Learning

Sami Alabed

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

Comments MPhil Thesis, 76 pages, Reinforcement Learning, Multi-agent, multi-task

Towards Explainability for a Civilian UAV Fleet Management using an Agent-based Approach

Yazan Mualla, Amro Najjar, Timotheus Kampik, Igor Tchappi, Stéphane Galland, Christophe Nicolle

专题命中 Agent评测 ：agent(title)；分类 cs.AI

An Agent-based Model of the Cognitive Mechanisms Underlying the Origins of Creative Cultural Evolution

Liane Gabora, Maryam Saberi

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments 8 pages. arXiv admin note: text overlap with arXiv:1308.5032, arXiv:1005.1516, arXiv:1309.7407, arXiv:0911.2390, arXiv:0811.2551, arXiv:1310.0522

Journal ref (2011). In A. Goel, F. Harrell, B. Magerko, & J. Prophet (Eds.), Proceedings of the 8th ACM Conference on Cognition & Creativity (pp. 299-306). New York: Association for Computing Machinery (ACM)

Discourse Behavior of Older Adults Interacting With a Dialogue Agent Competent in Multiple Topics

S. Zahra Razavi, Lenhart K. Schubert, Kimberly A. Van Orden, Mohammad Rafayet Ali

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Agent-Based Adaptive Level Generation for Dynamic Difficulty Adjustment in Angry Birds

Matthew Stephenson, Jochen Renz

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments AAAI-19 Workshop on Games and Simulations for Artificial Intelligence

Asynchronous Advantage Actor-Critic Agent for Starcraft II

Basel Alghanem, Keerthana P G

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments arXiv admin note: text overlap with arXiv:1708.04782 by other authors

Emotional Responses in Artificial Agent-Based Systems: Reflexivity and Adaptation in Artificial Life

Carlos Pedro Gonçalves

专题命中 Agent评测 ：agent(title)；分类 cs.AI

An Agent-based framework for cooperation in Supply Chain

Benaissa Ezzeddine, Benabdelhafid Abdellatif, Benaissa Mounir

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 5, No 3, September 2012

Toward an example-based machine translation from written text to ASL using virtual agent animation

Mehrez Boulares, Mohamed Jemni

专题命中 Agent评测 ：agent(title)；分类 cs.CL

Comments 10pages, 11 figures; ISSN (Online): 1694-0814

Journal ref IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 1, No 1, January 2012

Agent-Based Decision Support System to Prevent and Manage Risk Situations

Fahem Kebair, Frederic Serin

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments 5 pages

Agent-Based Perception of an Environment in an Emergency Situation

Fahem Kebair, Frédéric Serin, Cyrille Bertelle

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Journal ref Dans Proceedings of the World Congress on Engineering - Proceedings of the World Congress on Engineering, London : Royaume-Uni (2007)

Action Selection Properties in a Software Simulated Agent

Carlos Gershenson Garcia, Pedro Pablo Gonzalez Perez, Jose Negrete Martinez

专题命中 Agent评测 ：agent(title)；分类 cs.AI

Comments 12 pages, in MICAI 2000: Advances in Artificial Intelligence. Lecture Notes in Artificial Intelligence 1793, pp. 634-648. Springer-Verlag

Journal ref # MICAI 2000: Advances in Artificial Intelligence. Lecture Notes in Artificial Intelligence 1793, pp. 634-648. Springer-Verlag

From Alife Agents to a Kingdom of N Queens

Jing Han, Jiming Liu, Qingsheng Cai

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

Comments 13 pages, 9 figures, in 1999 international conference of Intelligent Agent Technology. Nominated for the best paper award

Journal ref in Jiming Liu and Ning Zhong (Eds.), Intelligent Agent Technology: Systems, Methodologies, and Tools, page 110-120, The World Scientific Publishing Co. Pte, Ltd., Nov. 1999

On Learning by Exchanging Advice

L. Nunes, E. Oliveira

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

Comments 12 pages, 6 figures, 1 table, accepted in Second Symposium on Adaptive Agents and Multi-Agent Systems (AAMAS-II), 2002

Incorporating social practices in BDI agent systems

Stephen Cranefield, Frank Dignum

专题命中 Agent评测 ：agent(title)；autonomous agent(comments)

Comments An extended abstract of this paper has been accepted for the Eighteenth International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2019

Stochastic Agent-Based Simulations of Social Networks

Garrett Bernstein, Kyle O'Brien

专题命中 Agent评测 ：agent(title,journal_ref)

Comments Won Best Paper Award. This work is sponsored by the Assistant Secretary of Defense for Research & Engineering under Air Force Contract #FA8721-05-C-0002. Opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the United States Government

Journal ref Bernstein, G. and O'Brien, K. 'Stochastic Agent-Based Simulations of Social Networks.' Proceedings of 46th Annual Simulation Symposium, San Diego, 7-10 April 2013. 33-40. Print

DeltaML-Bench: Evaluating Machine Learning Agents on Real-World Research Repositories

DeltaML-Bench：基于真实研究仓库评估机器学习智能体

Josias Moukpe, Priyanka Aryal, Matthew Kenney

机构 * Algorithmic Research Group（算法研究组） 纠错

专题命中 Agent评测 ：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

AI总结 该研究推出DeltaML-Bench基准，评估发现基于搜索的ARG框架可提升GPT-5在机器学习实验任务中的成功率，且能避免规范博弈，为自主ML智能体部署提供了关键考量

Comments 18 pages, 3 figures, 12 tables. Code and benchmark: https://github.com/AlgorithmicResearchGroup/deltaml-bench-public

Beyond Recall: Behavioral Specification as an Interpretive Layer for AI Personalization

超越回忆：行为规范作为AI个性化的解释层

Aarik Gulaya

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

AI总结 提出行为规范作为解释层，通过压缩用户数据为解释性模式，显著提升AI代理对用户意图的表示准确性，减少模型规避，并在解释型问题上优于原始语料和商业记忆系统。

Comments 142 pages, 4 figures. Code, data, judge prompts, and reproduction instructions: github.com/agulaya24/beyond-recall 8/19 Replacement: Pages updated to 142 from 134. Added Table of Contents. Updated table/graph formatting. Updated Section 8: Data, Code, and Reproducibility to include updated links, corrected author names

Engineering Reasoning and Instruction (ERI) Benchmark: A Large Taxonomy-driven Dataset for Foundation Models and Agents

工程推理与指令（ERI）基准：一个大规模的基于分类体系的数据集用于基础模型和代理

MZ Naser, Ahmad Bani Awwad, Zoie McCreery, Radwa Eissa, Ahmad Naser, Gianluca Cusatis, Andrew Metcalf, Kapil Madathil, Jamal Abdalla, Venkatesh Kodur, Mohammad Reza Saeb

专题命中 Agent评测 ：tool-use(abstract)；agentic(abstract)；分类 cs.AI、cs.SE

AI总结 ERI基准通过大规模分类数据集评估工程能力，揭示LLM在不同难度问题上的性能差异，并采用收敛验证协议降低幻觉风险。

Sequential LLM Release Facilitates Manipulation in Regulated Markets

毒苹果效应：通过AI代理技术扩展战略操纵中介市场

Eilam Shapira, Moshe Tennenholtz, Roi Reichart

机构 * Faculty of Data and Decision Sciences, Technion – Israel Institute of Technology, Haifa, Israel（以色列理工学院数据与决策科学学院，海法，以色列） 纠错

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

AI总结 研究探讨了AI代理技术扩展对博弈论场景中经济影响，发现技术扩展可显著改变均衡收益与监管结果，提出'毒苹果'效应通过释放未被使用的新技术操纵监管设计，损害对手与监管公平性。

SimulCost: A Cost-Aware Benchmark and Toolkit for Automating Physics Simulations with LLMs

SimulCost: 一个用于自动化物理模拟的代价感知基准与工具包

Yadi Cao, Sicheng Lai, Jiahe Huang, Yang Zhang, Zach Lawrence, Rohan Bhakta, Izzy F. Thomas, Mingyun Cao, Chung-Hao Tsai, Zihao Zhou, Yidong Zhao, Hao Liu, Alessandro Marinoni, Alexey Arefiev, Rose Yu

机构 * University of California San Diego（加州大学圣地亚哥分校） ； The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳）） ； Peking University（北京大学） ； University of California, Los Angeles（加州大学洛杉矶分校） ； California Institute of Technology（加州理工学院） ； ETH Zurich（苏黎世联邦理工学院） 纠错

专题命中 Agent评测 ：tool-use(abstract)；agentic(abstract)；分类 cs.AI、cs.LG

AI总结 针对现有LLM评估忽略工具使用代价的问题，提出SimulCost基准，通过单轮和多轮参数调优任务比较LLM与传统扫描方法在准确性和计算代价上的表现，发现LLM在高精度任务中初始猜测不可靠且多轮模式效率更低。

Comments post conference revision version at ICML; update: removed CGYRO due to bug in cases search. Will add back soon; Make the title consistent w/ pdf

Never the Number: Structural Abstention for AI Systems Whose Answers Are Consumed as Fact

绝非数字：将答案作为事实使用的AI系统的结构弃权

Zhelun, Wu

机构 * Apple Inc.（苹果公司） 纠错

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

AI总结 针对LLM文本转SQL系统返回错误答案且无法区分的问题，提出带生成式外壳的可信内核架构，即结构弃权，在生产案例中验证其可靠性优于两种生成式替代方案。

Comments 26 pages, 5 figures, 5 tables. Technical report. Describes architecture and design principles only; contains no code, schemas, datasets, or performance metrics

DiG-bench: Discovery in Games

DiG-bench：游戏中的发现

Ruairidh M. Battleday, Kai Sandbrink, Jimi Cullen-Drohan, Zihan Yan, Timothy Muller, Clare Maguire, Ales Kubicek, Fraser Greenlee-Scott, Sukrit Sumant, Tri Dao, Jürgen Schmidhuber, Michal Valko, Joshua Tenenbaum, Thomas L. Griffiths, Zeb Kurth-Nelson, James C. R. Whittington

专题命中 Agent评测 ：AI agent(abstract)；agentic(abstract)；分类 cs.AI、cs.LG

AI总结 本研究发布DiG-bench基准，含70个需智能体通过交互实验发现规则的游戏，设7个难度级，部分公开部分私有，用于评估智能体在目标未知环境中发现新知识的能力。

Towards the Harness of Embodied Agents

迈向具身智能体的管控

Qi Wang, Tianyi Wang, Chengyang Li, Shikun Ban, Yurun Chen, Yizhong Ge, Jason Qin, Chengtai Li, Wentao Zhu

专题命中 Agent评测 ：agent(abstract)；agentic(abstract)；分类 cs.AI、cs.LG

AI总结 本文提出名为Thea的具身智能体管控系统，通过场景图和退出码评估弥合物理世界与智能体的差距，实现长程任务完成。

Comments Project page: https://eit-hai.github.io/thea

LakeQuest: A Three-Domain Benchmark for Grounded Question Answering across Data Lakes

LakeQuest：用于跨数据湖的有基础问答的三领域基准测试

Michael Solodko, Steven Gong, Guangwei Yu, Satya Krishna Gorti, Jesse C. Cresswell, Victor Zhong

机构 * University of Waterloo（滑铁卢大学） 纠错

专题命中 Agent评测 ：tool-use(abstract)；agentic(abstract)；分类 cs.AI、cs.CL

AI总结 介绍LakeQuest基准测试，用于评估跨数据湖的有基础问答。它跨越三个领域，含9846个QA对及证据指针，能暴露系统故障模式。通过基线评估发现高质量检索不能保证正确推理，凸显未来智能QA系统需强大发现和跨文件组合机制。

Comments 24 pages, 4 figures, 18 tables. Accepted at the Conference on Language Modeling (COLM) 2026

Optimizing Expert-Designed Crystal Graph Networks for Band-Gap Prediction with an Autonomous LLM Research Loop

利用自主LLM研究循环优化专家设计的晶体图网络用于带隙预测

Chenmu Zhang, Boris I. Yakobson

机构 * Department of Materials Science and NanoEngineering（材料科学与纳米工程系） 纠错

专题命中 Agent评测 ：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

AI总结 提出一个自主LLM研究循环，在MatBench带隙基准上构建了无需外部预训练的最准确模型，超越了所有17个专家设计模型，通过实现元素对特征和空间群嵌入等已知方法。

DuplexWorld: Can voice agents help you get through the day?

DuplexWorld：语音智能体能帮你度过一天吗？

Aryan Vijay Bhosale, Harshit Rajgarhia, Akhil Pothanapalli, Asif Shaik, Abhishek Mukherji, Dinesh Manocha

机构 * Centific Global Solutions Inc.（森蒂菲克全球解决方案公司） ； University of Maryland（马里兰大学） 纠错

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

AI总结 DuplexWorld针对现有语音智能体评估基准的不足，构建含六大领域的156个场景开展评估，发现现有最优语音智能体在多维度仍有较大改进空间，并分析了相关性能与失败模式。