AI Agent - arXivDaily 专题

2210.06917 2022-10-14 cs.AI cs.LG 73%

A Direct Approximation of AIXI Using Logical State Abstractions

Samuel Yang-Zhao, Tianyu Wang, Kee Siong Ng

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2210.01907 2022-10-06 cs.LG cs.AI cs.GT 73%

A Self-Play Posterior Sampling Algorithm for Zero-Sum Markov Games

Wei Xiong, Han Zhong, Chengshuai Shi, Cong Shen, Tong Zhang

专题命中记忆与上下文管理：agent(abstract)；multi-agent(abstract)；分类 cs.AI、cs.LG

Comments Accepted to ICML 2022

2208.08713 2022-09-12 cs.LG cs.AI 73%

Learning Generative Models for Active Inference using Tensor Networks

Samuel T. Wauthier, Bram Vanhecke, Tim Verbelen, Bart Dhoedt

专题命中记忆与上下文管理：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

Comments Accepted as a workshop paper at IWAI2022 @ ECML/PKDD2022

2207.04361 2022-07-12 cs.RO cs.AI cs.LG 73%

State Dropout-Based Curriculum Reinforcement Learning for Self-Driving at Unsignalized Intersections

Shivesh Khaitan, John M. Dolan

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 6 pages, 6 figures

2105.06660 2021-05-17 cs.AI cs.LG 73%

Estimating Disentangled Belief about Hidden State and Hidden Task for Meta-RL

Kei Akuzawa, Yusuke Iwasawa, Yutaka Matsuo

专题命中记忆与上下文管理：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

Comments accepted to L4DC2021

2006.03662 2021-04-21 cs.LG cs.AI cs.NE stat.ML 73%

Rapid Task-Solving in Novel Environments

Sam Ritter, Ryan Faulkner, Laurent Sartran, Adam Santoro, Matt Botvinick, David Raposo

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2104.00540 2021-04-02 cs.LG cs.AI cs.SY eess.SY math.OC 73%

Reinforcement Learning Beyond Expectation

Bhaskar Ramasubramanian, Luyao Niu, Andrew Clark, Radha Poovendran

专题命中记忆与上下文管理：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

2101.04422 2021-03-16 cs.CE cs.AI cs.LG 73%

Automated Synthesis of Steady-State Continuous Processes using Reinforcement Learning

Quirin Göttl, Dominik G. Grimm, Jakob Burger

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2102.07764 2021-02-18 cs.RO cs.AI cs.CV cs.LG cs.NE 73%

End-to-End Egospheric Spatial Memory

Daniel Lenton, Stephen James, Ronald Clark, Andrew J. Davison

专题命中记忆与上下文管理：agent(abstract)；autonomous agent(abstract)；分类 cs.AI、cs.LG

Comments Conference paper at ICLR 2021. Implementation: https://github.com/ivy-dl/memory Project page: https://djl11.github.io/ESM/

2101.12446 2021-02-01 cs.AI cs.HC cs.LG 73%

Counterfactual State Explanations for Reinforcement Learning Agents via Generative Deep Learning

Matthew L. Olson, Roli Khanna, Lawrence Neal, Fuxin Li, Weng-Keen Wong

专题命中记忆与上下文管理：agent(abstract)；AI agent(abstract)；分类 cs.AI、cs.LG

Comments Full source code available at https://github.com/mattolson93/counterfactual-state-explanations

Journal ref Artificial Intelligence, 2021, 103455, ISSN 0004-3702

2010.01430 2020-10-06 cs.LG cs.AI q-bio.NC stat.ML 73%

Episodic Memory for Learning Subjective-Timescale Models

Alexey Zakharov, Matthew Crosby, Zafeirios Fountas

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2005.06061 2020-06-24 cs.LG cs.AI stat.ML 73%

TOMA: Topological Map Abstraction for Reinforcement Learning

Zhao-Heng Yin, Wu-Jun Li

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

2002.02286 2020-02-10 cs.LG cs.AI 73%

EgoMap: Projective mapping and structured egocentric memory for Deep RL

Edward Beeching, Christian Wolf, Jilles Dibangoye, Olivier Simonin

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

1802.04325 2018-06-12 cs.LG cs.AI stat.ML 73%

Efficient Model-Based Deep Reinforcement Learning with Variational State Tabulation

Dane Corneil, Wulfram Gerstner, Johanni Brea

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments Accepted at ICML 2018; camera-ready version

1707.00836 2017-07-05 cs.CV cs.AI cs.CL 73%

DeepStory: Video Story QA by Deep Embedded Memory Networks

Kyung-Min Kim, Min-Oh Heo, Seong-Ho Choi, Byoung-Tak Zhang

专题命中记忆与上下文管理：agent(abstract)；AI agent(abstract)；分类 cs.AI、cs.CL

Comments 7 pages, accepted for IJCAI 2017

1502.06132 2015-02-24 cs.AI cs.LG cs.RO math.MG 73%

Universal Memory Architectures for Autonomous Machines

Dan P. Guralnik, Daniel E. Koditschek

专题命中记忆与上下文管理：agent(abstract)；planning(abstract)；分类 cs.AI、cs.LG

Comments Technical report, 31 pages, 1 table, 14 figures, 2 appendices

2607.26072 2026-07-30 cs.IR cs.AI 新提交 72%

IFCMemoryBench: Evaluating Long-Term Memory of LLM-Based Agents in BIM Information Retrieval

IFCMemoryBench：评估基于大语言模型（LLM）的智能体在建筑信息模型（BIM）信息检索中的长期记忆能力

Changyu Du, Alexander Vosseler, Filippo Mazza, André Borrmann

专题命中记忆与上下文管理：agent(abstract)；workflow(abstract)；分类 cs.AI；agentic(comments)

AI总结本文推出IFCMemoryBench基准，评估基于LLM的智能体在BIM信息检索中的长期记忆，发现现有通用记忆系统在该场景下表现差，存在领域迁移差距。

Comments KDD 2026 Workshop on Evaluation and Trustworthiness of Agentic AI

2607.23931 2026-07-28 cs.CY 新提交 71%

State-dependent error correlations shape voting thresholds in committees of AI agents

状态依赖误差相关性塑造人工智能代理委员会中的投票阈值

Haifeng Li, Mo Hai

专题命中记忆与上下文管理：AI agent(title)

AI总结研究人工智能代理委员会投票阈值，结合萨赫 - 斯蒂格利茨筛选与误差依赖性，通过齐次可交换高斯 - copula模型及对大量投票数据的估计，发现考虑依赖关系可提升相关系数、降低损失，改进投票阈值选择。

2606.21842 2026-06-23 cs.CR 新提交 71%

Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG

代理辅助的非前缀KV缓存侧信道攻击

He Sun, Shinan Liu, Siyuan Ma, Junhao Li, Mingjun Xiao, Wenhao Jiang

专题命中记忆与上下文管理：agent(title)

AI总结针对RAG系统中非前缀KV缓存融合的侧信道漏洞，提出SpliceLeak攻击，利用“阶梯波”时序特征提取私有提示长度和语义内容，并设计SpliceDefense防御机制。

Comments 15 pages, 8 figures

2606.17480 2026-06-17 cs.CV cs.RO 新提交 71%

GeneralVLA-2: Geometry-Aware Reconstruction and Governed Memory for Robot Planning

GeneralVLA-2: 几何感知重建与受控记忆用于机器人规划

Haoyu Wang, Guoqing Ma, Zeyu Zhang, Yandong Guo, Boxin Shi, Hao Tang

机构 * School of Computer Science, Peking University（北京大学计算机科学学院）； CASIA（中国科学院自动化研究所）； AI 2 Robotics

专题命中记忆与上下文管理：planning(title)

AI总结针对机器人规划中3D物体重建幻觉和记忆质量不可控的问题，提出GeoFuse-MV3D几何先验引导重建分支和受控长期记忆系统，在GSO-30和Terminal-Bench等基准上显著提升性能。

2606.09915 2026-06-10 cs.AR cs.CR 新提交 71%

ARTA: Adaptive Reinforcement-Learning-Based Throttling Agent for RowHammer Vulnerabilities

ARTA: 基于自适应强化学习的节流代理应对RowHammer漏洞

Marco Ho, Michael S. Hsiao, Jeeho Ryoo

专题命中记忆与上下文管理：agent(title)

AI总结提出ARTA，一种基于Q学习的轻量级节流机制，通过监控DRAM刷新窗口内的内存访问行为并动态调整核心吞吐量，消除位翻转并提升性能。

1604.02943 2026-06-04 math.OC cs.RO cs.SY eess.SY 71%

Taming mismatches in inter-agent distances for the formation-motion control of second-order agents

消除交互距离不匹配以实现二阶代理的形成-运动控制

Hector Garcia de Marina, Bayu Jayawardhana, Ming Cao

专题命中记忆与上下文管理：agent(title)

AI总结本文分析距离不匹配对二阶代理标准梯度刚性形成控制的影响，指出其导致最终形状扭曲和群体稳态运动，并通过结合标准控制律与分布式估计器消除这些行为，同时利用不匹配参数控制群体的平移和旋转运动。

Comments 14 pages, conditionally accepted in Automatic Control, IEEE Transactions on

2605.17921 2026-06-02 cs.CV 71%

An Efficient Streaming Video Understanding Framework with Agentic Control

一种具有代理控制的高效流式视频理解框架

Jinming Liu, Jianguo Huang, Zhaoyang Jia, Jiahao Li, Xiaoyi Zhang, Zongyu Guo, Bin Li, Wenjun Zeng, Yan Lu, Xin Jin

机构 * Shanghai Jiao Tong University（上海交通大学）； Eastern Institute of Technology, Ningbo, China（宁波工程技术学院）； Microsoft Research Asia（微软亚洲研究院）

专题命中记忆与上下文管理：agentic(title)

AI总结提出R3-Streaming框架，通过级联控制（记忆压缩、响应判断、计算路由）和年龄感知遗忘策略及目标平衡强化学习（TB-GRPO），在严格延迟预算下实现流式视频理解，性能达到SOTA并减少95-96%视觉令牌使用。

2605.18729 2026-05-19 cs.RO cs.CV 71%

Robo-Cortex: A Self-Evolving Embodied Agent via Dual-Grain Cognitive Memory and Autonomous Knowledge Induction

Robo-Cortex: 通过双粒认知记忆和自主知识诱导实现自我进化具身智能体

Nga Teng Chan, Yi Zhang, Yechi Liu, Renwen Cui, Fanhu Zeng, Zeyuan Ding, Xiancong Ren, Zhang Zhang, Qifeng Chen, Jian Liu, Yong Dai, Xiaozhu Ju

机构 * The Hong Kong University of Science and Technology（香港科学与技术大学）； X-Humanoid ； Institute of Automation, CAS（中国科学院自动化研究所）； Beijing University of Aeronautics and Astronautics（北京航空航天大学）

专题命中记忆与上下文管理：agent(title)

AI总结本文提出Robo-Cortex框架，通过双粒认知记忆和自主知识诱导机制，使机器人能够自主诱导导航启发式方法并优化认知策略，从而在复杂环境中实现自主导航和探索。

2601.09377 2026-01-15 cs.RO 71%

ReflexDiffusion: Reflection-Enhanced Trajectory Planning for High-lateral-acceleration Scenarios in Autonomous Driving

ReflexDiffusion：面向自动驾驶高侧向加速度场景的反射增强轨迹规划

Xuemei Yao, Xiao Yang, Jianbin Sun, Liuwei Xie, Xuebin Shao, Xiyu Fang, Hang Su, Kewei Yang

专题命中记忆与上下文管理：planning(title)

AI总结 ReflexDiffusion通过反射调整机制提升自动驾驶在高侧向加速度场景下的轨迹规划安全性与可靠性。

Comments Accepted by AAAI 2026

2504.00741 2025-12-29 cond-mat.mtrl-sci 71%

Accelerated Inorganic Materials Design with Generative AI Agents

利用生成式AI代理加速无机材料设计

Izumi Takahara, Teruyasu Mizoguchi, Bang Liu

专题命中记忆与上下文管理：AI agent(title)

AI总结 MatAgent通过结合生成模型和预测模型，利用大语言模型的推理能力，实现高效、可解释的无机材料设计与发现。

2511.11370 2025-11-17 cs.IR 71%

SRLF: An Agent-Driven Set-Wise Reflective Learning Framework for Sequential Recommendation

Jiahao Wang, Bokang Fu, Yu Zhu, Yuli Liu

专题命中记忆与上下文管理：agent(title)

2307.08251 2025-09-17 cond-mat.stat-mech cond-mat.soft 71%

The transition to collective motion in nonreciprocal active matter: coarse graining agent-based models into fluctuating hydrodynamics

David Martin, Daniel Seara, Yael Avni, Michel Fruchart, Vincenzo Vitelli

专题命中记忆与上下文管理：agent(title)

2503.12871 2025-03-20 cs.NI 71%

A Reference Architecture for Autonomous Networks: An Agent-Based Approach

Joseph Sifakis, Dongming Li, Hairong Huang, Yong Zhang, Wenshuan Dang, River Huang, Yijun Yu

专题命中记忆与上下文管理：agent(title)

2302.12167 2025-02-11 econ.GN math.OC q-fin.EC 71%

A Principal-Agent Model for Optimal Incentives in Renewable Investments

René Aïd, Annika Kemper, Nizar Touzi

专题命中记忆与上下文管理：agent(title)

1. 记忆与上下文管理 4766 篇

A Direct Approximation of AIXI Using Logical State Abstractions

A Self-Play Posterior Sampling Algorithm for Zero-Sum Markov Games

Learning Generative Models for Active Inference using Tensor Networks

State Dropout-Based Curriculum Reinforcement Learning for Self-Driving at Unsignalized Intersections

Estimating Disentangled Belief about Hidden State and Hidden Task for Meta-RL

Rapid Task-Solving in Novel Environments

Reinforcement Learning Beyond Expectation

Automated Synthesis of Steady-State Continuous Processes using Reinforcement Learning

End-to-End Egospheric Spatial Memory

Counterfactual State Explanations for Reinforcement Learning Agents via Generative Deep Learning

Episodic Memory for Learning Subjective-Timescale Models

TOMA: Topological Map Abstraction for Reinforcement Learning

EgoMap: Projective mapping and structured egocentric memory for Deep RL

Efficient Model-Based Deep Reinforcement Learning with Variational State Tabulation

DeepStory: Video Story QA by Deep Embedded Memory Networks

Universal Memory Architectures for Autonomous Machines

IFCMemoryBench: Evaluating Long-Term Memory of LLM-Based Agents in BIM Information Retrieval

State-dependent error correlations shape voting thresholds in committees of AI agents

Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG

GeneralVLA-2: Geometry-Aware Reconstruction and Governed Memory for Robot Planning

ARTA: Adaptive Reinforcement-Learning-Based Throttling Agent for RowHammer Vulnerabilities

Taming mismatches in inter-agent distances for the formation-motion control of second-order agents

An Efficient Streaming Video Understanding Framework with Agentic Control

Robo-Cortex: A Self-Evolving Embodied Agent via Dual-Grain Cognitive Memory and Autonomous Knowledge Induction

ReflexDiffusion: Reflection-Enhanced Trajectory Planning for High-lateral-acceleration Scenarios in Autonomous Driving

Accelerated Inorganic Materials Design with Generative AI Agents

SRLF: An Agent-Driven Set-Wise Reflective Learning Framework for Sequential Recommendation

The transition to collective motion in nonreciprocal active matter: coarse graining agent-based models into fluctuating hydrodynamics

A Reference Architecture for Autonomous Networks: An Agent-Based Approach

A Principal-Agent Model for Optimal Incentives in Renewable Investments