AI Agent

Stationary Heterogeneous-Agent Models in Continuous Time

连续时间下的平稳异质性主体模型

Felix Höfer

专题命中 Agent评测 ：agent(title)

AI总结 研究连续时间下Bewley-Huggett-Aiyagari模型的平稳均衡存在性与多重性，基于财政价格水平理论，发现政府小规模赤字可导致任意偶数个均衡及价格水平多重性。

Agent based simulations visualize Adam Smith's invisible hand by solving Friedrich Hayek's Economic Calculus

基于代理的模拟通过解决弗里德里克·哈耶克的经济计算来可视化亚当·斯密的看不见的手

Klaus Jaffe

专题命中 Agent评测 ：agent(title)

AI总结 本文通过代理模拟展示经济中看不见的手如何在异质环境中通过劳动分工产生协同效应，揭示了经济协同的来源和性质。

Comments Econophysics, Complexity, Synergy

A para-model agent for dynamical systems

动力系统的参数模型智能体

Loïc Michel

专题命中 Agent评测 ：agent(title)

AI总结 提出一种广义的无模型控制方法，通过参数模型智能体实现非线性动力系统的控制与无导数优化，并验证其鲁棒性。

Comments 41 pages, 38 figures, partially presented at the French Symposium of Electrical Engineering in Grenoble, Jun. 2016 and at the Sparse days in St Girons III, Jul. 2015

Agnosiophobia in a virtual agent: behavioral and dynamical architecture in Lenia

虚拟智能体中的未知恐惧：Lenia中的行为与动力学架构

Jesse Cool, Benedikt Hartl, Michael Levin, Samantha Petti

专题命中 Agent评测 ：agent(title)

AI总结 本研究通过在Lenia环境中引入无感知区域，发现虚拟生物倾向于回避这些区域（称为未知恐惧），并通过动力学系统分析揭示其行为源于形态维持这一更根本的目标。

Comments 8 pages, 6 figures; under review at ALIFE 2026

Fine-Tuning Vision-Language Models for Understanding Current Damage and Scoring Priority with Quality Guard Agent

微调视觉语言模型以理解当前损伤并使用质量卫士代理进行优先级评分

Takato Yasuno

专题命中 Agent评测 ：agent(title)

AI总结 本文通过微调LLaVA-1.5-7B视觉语言模型，结合规则引擎和质量卫士代理，实现了桥梁损伤自动理解与修复优先级评分，有效降低了评分变异并提升了效率。

Comments 23 pages, 11 figures, 13 tables

Design and Evaluation of a Culturally Adapted Multimodal Virtual Agent for PTSD Screening

面向PTSD筛查的跨文化多模态虚拟代理设计与评估

Cengiz Ozel, Waleed Nadeem, Samuel Potter, Yahya Bokhari, Bdour Alwuqaysi, Wejdan Alotaibi, Rahaf Fahad Alnufaie, Sabri Boughorbel, Abdulrhman Aljouie, Rakan Altasan, Ehsan Hoque

专题命中 Agent评测 ：agent(title)

AI总结 本文设计并评估了适用于军事医疗场景的跨文化多模态虚拟代理Molhim，通过可配置的对话流程实现特定目的的交互，支持结构化多轮对话和自动会后分析，集成PCL-5量表进行PTSD筛查。

CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance

CountLoop：通过迭代代理指导实现无训练的高实例图像生成

Anindya Mondal, Ayan Banerjee, Sauradip Nag, Josep Llados, Xiatian Zhu, Anjan Dutta

机构 * University of Surrey（萨里大学） ； Universitat Autònoma de Barcelona（巴塞罗那自治大学） ； Simon Fraser University（西蒙菲莎大学） 纠错

专题命中 Agent评测 ：agent(title)

AI总结 CountLoop通过迭代反馈机制实现高密度场景下的精确实例控制，结合视觉语言模型生成布局和评估反馈，减少计数误差并提升空间质量。

Simulating Online Social Media Conversations on Controversial Topics Using AI Agents Calibrated on Real-World Data

利用真实数据校准的AI代理模拟争议性话题的在线社交媒体对话

Elisa Composta, Nicolo' Fontana, Francesco Corso, Francesco Pierri

专题命中 Agent评测 ：AI agent(title)

AI总结 本文研究了基于LLM的代理在模拟微博客社交网络中的行为，探讨了其在不同场景下生成内容、互动及意见演变的机制，发现其生成内容在语气和毒性方面不如真实数据多样，需更精细的认知建模以模拟人类行为。

MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agent

MergeVLA: 朝着通用视觉-语言-动作代理的跨技能模型合并

Yuxia Fu, Zhizhen Zhang, Yuqi Zhang, Zijian Wang, Zi Huang, Yadan Luo

机构 * UQMM Lab, The University of Queensland（昆士兰大学UQMM实验室） 纠错

专题命中 Agent评测 ：agent(title)

AI总结 MergeVLA通过设计保留模型合并性，利用稀疏激活的LoRA适配器和跨注意力机制，实现多技能任务的高效泛化。

Comments Accepted to CVPR 2026

Sparse probabilistic evaluation for treatment planning: a feasibility study in IMPT head & neck patients

稀疏概率评估用于治疗计划：一项在IMPT头颈患者中的可行性研究

Jenneke I. de Jong, Steven J. M. Habraken, Albin Fredriksson, Johan Sundström, Erik Engwall, Sebastiaan Breedveld, Mischa S. Hoogeman

专题命中 Agent评测 ：planning(title)

AI总结 本研究提出稀疏概率评估方法，用于提高IMPT治疗计划中目标覆盖与器官保护的平衡，通过优化网格设置实现高效计算，验证了其在头颈癌患者中的可行性。

Higgs Field as Architect of a Geodesically Complete Universe and Agent for New Physics in Interiors of Black Holes

希格斯场作为几何完备宇宙的架构师及黑洞内部新物理的代理

Itzhak Bars

专题命中 Agent评测 ：agent(title)

AI总结 希格斯场在极端引力区域创造反引力区域，解决黑洞信息悖论并恢复电弱对称性。

Comments 7 pages, 5 figures

PrefIx: Understand and Adapt to User Preference in Human-Agent Interaction

PrefIx: 在人机交互中理解并适应用户偏好

Jialin Li, Zhenhao Chen, Hanjun Luo, Hanan Salam

专题命中 Agent评测 ：agent(title)

AI总结 PrefIx通过交互作为工具范式，评估智能体在人机交互中的偏好适应能力，提升用户体验和偏好对齐度。

An LLM Agent-based Framework for Whaling Countermeasures

基于LLM代理的鲸鱼攻击防御框架

Daisuke Miyamoto, Takuji Iimura, Narushige Michishita

专题命中 Agent评测 ：agent(title)

AI总结 本研究提出基于LLM代理的鲸鱼攻击防御框架，通过构建个性化防御资料和分析电子邮件，提升对高权威目标的防御能力。

Ahead of the Spread: Agent-Driven Virtual Propagation for Early Fake News Detection

提前传播：基于代理的虚拟传播用于早期虚假新闻检测

Bincheng Gu, Min Gao, Junliang Yu, Zongwei Wang, Zhiyi Liu, Kai Shu, Hongyu Zhang

专题命中 Agent评测 ：agent(title)

AI总结 AVOID通过基于代理的虚拟传播方法，主动模拟早期虚假新闻的扩散行为，从而提升早期检测性能。

Post-Training and Test-Time Scaling of Generative Agent Behavior Models for Interactive Autonomous Driving

生成代理行为模型在交互式自动驾驶中的训练和测试时间缩放

Hyunki Seong, Jeong-Kyun Lee, Heesoo Myeong, Yongho Shin, Hyun-Mook Cho, Duck Hoon Kim, Pranav Desai, Monu Surana

专题命中 Agent评测 ：agent(title)

AI总结 本文提出GRBO和Warm-K两种方法，用于提升交互式自动驾驶中生成代理行为模型的训练和测试时间性能，提高安全性和鲁棒性。

Comments 11 pages, 5 figures

Fate of gadolinium in inflamed mouse brain: Release and phosphate interaction post-contrast agent administration

钆在炎症小鼠脑中的命运：对比剂给药后释放及与磷酸盐的相互作用

Lina Anderhalten, Nicole Höfer, Daria Dymnikova, Julia Hahndorf, Matthias Taupitz, Heike Traub, Christian Teutloff, Carmen Infante-Duarte, Robert Bittl

专题命中 Agent评测 ：agent(title)

AI总结 研究通过EPR和ENDOR光谱学揭示炎症条件下小鼠脑中钆的释放及与磷酸盐的相互作用，表明体内和体外分析结合能更准确评估长期钆保留机制。

Comments 50 pages (41 main part, 9 supplementary material), 13 figures (6 main part, 7 supplementary material)

AI/ML Model Cards in Edge AI Cyberinfrastructure: towards Agentic AI

边缘AI计算基础设施中的AI/ML模型卡片：迈向代理AI

Beth Plale, Neelesh Karthikeyan, Isuru Gamage, Joe Stubbs, Sachith Withana

专题命中 Agent评测 ：agentic(title)

AI总结 本文探讨了在边缘AI计算基础设施中使用模型卡片和模型上下文协议，以提升AI/ML模型的动态管理和使用效率。

The Emerged Security and Privacy of LLM Agent: A Survey with Case Studies

Feng He, Tianqing Zhu, Dayong Ye, Bo Liu, Wanlei Zhou, Philip S. Yu

专题命中 Agent评测 ：agent(title)

Comments 35 pages, 19 figures. Accepted to ACM Computing Surveys (CSUR), 2025

Cybersecurity AI Benchmark (CAIBench): A Meta-Benchmark for Evaluating Cybersecurity AI Agents

María Sanz-Gómez, Víctor Mayoral-Vilches, Francesco Balassone, Luis Javier Navarrete-Lozano, Cristóbal R. J. Veas Chavez, Maite del Mundo de Torres

专题命中 Agent评测 ：AI agent(title)

DualTHOR: A Dual-Arm Humanoid Simulation Platform for Contingency-Aware Planning

Boyu Li, Siyuan He, Hang Xu, Haoqi Yuan, Yu Zang, Liwei Hu, Junpeng Yue, Zhenxiong Jiang, Pengbo Hu, Börje F. Karlsson, Yehui Tang, Zongqing Lu

专题命中 Agent评测 ：planning(title)

Comments The experiments in the paper need to be further supplemented, and more methods should be considered for expansion

Pattern Formation in Agent-Based and PDE Models for Evolutionary Games with Payoff-Driven Motion

Tianyong Yao, Chenning Xu, Daniel B. Cooney

专题命中 Agent评测 ：agent(title)

Comments 56 pages, 15 figures, equal contribution from TY and CX

The Impact of Sequential versus Parallel Clearing Mechanisms in Agent-Based Simulations of Artificial Limit Order Book Exchanges

Matej Steinbacher, Mitja Steinbacher, Matjaz Steinbacher

专题命中 Agent评测 ：agent(title)

ChatVis: Large Language Model Agent for Generating Scientific Visualizations

Tom Peterka, Tanwi Mallick, Orcun Yildiz, David Lenz, Cory Quammen, Berk Geveci

专题命中 Agent评测 ：agent(title)

Trusted Routing for Blockchain-Enabled Low-Altitude Intelligent Networks

Sijie He, Ziye Jia, Qiuming Zhu, Fuhui Zhou, Qihui Wu

专题命中 Agent评测 ：agent(abstract,comments)；multi-agent(abstract,comments)

Comments Low-altitude intelligent networks, trusted routing, blockchain, soft hierarchical experience replay buffer, multi-agent deep reinforcement learning

Understanding Online Polarization Through Human-Agent Interaction in a Synthetic LLM-Based Social Network

Tim Donkers, Jürgen Ziegler

专题命中 Agent评测 ：agent(title)

Comments Accepted for publication in the Proceedings of the Nineteenth International AAAI Conference on Web and Social Media (ICWSM 2025). This is the authors' version of the work, with corrections to table cross-references. The definitive Version of Record is available at https://doi.org/10.1609/icwsm.v19i1.35826. arXiv admin note: substantial text overlap with arXiv:2502.01340

Journal ref Proceedings of the Nineteenth International AAAI Conference on Web and Social Media (ICWSM 2025)

Human-Agent Interaction in Synthetic Social Networks: A Framework for Studying Online Polarization

Tim Donkers, Jürgen Ziegler

专题命中 Agent评测 ：agent(title)

What Limits Virtual Agent Application? OmniBench: A Scalable Multi-Dimensional Benchmark for Essential Virtual Agent Capabilities

Wendong Bu, Yang Wu, Qifan Yu, Minghe Gao, Bingchen Miao, Zhenkui Zhang, Kaihang Pan, Yunfei Li, Mengze Li, Wei Ji, Juncheng Li, Siliang Tang, Yueting Zhuang

专题命中 Agent评测 ：agent(title)

Comments Accepted by ICML 2025 (Oral)

A Complete Mental Temporal Logic for Intelligent Agent

Zining Cao

专题命中 Agent评测 ：agent(title)

Comments 18 pages

Opinion dynamics for an increasing population of agents. A symmetric continuous agent model

Ioannis Markou

专题命中 Agent评测 ：agent(title)

A Bayesian design for dual-agent dose optimization with targeted therapies

José L. Jiménez, Mourad Tighiouart

专题命中 Agent评测 ：agent(title)