Memory-Augmented Agents

Follow this topic Ask about this topic

Momentum

41 papers in the last four weeks, up 193% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 235

All topics

MemOPD: On-Policy Distillation through Memory State Alignment for Long-Horizon Agents
Aug 7, 2026Zhiyuan Liu, Tinghong Ye, Chenghao Liu +2On-Policy Distillation Memory-Augmented Agents
School of Advanced Manufacturing and Robotics, Peking University · Zhejiang University · Harbin Institute of Technology
MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents
Aug 7, 2026Zhisheng Chen, Bingfan Zeng, Bangde Cao +8Long-Horizon Agent Tasks Memory-Augmented Agents
Nanyang Technological University · South China University of Technology · Beijing University of Posts and Telecommunications +5
SkillMemo: Expert-guided Skill Memory Framework for Compositional Embodied Manipulation
Aug 6, 2026Changyuan Wang, Chubin Zhang, Zhenyu Wu +8Long-Horizon Robotic Manipulation Robot Skill Learning
Shenzhen International Graduate School, Tsinghua University · Department of Automation, Tsinghua University · Nanyang Technological University +1
Unified Agent: Managing Interactions across Devices
Aug 6, 2026Xinshuang Liu, Runfa Blark Li, Shaoxiu Wei +2AI Agent Benchmarks State Tracking
University of California, San Diego San Diego, CA, USA
Mimir: A Neuro-Symbolic Memory System with Dynamic Grounding for Embodied Agents in Interactive Environments
Aug 5, 2026Haoming Xu, Zhenlin He, Hengyi Wang +2Neuro-Symbolic Reasoning Long-Horizon Agent Tasks
PrimeBot Research Institute · Peking University
InsightEmb: Learning Action-Intent Embeddings for Agentic Insight Retrieval
Aug 5, 2026Tsz Ting Chung, Jiangnan Li, Jie Zhou +1Contrastive Learning Agentic Retrieval
1The Hong Kong University of Science and Technology · 2WeChat AI, Tencent
FocusMem: Factorizing Content, Readout, and Trust in Latent GUI Memory
Aug 5, 2026Zhuoran Zhang, Bowen Li, Jingcheng Ju +5GUI Agents Episodic Memory
School of Computer Science, Peking University · 2Key Lab of High Confidence Software Technologies, Peking University · Institute of Information Engineering, Chinese Academy of Sciences +1
GROVE: Growing and Reasoning over Temporally Stratified Memory from Streaming Video Experience
Aug 3, 2026Sitong Gong, Caixin Kang, Tianyu Yan +7Streaming Video Understanding Video QA
1Dalian University of Technology · 2Alaya Lab · 3The University of Tokyo +1
KC-Agent: A Dual-Process Cognitive Architecture for Efficient ML Model Improvement
Aug 3, 2026Gusseppe Bravo-Rocca, Jordi Guitart, Ajay Dholakia +2Cognitive Modeling Concept Drift
Barcelona Supercomputing Center, Barcelona, Spain · Cotiviti, South Jordan, UT, United States · Universitat Polit`ecnica de Catalunya, Barcelona, Spain +1
Salami Attack: Stealthy Collusive Memory Poisoning against OpenClaw
Aug 3, 2026Zheng Lin, Yuzhe Huang, Zhenxing Niu +2Agent Memory Poisoning Automated Red Teaming
Xidian University Xi’an, Shaanxi, China
Long-Horizon Embodied Decision-Making via Multimodal Memory Compression
Aug 2, 2026Bingxuan Li, Rui Yang, Cheng Qian +6Long-Horizon Agent Evaluation AI Agent Benchmarks
University of Illinois Urbana-Champaign · Northwestern University
PATH-Bench: Path-Dependent Evaluation of Lifelong Agents
Aug 2, 2026Xidong Yang, Xingyi Zhang, Wenhao Li +7Continual Learning for LLM Agents Lifelong Learning Agents
East China Normal University · Tongji University · Ant Group +5
HAM-VLN: Harnessing Hierarchical Agentic Memory for Zero-Shot Vision-and-Language Navigation
Jul 31, 2026An Liu, Bingxi Liu, Hongyu Ding +6Memory-Augmented VLMs Vision-Language Navigation
Institute of Automation, Chinese Academy of Sciences, Beijing, China · Southern University of Science and Technology, Shenzhen, China · Pengcheng Laboratory, Shenzhen, China +3
RRM: Experience-Driven Reflective Retrieval Memory for Long-Horizon Multimodal Reasoning
Jul 30, 2026Jingxiang Fan, Junbao Zhuo, Bochao ZouMultimodal Memory Multimodal Reasoning
University of Science and Technology Beijing 30 Xueyuan Road, Haidian District, Beijing 100083, China
Bridging Inference-Time Scaling and Episodic Memory with Action-Centric Graphs
Jul 29, 2026Xu Zheng, Chaohao Lin, Zhuomin Chen +4Agent Memory Inference-Time Search
Florida International University, Miami, FL, USA · Stanford University, Stanford, CA, USA · NEC Laboratories America, Princeton, NJ, USA +1
Living-Harness Is an Interactive-Agent Evolver
Jul 29, 2026Yuetian Du, Yucheng Wang, He Xu +9AI Agent Reliability Agent Harness Optimization
Zhejiang University · The Hong Kong University of Science and Technology · Alibaba Group +2
Voice Memory for Agentic Speech Recognition
Jul 29, 2026Chao-Han Huck Yang, Zih-Ching Chen, Piotr Zelasko +3Memory-Augmented Language Models Automatic Speech Recognition
NVIDIA
Memory for Attention: Language-Conditioned Re-Perception with a Vision--Language--Motion Map
Jul 26, 2026Dibyendu GhoshActive Perception Visual Attention
Dibyendu Ghosh
Try Once, Then Optimal: De-Redundified Procedure Memory for Cross-Episode Exploration Amortization
Jul 26, 2026Haizhou Ge, Haochen Ouyang, Zhixing Chen +6RL Exploration Robotic Manipulation
1Tsinghua University. · 3Northeast Agricultural University. · 2DISCOVER Robotics.
Co-Evolving Graph and Text Memory for Training-Free Multi-Hop Question Answering
Jul 25, 2026Hieu Man, Thien Huu NguyenMulti-Hop QA Graph Retrieval-Augmented Generation
University of Oregon, OR, USA
Harnessing agent memory to build lifelong AI partners for materials scientists
Jul 25, 2026Siyu Liu, Bo Hu, Beilin Ye +3Agent Memory AI Agents for Scientific Discovery
Center for Structural Materials, Department of Mechanical Engineering, The University of Hong Kong, Hong Kong, China · International Digital Economy Academy (IDEA), Shenzhen, China · Materials Innovation Institute for Life Sciences and Energy (MILES), HKU-SIRI, Shenzhen, China
Ground Truth First: A Longitudinal Evaluation Instrument for Agent Memory, and the Tenure Crossover in Memory-Architecture Rankings
Jul 24, 2026Quentin SpencerLong-Horizon Agent Evaluation Memory Agent Benchmarks
Independent Researcher
AREX: Towards a Recursively Self-Improving Agent for Deep Research
Jul 23, 2026Shuqi Lu, Chaofan Li, Kun Luo +21Agentic RL Deep Research Agents
Beijing Academy of Artificial Intelligence (BAAI)
Causal-AgentIR: Self-Evolving Causal Memory for Adaptive Image Restoration Agents
Jul 23, 2026Hu Gao, Yulong Chen, Lizhuang MaImage Restoration Multi-Agent Reasoning
Department of Computer Science, Shanghai Jiao Tong University, Shanghai 200240, China · Department of Architecture and Design, Harbin Institute of Technology, Heilongjiang, China
Coordinating from Memory: Graph-Structured Experience Reuse for Multi-Agent Adaptation in Dynamic Manufacturing
Jul 22, 2026Chengxiao Dai, Zhanhui Lin, Zhaokun Yan +3Multi-Agent Coordination Combinatorial Optimization
School of Computer Science, University of Sydney, Sydney, Australia. · School of Computation, Information and Technology, Technical University of Munich, Munich, Germany. · School of Life and Environmental Sciences, University of Sydney, Sydney, Australia. +2
Knowledge-Centric Self-Improvement
Jul 21, 2026Xuefei Julie Wang, Lauren Hyoseo Yoon, Chengrui Qu +4Self-Improving Agents Memory-Augmented Agents
Caltech
Neuro-Symbolic Meta-Policies for Temporal Knowledge-Graph Memory under Partial Observability
Jul 20, 2026Taewoon Kim, Vincent François-Lavet, Michael CochezReinforcement Learning Partially Observable RL
HumemAI, The Netherlands · Vrije Universiteit Amsterdam, The Netherlands · ELLIS Institute Finland & Abo Akademi University, Finland
MemoGuard: An Adaptive Runtime for Guarding Against Memory Traps in Communication-Limited Robot Navigation
Jul 17, 2026Rajat Bhattacharjya, Hyeonjong Ju, Sing-Yao Wu +2Robot Navigation Safe Robot Navigation
Department of Computer Science, University of California, Irvine · Department of Computer Science, Yonsei University, South Korea
Bad Memory: Evaluating Prompt Injection Risks from Memory in Agentic Systems
Jul 16, 2026Soham Gadgil, David Alexander, Sai Sunku +1AI Agent Security Agent Memory Poisoning
University of Washington Seattle, WA, USA
MEMORA: Embodied Action Memory from Egocentric Videos for Reasoning and Planning
Jul 15, 2026Zihao Yu, Xiu Yuan, Chongjie ZhangEmbodied AI Memory-Augmented Agents
Washington University in St. Louis

Aug 7, 2026, cs.AIJ/K move · Enter open · S save

MemOPD: On-Policy Distillation through Memory State Alignment for Long-Horizon Agents

Zhiyuan Liu, Tinghong Ye, Chenghao Liu, Yizhuo Li+1

School of Advanced Manufacturing and Robotics, Peking University · Zhejiang University · Harbin Institute of Technology

On-Policy Distillation Memory-Augmented Agents RL for LLM Agents

Open paper AskPDF arXiv

Long-horizon agents accumulate growing contexts during interaction, impairing performance and stability. Compact memory mitigates this problem by compressing and rewriting the history retained between model invocations. Learning what to retain typically relies on proximal policy optimization (PPO) with final task rewards, but sparse rewards provide little guidance for individual memory updates. This limitation motivates on-policy distillation (OPD), which supplies dense teacher supervision on student rollouts. For such supervision to be valid, the teacher must evaluate each sampled action under the same state in which it was generated. However, the context rewriting performed during memory compression can break this alignment. When sampled responses are retained and re-encoded for later invocations, flattening the interaction into a persistent history may cause the teacher to score the action under a state that the student never visited during rollout. The action therefore remains on-policy by provenance, but not necessarily by state. We therefore propose Memory-Aligned On-Policy Distillation (MemOPD). MemOPD records the inputs and sampled outputs of each model invocation, restores its original token positions and causal visibility, and packs the reconstructed invocations for efficient teacher scoring. The teacher provides full-vocabulary supervision at the sampled action positions, while PPO preserves the final task objective. Experiments verify state alignment across several context updates and show that it improves F1 by 7.0% over persistent-history teacher scoring in a matched control. Overall, MemOPD-3B improves F1 over PPO by up to 416.2%, while packing yields up to a 1.63x speedup in actor computation during training. The code for this work is publicly available at: https://github.com/TPssp/MemOPD.

Momentum

Related topics

Latest papers 235