Pith. sign in

REVIEW 3 major objections 1 minor 154 references

Learning How and What to Memorize: Cognition-Inspired Two-Stage Optimization for Evolving Memory

T0 review · 3 major / 1 minor · reviewed 2026-05-09 · grok-4.3

Challenge this review Download PDF Read on arXiv ↗

Pith's one-line read Cognition-inspired two-stage optimization learns memory guidelines then policies for evolving LLM personalization.

desk verdict The paper's two-stage MemCoE framework separates guideline induction from RL policy learning for LLM memory evolution, but the abstract supplies no numbers or ablations to show the split actually fixes sparse rewards. read the letter →

arxiv 2605.00702 v1 submitted 2026-05-01 cs.CL

Derong Xu , Shuochen Liu , Pengfei Luo , Pengyue Jia , Yingyi Zhang , Yi Wen , Yimin Deng , Wenlin Zhang
show 3 more authors
Enhong Chen Xiangyu Zhao Tong Xu
This is my paper · ORCID
classification cs.CL
keywords memoryevolutionLLMagentsreinforcementlearningpersonalizationtextualgradientsprocessrewardscognition-inspired
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper targets the challenge of maintaining consistent, evolving user preferences in LLM agents across long interactions where context windows are limited. Existing static rules lack adaptability while direct RL suffers from weak sparse rewards that destabilize learning. MemCoE addresses this by first inducing a global memory guideline from contrastive feedback treated as textual gradients, then applying that guideline to generate structured process rewards for multi-turn RL training of a memory-update policy. A sympathetic reader would care because reliable long-term memory could make personalization more stable and less dependent on hand-engineering or short-horizon signals.

What carries the argument

MemCoE's two-stage framework: Memory Guideline Induction, which derives a global memory organization rule from contrastive textual feedback, followed by Guideline-Aligned Memory Policy Optimization, which converts the guideline into process rewards to guide RL-based learning of memory update actions.

What would settle it

An ablation experiment in which the guideline induction stage is removed or replaced with random guidelines, resulting in no performance gain or greater instability during RL training on the same benchmarks, would falsify the claim that the two-stage separation is necessary for the observed improvements.

Watch

Extended reading notes

Core claim

We introduce MemCoE, a cognition-inspired two-stage optimization framework that learns how memory should be organized and what information to update. In the first stage, Memory Guideline Induction optimizes a global guideline via contrastive feedback interpreted as textual gradients; in the second stage, Guideline-Aligned Memory Policy Optimization uses the induced guideline to define structured process rewards and performs multi-turn RL to learn a guideline-following memory evolution policy. We evaluate on three personalization memory benchmarks, covering explicit/implicit preference and different sizes and noise, and observe consistent improvements over strong baselines with favorable ro

Load-bearing premise

That contrastive feedback interpreted as textual gradients can reliably induce an optimal global memory guideline, and that this guideline then supplies sufficiently informative process rewards to stabilize multi-turn RL for the memory evolution policy.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 1 minor

Summary. The manuscript proposes MemCoE, a cognition-inspired two-stage optimization framework for evolving memory in LLM agents. Stage 1 induces a global memory guideline from contrastive feedback interpreted as textual gradients. Stage 2 uses the guideline to supply structured process rewards for multi-turn RL, training a guideline-following memory evolution policy. The approach is evaluated on three personalization memory benchmarks covering explicit/implicit preferences and varying sizes/noise levels, with claims of consistent improvements over strong baselines along with favorable robustness, transferability, and efficiency.

Significance. If the two-stage separation reliably converts contrastive feedback into stable, dense process rewards that resolve the sparse-outcome instability identified in prior RL memory work, the framework could advance long-horizon memory management for personalized LLM agents by providing both interpretability (via the explicit guideline) and empirical gains. The cognitive analogy and explicit decoupling of guideline induction from policy optimization are conceptually attractive strengths.

major comments (3)
  1. [Abstract] Abstract: the central claims of 'consistent improvements over strong baselines with favorable robustness, transferability, and efficiency' are stated without any quantitative metrics, error bars, baseline descriptions, or ablation results, preventing assessment of effect sizes or whether the two-stage procedure actually outperforms direct RL or static rules.
  2. [Method] Description of the two-stage framework: the guideline produced by contrastive feedback in stage 1 is used directly to define the process rewards in stage 2, creating a circular dependency in which the RL optimization target is shaped by the same learned component without an independent external benchmark or validation that the guideline is optimal or general.
  3. [Experiments] Evaluation section: no ablation, convergence argument, or analysis is supplied demonstrating that the induced guideline measurably reduces reward sparsity or variance in the multi-turn RL stage, despite the abstract explicitly identifying sparse outcome rewards as the source of instability in prior work.
minor comments (1)
  1. The term 'textual gradients' is used without a formal definition or worked example showing how contrastive pairs are converted into an update rule for the guideline.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments and positive recognition of the conceptual contributions. We address each major comment point by point below, clarifying our approach where needed and committing to revisions that strengthen the presentation without altering the core claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claims of 'consistent improvements over strong baselines with favorable robustness, transferability, and efficiency' are stated without any quantitative metrics, error bars, baseline descriptions, or ablation results, preventing assessment of effect sizes or whether the two-stage procedure actually outperforms direct RL or static rules.

    Authors: We agree that the abstract would benefit from greater specificity to convey effect sizes. In the revised version we will insert concise quantitative highlights drawn from the experimental results (e.g., average accuracy gains across the three benchmarks and reference to statistical robustness), while remaining within the abstract length constraint. Baseline families and the two-stage versus direct-RL comparison will be mentioned at a high level. revision: yes

  2. Referee: [Method] Description of the two-stage framework: the guideline produced by contrastive feedback in stage 1 is used directly to define the process rewards in stage 2, creating a circular dependency in which the RL optimization target is shaped by the same learned component without an independent external benchmark or validation that the guideline is optimal or general.

    Authors: The two stages are strictly sequential and non-circular. Stage 1 performs an independent optimization of the guideline using only contrastive textual feedback; once induced, the guideline is frozen and supplied as a fixed reward-shaping function to Stage 2. The RL policy is then optimized against this fixed external signal. We will add explicit wording and a diagram annotation in the method section to emphasize the one-way information flow and note that the guideline itself can be inspected or validated on held-out data independently of the RL stage. revision: partial

  3. Referee: [Experiments] Evaluation section: no ablation, convergence argument, or analysis is supplied demonstrating that the induced guideline measurably reduces reward sparsity or variance in the multi-turn RL stage, despite the abstract explicitly identifying sparse outcome rewards as the source of instability in prior work.

    Authors: We acknowledge that a direct quantitative demonstration of reduced reward sparsity or variance would strengthen the link to the motivating problem. Although overall performance gains are reported, we did not include a dedicated ablation isolating this mechanism. In the revision we will add a short analysis (main text or appendix) comparing reward variance and convergence behavior between the guideline-aligned RL and a direct-outcome-reward baseline, using the same multi-turn setup. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: two-stage framework validated on external benchmarks

full rationale

The paper proposes MemCoE as a two-stage process where stage 1 induces a global guideline from contrastive feedback and stage 2 uses it to shape process rewards for RL-based policy learning. No equations, definitions, or self-citations in the provided text reduce the final performance claims to the inputs by construction. The central results rest on empirical improvements over baselines across three independent personalization memory benchmarks with varying preference types, sizes, and noise levels, providing external falsifiability outside the fitted guideline itself.

Assumptions & free parameters 2 free parameters · 1 assumptions · 1 invented entities

The framework rests on a domain assumption from cognitive science and introduces new learned components whose values are fitted during the two optimization stages.

free parameters (2)
  • guideline induction parameters
    Parameters controlling contrastive feedback and textual gradient updates in stage one.
  • RL process reward scaling
    Scaling factors that turn the induced guideline into structured rewards for the policy optimization.
assumptions (1)
  • domain assumption Memory schema theory and functional division between prefrontal regions and hippocampus regions
    Invoked to justify the separation into guideline induction and policy optimization stages.
invented entities (1)
  • MemCoE two-stage framework
    purpose: To jointly learn memory organization and update policy
    New composite system whose components are defined and optimized within the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning How and What to Memorize: Cognition-Inspired Two-Stage Optimization for Evolving Memory." pith.science (2026). https://pith.science/paper/2605.00702

@misc{pith2026260500702,
  author       = {Pith},
  title        = {Pith review of: Learning How and What to Memorize: Cognition-Inspired Two-Stage Optimization for Evolving Memory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2605.00702}},
  note         = {Machine review of arXiv:2605.00702}
}
read the original abstract

Large language model (LLM) agents require long-term user memory for consistent personalization, but limited context windows hinder tracking evolving preferences over long interactions. Existing memory systems mainly rely on static, hand-crafted update rules; although reinforcement learning (RL)-based agents learn memory updates, sparse outcome rewards provide weak supervision, resulting in unstable long-horizon optimization. Drawing on memory schema theory and the functional division between prefrontal regions and hippocampus regions, we introduce MemCoE, a cognition-inspired two-stage optimization framework that learns how memory should be organized and what information to update. In the first stage, we propose Memory Guideline Induction to optimize a global guideline via contrastive feedback interpreted as textual gradients; in the second stage, Guideline-Aligned Memory Policy Optimization uses the induced guideline to define structured process rewards and performs multi-turn RL to learn a guideline-following memory evolution policy. We evaluate on three personalization memory benchmarks, covering explicit/implicit preference and different sizes and noise, and observe consistent improvements over strong baselines with favorable robustness, transferability, and efficiency.

Figures

Figures reproduced from arXiv: 2605.00702 by the authors.

Figure 1
Figure 1. Top: With a limited context window, the agent fails to capture preferences. Bottom: Inspired by Prefrontal → Hippocampus, we decouple agent memory into Memory Guideline (for organizing) → Agent Memory (for updating). user’s evolving preferences and habits (Jiang et al., 2025a,b). However, the context window prevents LLMs from retaining and exploiting the full history of dialogue (Yen et al., 2024), and simply storin… view at source ↗
Figure 2
Figure 2. Overview of our proposed MemCoE. It performs two-stage optimization for evolving user memory: (1) Memory Guideline Induction (MGI) iteratively refines a natural-language guideline; (2) Guideline-Aligned Memory Policy Optimization (GMPO) fixes the induced guideline to define guideline-aligned rewards and applies multi-turn GRPO to learn what information to update in evolving memory bank. 4.2 Guideline-Aligned Memory … view at source ↗
Figure 3
Figure 3. Efficiency analysis on PersonaMem. We re view at source ↗
Figures from the paper (17 more)
Figure 4
Figure 4. Figure 4: Retrieval Top-K on PersonaMem (32K). sults show that, across all four LLMs, both MGI vari￾ants consistently outperform the baselines (RAG and A-Mem), indicating that the learned guideline captures model-agnostic memory-update principles rather than overfitting to a spe…
Figure 6
Figure 6. Figure 6: Impact of prompt quality on PersonaMem, averaged over three random runs with different seeds; error bars indicate standard deviation. 6 Conclusion Inspired by memory schema theory that highlights prefrontal regions and hippocampus regions, we present MemCoE, a two-stag…
Figure 7
Figure 7. Figure 7: Preference retention during multi-round memory evolution on PrefEval (Explicit). We insert a user
Figure 8
Figure 8. Figure 8: Error analysis decomposing failures into the
Figure 9
Figure 9. Figure 9: Scaling analysis on PersonaMem (128K). We increase the total dialogue tokens (each evolution round
Figure 10
Figure 10. Figure 10: Hyperparameter analysis of optimization steps. The relative gains (
Figure 11
Figure 11. Figure 11: Test set Accuracy of RL-based baselines across RL training steps. imental setting: 300 sampled PersonaMem train￾ing examples, batch size 4, and evaluation on PersonaMem-32K, varying only the number of RL update steps up to 200. As shown in
Figure 12
Figure 12. Figure 12: PersonaMem case study (MCQ). Our method matches the ground-truth option (d), while all baselines select different options. CASE STUDY (PersonaMem-2): Suggesting New Ideas for Music Expression (MCQ) User question: How can I find a more fulfilling way to express my love…
Figure 13
Figure 13. Figure 13: PersonaMem case study (MCQ). Our method selects the correct option (a), whereas each baseline chooses a different option
Figure 14
Figure 14. Figure 14: PersonaBench case study (open-form QA). Our method outputs the correct age (39); baselines answer with missing/unknown information. CASE STUDY (PersonaBench-2): Work Location (Open-form QA) User question: What is the address of my work location? Question type: Basic i…
Figure 15
Figure 15. Figure 15: PersonaBench case study (open-form QA). Our method recovers the correct work location (University), while baselines report missing or unknown information. mary, resolving inconsistencies and retaining the most consistent actionable points
Figure 16
Figure 16. Figure 16: Meta prompt for generating contrastive feedback by comparing a correct and an incorrect memory-update
Figure 17
Figure 17. Figure 17: Meta prompt for synthesizing multiple analysis outputs into a single consolidated feedback summary
Figure 18
Figure 18. Figure 18: Meta prompt for updating template_evolve using aggregated feedback while strictly preserving required placeholders and forbidding new ones (TEMPLATE_OPTIM). TEMPLATE_EVOLVE_STEP50: Evidence-Bound Memory Update from a New Chunk TEMPLATE_EVOLVE_STEP50 = """ You are give…
Figure 19
Figure 19. Figure 19: Meta prompt for updating the user memory profile from a newly observed dialogue chunk with
Figure 20
Figure 20. Figure 20: Shared prompt for final answer generation on multiple-choice benchmarks (PersonaMem, and PrefEval).
Figure 21
Figure 21. Figure 21: Shared prompt for final answer generation on PersonaBench, constraining outputs to only the relevant

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

154 extracted references · 154 canonical work pages

  1. [2]

    M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation

    Bge m3-embedding: Multi-lingual, multi-functionality, multi-granularity text embeddings through self-knowledge distillation , author=. arXiv preprint arXiv:2402.03216 , year=

    work page Pith review arXiv

  2. [3]

    Pandya, M

    Automating Customer Service using LangChain: Building custom open-source GPT Chatbot for organizations , author=. arXiv preprint arXiv:2310.05421 , year=

    work page arXiv

  3. [4]

    The Thirteenth International Conference on Learning Representations , year=

    SeCom: On Memory Construction and Retrieval for Personalized Conversational Agents , author=. The Thirteenth International Conference on Learning Representations , year=

    work page

  4. [5]

    Neurocomputing , pages=

    Recursively summarizing enables long-term dialogue memory in large language models , author=. Neurocomputing , pages=. 2025 , publisher=

    work page 2025

  5. [6]

    Proceedings of the 10th SIGHAN Workshop on Chinese Language Processing (SIGHAN-10) , pages=

    PerLTQA: A Personal Long-Term Memory Dataset for Memory Classification, Retrieval, and Fusion in Question Answering , author=. Proceedings of the 10th SIGHAN Workshop on Chinese Language Processing (SIGHAN-10) , pages=

    work page

  6. [7]

    Madial- bench: Towards real-world evaluation of memory-augmented dialogue generation

    MADial-Bench: Towards Real-world Evaluation of Memory-Augmented Dialogue Generation , author=. arXiv preprint arXiv:2409.15240 , year=

    work page arXiv

  7. [8]

    MemSim: A Bayesian simulator for evaluating memory of LLM-based personal assistants

    Memsim: A bayesian simulator for evaluating memory of llm-based personal assistants , author=. arXiv preprint arXiv:2409.20163 , year=

    work page arXiv

  8. [9]

    Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , year=

    Beyond Goldfish Memory: Long-Term Open-Domain Conversation , author=. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , year=

    work page

Show all 154 references
  1. [10]

    Proceedings of the 31st International Conference on Computational Linguistics: Industry Track , pages=

    CarMem: Enhancing Long-Term Memory in LLM Voice Assistants through Category-Bounding , author=. Proceedings of the 31st International Conference on Computational Linguistics: Industry Track , pages=

  2. [11]

    Foundations and Trends

    The probabilistic relevance framework: BM25 and beyond , author=. Foundations and Trends. 2009 , publisher=

    2009

  3. [13]

    LangChain Team , title =

  4. [15]

    arXiv preprint arXiv:2210.08750 , year=

    Keep me updated! memory management in long-term conversations , author=. arXiv preprint arXiv:2210.08750 , year=

  5. [16]

    arXiv preprint arXiv:2308.03281 , year=

    Towards general text embeddings with multi-stage contrastive learning , author=. arXiv preprint arXiv:2308.03281 , year=

  6. [17]

    arXiv preprint arXiv:2402.11573 , year=

    Bge landmark embedding: A chunking-free embedding method for retrieval augmented long-context large language models , author=. arXiv preprint arXiv:2402.11573 , year=

  7. [18]

    arXiv preprint arXiv:2212.03533 , year=

    Text embeddings by weakly-supervised contrastive pre-training , author=. arXiv preprint arXiv:2212.03533 , year=

  8. [19]

    arXiv preprint arXiv:2112.09118 , year=

    Unsupervised dense information retrieval with contrastive learning , author=. arXiv preprint arXiv:2112.09118 , year=

  9. [20]

    , author=

    Dense Passage Retrieval for Open-Domain Question Answering. , author=. EMNLP (1) , pages=

  10. [21]

    Proceedings of the 31st International Conference on Computational Linguistics , pages=

    Compress to Impress: Unleashing the Potential of Compressive Memory in Real-World Long-Term Conversations , author=. Proceedings of the 31st International Conference on Computational Linguistics , pages=

  11. [22]

    Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing , pages=

    Crafting Personalized Agents through Retrieval-Augmented Generation on Editable Memory Graphs , author=. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing , pages=

    2024

  12. [23]

    arXiv e-prints , pages=

    Theanine: Revisiting memory management in long-term conversations with timeline-augmented response generation , author=. arXiv e-prints , pages=

  13. [25]

    Towards Lifelong Dialogue Agents via Timeline-based Memory Management , author=. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) , pages=

    2025

  14. [26]

    arXiv preprint arXiv:2410.17509 , year=

    WAGLE: Strategic weight attribution for effective and modular unlearning in large language models , author=. arXiv preprint arXiv:2410.17509 , year=

  15. [29]

    and Stoica, Ion and Gonzalez, Joseph E

    Packer, Charles and Wooders, Sarah and Lin, Kevin and Fang, Vivian and Patil, Shishir G. and Stoica, Ion and Gonzalez, Joseph E. , journal=

  16. [30]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Memorybank: Enhancing large language models with long-term memory , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  17. [31]

    Hipporag: Neurobiologically inspired long-term memory for large language models , author=

  18. [32]

    arXiv preprint arXiv:2404.07103 , year=

    Graph chain-of-thought: Augmenting large language models by reasoning on graphs , author=. arXiv preprint arXiv:2404.07103 , year=

  19. [35]

    Advances in Neural Information Processing Systems , volume=

    G-retriever: Retrieval-augmented generation for textual graph understanding and question answering , author=. Advances in Neural Information Processing Systems , volume=

  20. [36]

    Advances in Neural Information Processing Systems , volume=

    Tablerag: Million-token table understanding with language models , author=. Advances in Neural Information Processing Systems , volume=

  21. [37]

    Parth Sarthi and Salman Abdullah and Aditi Tuli and Shubh Khanna and Anna Goldie and Christopher D Manning , booktitle=

  22. [38]

    Lightrag: Simple and fast retrieval-augmented generation , author=

  23. [39]

    arXiv preprint arXiv:2404.16130 , year=

    From local to global: A graph rag approach to query-focused summarization , author=. arXiv preprint arXiv:2404.16130 , year=

  24. [40]

    arXiv preprint arXiv:2410.08815 , year=

    Structrag: Boosting knowledge intensive reasoning of llms via inference-time hybrid information structurization , author=. arXiv preprint arXiv:2410.08815 , year=

  25. [42]

    Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

    Evaluating Very Long-Term Conversational Memory of LLM Agents , author=. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

  26. [43]

    The Thirteenth International Conference on Learning Representations , year=

    LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory , author=. The Thirteenth International Conference on Learning Representations , year=

  27. [45]

    Findings of the Association for Computational Linguistics: ACL 2023 , pages=

    Prompted LLMs as Chatbot Modules for Long Open-domain Conversation , author=. Findings of the Association for Computational Linguistics: ACL 2023 , pages=

    2023

  28. [47]

    Frontiers of Computer Science , volume=

    A survey on large language model based autonomous agents , author=. Frontiers of Computer Science , volume=. 2024 , publisher=

    2024

  29. [48]

    arXiv preprint arXiv:2504.10147 , year=

    A Survey of Personalization: From RAG to Agent , author=. arXiv preprint arXiv:2504.10147 , year=

  30. [50]

    Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

    ChatDev: Communicative Agents for Software Development , author=. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

  31. [51]

    arXiv preprint arXiv:2307.03172 , year=

    Lost in the middle: How language models use long contexts , author=. arXiv preprint arXiv:2307.03172 , year=

    arXiv

  32. [52]

    Proceedings of the 36th annual acm symposium on user interface software and technology , pages=

    Generative agents: Interactive simulacra of human behavior , author=. Proceedings of the 36th annual acm symposium on user interface software and technology , pages=

  33. [53]

    Advances in neural information processing systems , volume=

    Mpnet: Masked and permuted pre-training for language understanding , author=. Advances in neural information processing systems , volume=

  34. [54]

    P er LTQA : A Personal Long-Term Memory Dataset for Memory Classification, Retrieval, and Fusion in Question Answering

    Du, Yiming and Wang, Hongru and Zhao, Zhengyi and Liang, Bin and Wang, Baojun and Zhong, Wanjun and Wang, Zezhong and Wong, Kam-Fai. P er LTQA : A Personal Long-Term Memory Dataset for Memory Classification, Retrieval, and Fusion in Question Answering. Proceedings of the 10th ...

    2024

  35. [55]

    Gonzalez and Ion Stoica , booktitle=

    Lianmin Zheng and Wei-Lin Chiang and Ying Sheng and Siyuan Zhuang and Zhanghao Wu and Yonghao Zhuang and Zi Lin and Zhuohan Li and Dacheng Li and Eric Xing and Hao Zhang and Joseph E. Gonzalez and Ion Stoica , booktitle=. Judging

  36. [56]

    , author=

    The hippocampal memory indexing theory. , author=. Behavioral neuroscience , volume=. 1986 , publisher=

    1986

  37. [57]

    2025 , eprint=

    MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentation , author=. 2025 , eprint=

    2025

  38. [58]

    Advances in neural information processing systems , volume=

    The infinite Gaussian mixture model , author=. Advances in neural information processing systems , volume=

  39. [59]

    IEEE Transactions on Knowledge and Data Engineering , year=

    Efficient algorithms for personalized pagerank computation: A survey , author=. IEEE Transactions on Knowledge and Data Engineering , year=

  40. [60]

    Proceedings of the 11th international conference on World Wide Web , pages=

    Topic-sensitive pagerank , author=. Proceedings of the 11th international conference on World Wide Web , pages=

  41. [61]

    arXiv preprint arXiv:2312.17257 , year=

    Personalized Large Language Model Assistant with Evolving Conditional Memory , author=. arXiv preprint arXiv:2312.17257 , year=

  42. [62]

    arXiv preprint arXiv:2112.01488 , year=

    Colbertv2: Effective and efficient retrieval via lightweight late interaction , author=. arXiv preprint arXiv:2112.01488 , year=

  43. [63]

    arXiv preprint arXiv:2501.13958 , year=

    A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models , author=. arXiv preprint arXiv:2501.13958 , year=

  44. [70]

    arXiv preprint arXiv:2507.22925 , year=

    Hierarchical memory for high-efficiency long-term reasoning in llm agents , author=. arXiv preprint arXiv:2507.22925 , year=

  45. [71]

    Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

    Long-context language modeling with parallel context encoding , author=. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

  46. [78]

    First Workshop on Multi-Turn Interactions in Large Language Models , year=

    PersonaAgent: When Large Language Model Agents Meet Personalization at Test Time , author=. First Workshop on Multi-Turn Interactions in Large Language Models , year=

  47. [81]

    Psychological Bulletin , volume=

    Is memory schematic? , author=. Psychological Bulletin , volume=. 1983 , publisher=

    1983

  48. [82]

    2025 , eprint=

    Memory in the Age of AI Agents , author=. 2025 , eprint=

    2025

  49. [83]

    arXiv preprint arXiv:2505.15456 , year=

    Teaching Language Models to Evolve with Users: Dynamic Profile Modeling for Personalized Alignment , author=. arXiv preprint arXiv:2505.15456 , year=

  50. [85]

    Forty-second International Conference on Machine Learning , year=

    Agent Workflow Memory , author=. Forty-second International Conference on Machine Learning , year=

  51. [88]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Expel: Llm agents are experiential learners , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  52. [89]

    The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

    G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems , author=. The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

  53. [91]

    Findings of the Association for Computational Linguistics: ACL 2025 , pages=

    Personabench: Evaluating ai models on understanding personal information through accessing (synthetic) private user data , author=. Findings of the Association for Computational Linguistics: ACL 2025 , pages=

    2025

  54. [92]

    Advances in neural information processing systems , volume=

    Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers , author=. Advances in neural information processing systems , volume=

  55. [94]

    arXiv preprint arXiv:2505.09388 , year=

    Qwen3 technical report , author=. arXiv preprint arXiv:2505.09388 , year=

    arXiv

  56. [95]

    2025 , eprint=

    The Landscape of Agentic Reinforcement Learning for LLMs: A Survey , author=. 2025 , eprint=

    2025

  57. [96]

    Evoking User Memory: Personalizing

    Yingyi Zhang and Junyi Li and Wenlin Zhang and Pengyue Jia and Xianneng Li and Yichao Wang and Derong Xu and Yi Wen and Huifeng Guo and Yong Liu and Xiangyu Zhao , booktitle=. Evoking User Memory: Personalizing. 2026 , url=

    2026

  58. [97]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Personalize before retrieve: Llm-based personalized query expansion for user-centric retrieval , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  59. [98]

    The Fourteenth International Conference on Learning Representations , year=

    From Single to Multi-Granularity: Toward Long-Term Memory Association and Selection of Conversational Agents , author=. The Fourteenth International Conference on Learning Representations , year=

  60. [99]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Harnessing large language models for knowledge graph question answering via adaptive multi-aspect retrieval-augmentation , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  61. [101]

    Multi-perspective Improvement of Knowledge Graph Completion with Large Language Models

    Xu, Derong and Zhang, Ziheng and Lin, Zhenxi and Wu, Xian and Zhu, Zhihong and Xu, Tong and Zhao, Xiangyu and Zheng, Yefeng and Chen, Enhong. Multi-perspective Improvement of Knowledge Graph Completion with Large Language Models. Proceedings of the 2024 Joint International Con...

    2024

  62. [102]

    Proceedings of the ACM on Web Conference 2025 , pages=

    Llm4rerank: Llm-based auto-reranking framework for recommendations , author=. Proceedings of the ACM on Web Conference 2025 , pages=

    2025

  63. [103]

    Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V

    Notellm-2: Multimodal large representation models for recommendation , author=. Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 1 , pages=

  64. [104]

    ACM Transactions on Information Systems , volume=

    A unified framework for multi-domain ctr prediction via large language models , author=. ACM Transactions on Information Systems , volume=. 2025 , publisher=

    2025

  65. [105]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Llm-powered user simulator for recommender system , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  66. [106]

    Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V

    Large Language Model Enhanced Recommender Systems: Methods, Applications and Trends , author=. Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 2 , pages=

  67. [108]

    Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) , pages=

    Mill: Mutual verification with large language models for zero-shot query expansion , author=. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) , pages=

    2024

  68. [109]

    Nature , volume=

    Optimizing generative ai by backpropagating language model feedback , author=. Nature , volume=. 2025 , publisher=

    2025

  69. [110]

    The Twelfth International Conference on Learning Representations , year=

    Large language models as optimizers , author=. The Twelfth International Conference on Learning Representations , year=

  70. [111]

    gradient descent

    Automatic prompt optimization with “gradient descent” and beam search , author=. Proceedings of the 2023 conference on empirical methods in natural language processing , pages=

    2023

  71. [112]

    Advances in neural information processing systems , volume=

    Reflexion: Language agents with verbal reinforcement learning , author=. Advances in neural information processing systems , volume=

  72. [113]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Unleashing the potential of large language models as prompt optimizers: Analogical analysis with gradient-based model optimizers , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  73. [114]

    Forty-first International Conference on Machine Learning , year=

    Offline training of language model agents with functions as learnable weights , author=. Forty-first International Conference on Machine Learning , year=

  74. [118]

    Joseph W Alba and Lynn Hasher. 1983. Is memory schematic? Psychological Bulletin, 93(2):203

    1983

  75. [119]

    Nuo Chen, Hongguang Li, Jianhui Chang, Juhua Huang, Baoyuan Wang, and Jia Li. 2025 a . Compress to impress: Unleashing the potential of compressive memory in real-world long-term conversations. In Proceedings of the 31st International Conference on Computational Linguistics, p...

    2025

  76. [120]

    Nuo Chen, Hongguang Li, Jianhui Chang, Juhua Huang, Baoyuan Wang, and Jia Li. 2025 b . Compress to impress: Unleashing the potential of compressive memory in real-world long-term conversations. In Proceedings of the 31st International Conference on Computational Linguistics, p...

    2025

  77. [121]

    Prateek Chhikara, Dev Khant, Saket Aryan, Taranjeet Singh, and Deshraj Yadav. 2025. Mem0: Building production-ready ai agents with scalable long-term memory. arXiv preprint arXiv:2504.19413

    2025

  78. [122]

    Yimin Deng, Yuqing Fu, Derong Xu, Yejing Wang, Wei Ni, Jingtong Gao, Xiaopeng Li, Chengxu Liu, Xiao Han, Guoshuai Zhao, and 1 others. 2026. Enhancing conversational agents via task-oriented adversarial memory adaptation. arXiv preprint arXiv:2601.21797

    2026

  79. [123]

    Yiming Du, Hongru Wang, Zhengyi Zhao, Bin Liang, Baojun Wang, Wanjun Zhong, Zezhong Wang, and Kam-Fai Wong. 2024. Perltqa: A personal long-term memory dataset for memory classification, retrieval, and fusion in question answering. In Proceedings of the 10th SIGHAN Workshop on ...

    2024

  80. [124]

    Jizhan Fang, Xinle Deng, Haoming Xu, Ziyan Jiang, Yuqi Tang, Ziwen Xu, Shumin Deng, Yunzhi Yao, Mengru Wang, Shuofei Qiao, and 1 others. 2025. Lightmem: Lightweight and efficient memory-augmented generation. arXiv preprint arXiv:2510.18866

    2025

  81. [125]

    Jingtong Gao, Bo Chen, Xiangyu Zhao, Weiwen Liu, Xiangyang Li, Yichao Wang, Wanyu Wang, Huifeng Guo, and Ruiming Tang. 2025. Llm4rerank: Llm-based auto-reranking framework for recommendations. In Proceedings of the ACM on Web Conference 2025, pages 228--239

    2025

  82. [126]

    Bernal Jim \'e nez Guti \'e rrez, Yiheng Shu, Weijian Qi, Sizhe Zhou, and Yu Su. 2025. From rag to memory: Non-parametric continual learning for large language models. arXiv preprint arXiv:2502.14802

    2025

  83. [127]

    Yuyang Hu, Shichun Liu, Yanwei Yue, Guibin Zhang, Boyang Liu, Fangyi Zhu, Jiahang Lin, Honglin Guo, Shihan Dou, Zhiheng Xi, Senjie Jin, Jiejun Tan, Yanbin Yin, Jiongnan Liu, Zeyu Zhang, Zhongxiang Sun, Yutao Zhu, Hao Sun, Boci Peng, and 28 others. 2025. https://arxiv.org/abs/2...

    2025

  84. [128]

    Pengyue Jia, Yiding Liu, Xiangyu Zhao, Xiaopeng Li, Changying Hao, Shuaiqiang Wang, and Dawei Yin. 2024. Mill: Mutual verification with large language models for zero-shot query expansion. In Proceedings of the 2024 Conference of the North American Chapter of the Association f...

    2024

  85. [129]

    Bowen Jiang, Zhuoqun Hao, Young-Min Cho, Bryan Li, Yuan Yuan, Sihao Chen, Lyle Ungar, Camillo J Taylor, and Dan Roth. 2025 a . Know me, respond to me: Benchmarking llms for dynamic user profiling and personalized responses at scale. arXiv preprint arXiv:2504.14225

    2025

  86. [130]

    Bowen Jiang, Yuan Yuan, Maohao Shen, Zhuoqun Hao, Zhangchen Xu, Zichen Chen, Ziyi Liu, Anvesh Rao Vijjini, Jiashu He, Hanchao Yu, and 1 others. 2025 b . Personamem-v2: Towards personalized intelligence via learning implicit user personas and agentic memory. arXiv preprint arXi...

    2025

  87. [131]

    Seo Hyun Kim, Tzu-iunn Ong, Taeyoon Kwon, Namyoung Kim, Keummin Ka, SeongHyeon Bae, Yohan Jo, Seung-won Hwang, Dongha Lee, Jinyoung Yeo, and 1 others. 2024. Theanine: Revisiting memory management in long-term conversations with timeline-augmented response generation. arXiv e-p...

    2024

  88. [132]

    Kuang-Huei Lee, Xinyun Chen, Hiroki Furuta, John Canny, and Ian Fischer. 2024. A human-inspired reading agent with gist memory of very long contexts. arXiv preprint arXiv:2402.09727

    2024

  89. [133]

    Hao Li, Chenghao Yang, An Zhang, Yang Deng, Xiang Wang, and Tat-Seng Chua. 2024 a . Hello again! llm-powered personalized agent for long-term dialogue. arXiv preprint arXiv:2406.05925

    2024

  90. [134]

    Yuanchun Li, Hao Wen, Weijun Wang, Xiangyu Li, Yizhen Yuan, Guohong Liu, Jiacheng Liu, Wenxing Xu, Xiang Wang, Yi Sun, and 1 others. 2024 b . Personal llm agents: Insights and survey about the capability, efficiency and security. arXiv preprint arXiv:2401.05459

    2024

  91. [135]

    Yuchen Li, Hengyi Cai, Rui Kong, Xinran Chen, Jiamin Chen, Jun Yang, Haojie Zhang, Jiayi Li, Jiayi Wu, Yiqun Chen, and 1 others. 2025 a . Towards ai search paradigm. arXiv preprint arXiv:2506.17188

    2025

  92. [136]

    Zhiyu Li, Shichao Song, Chenyang Xi, Hanyu Wang, Chen Tang, Simin Niu, Ding Chen, Jiawei Yang, Chunyu Li, Qingchen Yu, and 1 others. 2025 b . Memos: A memory os for ai system. arXiv preprint arXiv:2507.03724

    2025

  93. [137]

    Lei Liu, Xiaoyan Yang, Yue Shen, Binbin Hu, Zhiqiang Zhang, Jinjie Gu, and Guannan Zhang. 2023. Think-in-memory: Recalling and post-thinking enable llms with long-term memory. arXiv preprint arXiv:2311.08719

    2023

  94. [138]

    Qidong Liu, Xiangyu Zhao, Yuhao Wang, Yejing Wang, Zijian Zhang, Yuqi Sun, Xiang Li, Maolin Wang, Pengyue Jia, Chong Chen, and 1 others. 2025. Large language model enhanced recommender systems: Methods, applications and trends. In Proceedings of the 31st ACM SIGKDD Conference ...

    2025

  95. [139]

    Lin Long, Yichen He, Wentao Ye, Yiyuan Pan, Yuan Lin, Hang Li, Junbo Zhao, and Wei Li. 2025. Seeing, listening, remembering, and reasoning: A multimodal agent with long-term memory. arXiv preprint arXiv:2508.09736

    2025

  96. [140]

    Junru Lu, Siyu An, Mingbao Lin, Gabriele Pergola, Yulan He, Di Yin, Xing Sun, and Yunsheng Wu. 2023. Memochat: Tuning llms to use memos for consistent long-range open-domain conversation. arXiv preprint arXiv:2308.08239

    2023

  97. [141]

    Siru Ouyang, Jun Yan, I Hsu, Yanfei Chen, Ke Jiang, Zifeng Wang, Rujun Han, Long T Le, Samira Daruki, Xiangru Tang, and 1 others. 2025. Reasoningbank: Scaling agent self-evolving with reasoning memory. arXiv preprint arXiv:2509.25140

    2025

  98. [142]

    Patil, Ion Stoica, and Joseph E

    Charles Packer, Sarah Wooders, Kevin Lin, Vivian Fang, Shishir G. Patil, Ion Stoica, and Joseph E. Gonzalez. 2023. MemGPT : Towards llms as operating systems. arXiv preprint arXiv:2310.08560

    2023

  99. [143]

    Vicky Zhao, Lili Qiu, and Jianfeng Gao

    Zhuoshi Pan, Qianhui Wu, Huiqiang Jiang, Xufang Luo, Hao Cheng, Dongsheng Li, Yuqing Yang, Chin-Yew Lin, H. Vicky Zhao, Lili Qiu, and Jianfeng Gao. 2025. Secom: On memory construction and retrieval for personalized conversational agents. In The Thirteenth International Confere...

    2025

  100. [144]

    gradient descent

    Reid Pryzant, Dan Iter, Jerry Li, Yin Lee, Chenguang Zhu, and Michael Zeng. 2023. Automatic prompt optimization with “gradient descent” and beam search. In Proceedings of the 2023 conference on empirical methods in natural language processing, pages 7957--7968

    2023

  101. [145]

    Hongjin Qian, Zheng Liu, Peitian Zhang, Kelong Mao, Defu Lian, Zhicheng Dou, and Tiejun Huang. 2025. https://arxiv.org/abs/2409.05591 Memorag: Boosting long context processing with global memory-enhanced retrieval augmentation . Preprint, arXiv:2409.05591

    2025

  102. [146]

    Preston Rasmussen, Pavlo Paliychuk, Travis Beauvais, Jack Ryan, and Daniel Chalef. 2025. Zep: a temporal knowledge graph architecture for agent memory. arXiv preprint arXiv:2501.13956

    2025

  103. [147]

    Alireza Rezazadeh, Zichao Li, Wei Wei, and Yujia Bao. 2024. From isolated conversations to hierarchical schemas: Dynamic tree memory representation for llms. arXiv preprint arXiv:2410.14052

    2024

  104. [148]

    Parth Sarthi, Salman Abdullah, Aditi Tuli, Shubh Khanna, Anna Goldie, and Christopher D Manning. 2024. RAPTOR : Recursive abstractive processing for tree-organized retrieval. In The Twelfth International Conference on Learning Representations

    2024

  105. [149]

    Noah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. 2023. Reflexion: Language agents with verbal reinforcement learning. Advances in neural information processing systems, 36:8634--8652

    2023

  106. [150]

    Juntao Tan, Liangwei Yang, Zuxin Liu, Zhiwei Liu, Rithesh RN, Tulika Manoj Awalgaonkar, Jianguo Zhang, Weiran Yao, Ming Zhu, Shirley Kokane, and 1 others. 2025 a . Personabench: Evaluating ai models on understanding personal information through accessing (synthetic) private us...

    2025

  107. [151]

    Zhen Tan, Jun Yan, I Hsu, Rujun Han, Zifeng Wang, Long T Le, Yiwen Song, Yanfei Chen, Hamid Palangi, George Lee, and 1 others. 2025 b . In prospect and retrospect: Reflective memory management for long-term personalized dialogue agents. arXiv preprint arXiv:2503.08026

    2025

  108. [152]

    Xiangru Tang, Tianrui Qin, Tianhao Peng, Ziyang Zhou, Daniel Shao, Tingting Du, Xinming Wei, Peng Xia, Fang Wu, He Zhu, and 1 others. 2025 a . Agent kb: Leveraging cross-domain experience for agentic problem solving. arXiv preprint arXiv:2507.06229

    2025

  109. [153]

    Xinyu Tang, Xiaolei Wang, Wayne Xin Zhao, Siyuan Lu, Yaliang Li, and Ji-Rong Wen. 2025 b . Unleashing the potential of large language models as prompt optimizers: Analogical analysis with gradient-based model optimizers. In Proceedings of the AAAI Conference on Artificial Inte...

    2025

  110. [154]

    LangChain Team. 2023. Conversation summary memory

    2023

  111. [155]

    Timothy J Teyler and Pascal DiScenna. 1986. The hippocampal memory indexing theory. Behavioral neuroscience, 100(2):147

    1986

  112. [156]

    Qingyue Wang, Yanhe Fu, Yanan Cao, Shuai Wang, Zhiliang Tian, and Liang Ding. 2025 a . Recursively summarizing enables long-term dialogue memory in large language models. Neurocomputing, page 130193

    2025

  113. [157]

    Wenhui Wang, Furu Wei, Li Dong, Hangbo Bao, Nan Yang, and Ming Zhou. 2020. Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. Advances in neural information processing systems, 33:5776--5788

    2020

  114. [158]

    Yu Wang and Xi Chen. 2025. Mirix: Multi-agent memory system for llm-based agents. arXiv preprint arXiv:2507.07957

    2025

  115. [159]

    Yu Wang, Ryuichi Takanobu, Zhiqi Liang, Yuzhen Mao, Yuanzhe Hu, Julian McAuley, and Xiaojian Wu. 2025 b . Mem- \ alpha \ : Learning memory construction via reinforcement learning. arXiv preprint arXiv:2509.25911

    2025

  116. [160]

    Zora Zhiruo Wang, Jiayuan Mao, Daniel Fried, and Graham Neubig. 2025 c . https://openreview.net/forum?id=NTAhi2JEEE Agent workflow memory . In Forty-second International Conference on Machine Learning

    2025

  117. [161]

    Qingsong Wen, Jing Liang, Carles Sierra, Rose Luckin, Richard Tong, Zitao Liu, Peng Cui, and Jiliang Tang. 2024. https://doi.org/10.1145/3637528.3671498 Ai for education (ai4edu): Advancing personalized education with llm and adaptive learning . In Proceedings of the 30th ACM ...

    2024 doi

  118. [162]

    Yaxiong Wu, Sheng Liang, Chen Zhang, Yichao Wang, Yongyue Zhang, Huifeng Guo, Ruiming Tang, and Yong Liu. 2025. From human memory to ai memory: A survey on memory mechanisms in the era of llms. arXiv preprint arXiv:2504.15965

    2025

  119. [163]

    Derong Xu, Xinhang Li, Ziheng Zhang, Zhenxi Lin, Zhihong Zhu, Zhi Zheng, Xian Wu, Xiangyu Zhao, Tong Xu, and Enhong Chen. 2025 a . Harnessing large language models for knowledge graph question answering via adaptive multi-aspect retrieval-augmentation. In Proceedings of the AA...

    2025

  120. [164]

    Derong Xu, Yi Wen, Pengyue Jia, Yingyi Zhang, Wenlin Zhang, Yichao Wang, Huifeng Guo, Ruiming Tang, Xiangyu Zhao, Enhong Chen, and Tong Xu. 2026. https://openreview.net/forum?id=i2yIvZARnG From single to multi-granularity: Toward long-term memory association and selection of c...

    2026

  121. [165]

    Derong Xu, Ziheng Zhang, Zhenxi Lin, Xian Wu, Zhihong Zhu, Tong Xu, Xiangyu Zhao, Yefeng Zheng, and Enhong Chen. 2024. https://aclanthology.org/2024.lrec-main.1044/ Multi-perspective improvement of knowledge graph completion with large language models . In Proceedings of the 2...

    2024

  122. [166]

    Fangyuan Xu, Weijia Shi, and Eunsol Choi. 2023. Recomp: Improving retrieval-augmented lms with compression and selective augmentation. arXiv preprint arXiv:2310.04408

    2023

  123. [167]

    Wujiang Xu, Kai Mei, Hang Gao, Juntao Tan, Zujie Liang, and Yongfeng Zhang. 2025 b . A-mem: Agentic memory for llm agents. arXiv preprint arXiv:2502.12110

    2025

  124. [168]

    Sikuan Yan, Xiufeng Yang, Zuchao Huang, Ercong Nie, Zifeng Ding, Zonggen Li, Xiaowen Ma, Kristian Kersting, Jeff Z Pan, Hinrich Sch \"u tze, and 1 others. 2025. Memory-r1: Enhancing large language model agents to manage and utilize memories via reinforcement learning. arXiv pr...

    2025

  125. [169]

    An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, and 1 others. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115

    2024 arXiv

  126. [170]

    Chengrun Yang, Xuezhi Wang, Yifeng Lu, Hanxiao Liu, Quoc V Le, Denny Zhou, and Xinyun Chen. 2023. Large language models as optimizers. In The Twelfth International Conference on Learning Representations

    2023

  127. [171]

    Howard Yen, Tianyu Gao, and Danqi Chen. 2024. Long-context language modeling with parallel context encoding. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2588--2610

    2024

  128. [172]

    Hongli Yu, Tinghong Chen, Jiangtao Feng, Jiangjie Chen, Weinan Dai, Qiying Yu, Ya-Qin Zhang, Wei-Ying Ma, Jingjing Liu, Mingxuan Wang, and 1 others. 2025. Memagent: Reshaping long-context llm with multi-conv rl-based memory agent. arXiv preprint arXiv:2507.02259

    2025

  129. [173]

    Qianhao Yuan, Jie Lou, Zichao Li, Jiawei Chen, Yaojie Lu, Hongyu Lin, Le Sun, Debing Zhang, and Xianpei Han. 2025. Memsearcher: Training llms to reason, search and manage memory via end-to-end reinforcement learning. arXiv preprint arXiv:2511.02805

    2025

  130. [174]

    Mert Yuksekgonul, Federico Bianchi, Joseph Boen, Sheng Liu, Pan Lu, Zhi Huang, Carlos Guestrin, and James Zou. 2025. Optimizing generative ai by backpropagating language model feedback. Nature, 639(8055):609--616

    2025

  131. [175]

    Chao Zhang, Haoxin Zhang, Shiwei Wu, Di Wu, Tong Xu, Xiangyu Zhao, Yan Gao, Yao Hu, and Enhong Chen. 2025 a . Notellm-2: Multimodal large representation models for recommendation. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 1, pag...

    2025

  132. [176]

    Guibin Zhang, Muxin Fu, Kun Wang, Guancheng Wan, Miao Yu, and Shuicheng YAN. 2025 b . https://openreview.net/forum?id=mmIAp3cVS0 G-memory: Tracing hierarchical memory for multi-agent systems . In The Thirty-ninth Annual Conference on Neural Information Processing Systems

    2025

  133. [177]

    Guibin Zhang, Muxin Fu, and Shuicheng Yan. 2025 c . Memgen: Weaving generative latent memory for self-evolving agents. arXiv preprint arXiv:2509.24704

    2025

  134. [178]

    Guibin Zhang, Hejia Geng, Xiaohang Yu, Zhenfei Yin, Zaibin Zhang, Zelin Tan, Heng Zhou, Zhongzhi Li, Xiangyuan Xue, Yijiang Li, Yifan Zhou, Yang Chen, Chen Zhang, Yutao Fan, Zihu Wang, Songtao Huang, Francisco Piedrahita-Velez, Yue Liao, Hongru Wang, and 6 others. 2025 d . htt...

    2025 arXiv

  135. [179]

    Jiayi Zhang, Jinyu Xiang, Zhaoyang Yu, Fengwei Teng, Xionghui Chen, Jiaqi Chen, Mingchen Zhuge, Xin Cheng, Sirui Hong, Jinlin Wang, and 1 others. 2024 a . Aflow: Automating agentic workflow generation. arXiv preprint arXiv:2410.10762

    2024

  136. [180]

    Peiyan Zhang, Haibo Jin, Leyang Hu, Xinnuo Li, Liying Kang, Man Luo, Yangqiu Song, and Haohan Wang. 2024 b . Revolve: Optimizing ai systems by tracking response evolution in textual optimization. arXiv preprint arXiv:2412.03092

    2024

  137. [181]

    Shaokun Zhang, Jieyu Zhang, Jiale Liu, Linxin Song, Chi Wang, Ranjay Krishna, and Qingyun Wu. 2024 c . Offline training of language model agents with functions as learnable weights. In Forty-first International Conference on Machine Learning

    2024

  138. [182]

    Yu, and Xian Li

    Weizhi Zhang, Xinyang Zhang, Chenwei Zhang, Liangwei Yang, Jingbo Shang, Zhepei Wei, Henry Peng Zou, Zijie Huang, Zhengyang Wang, Yifan Gao, Xiaoman Pan, Lian Xiong, Jingguo Liu, Philip S. Yu, and Xian Li. 2025 e . Personaagent: When large language model agents meet personaliz...

    2025

  139. [183]

    Yingyi Zhang, Pengyue Jia, Derong Xu, Yi Wen, Xianneng Li, Yichao Wang, Wenlin Zhang, Xiaopeng Li, Weinan Gan, Huifeng Guo, and 1 others. 2026 a . Personalize before retrieve: Llm-based personalized query expansion for user-centric retrieval. In Proceedings of the AAAI Confere...

    2026

  140. [184]

    Yingyi Zhang, Junyi Li, Wenlin Zhang, Pengyue Jia, Xianneng Li, Yichao Wang, Derong Xu, Yi Wen, Huifeng Guo, Yong Liu, and Xiangyu Zhao. 2026 b . https://openreview.net/forum?id=f7p0F2X6XN Evoking user memory: Personalizing LLM via recollection-familiarity adaptive retrieval ....

    2026

  141. [185]

    Yuxiang Zhang, Jiangming Shu, Ye Ma, Xueyuan Lin, Shangxi Wu, and Jitao Sang. 2025 f . Memory as action: Autonomous context curation for long-horizon agentic tasks. arXiv preprint arXiv:2510.12635

    2025 arXiv

  142. [186]

    Zeyu Zhang, Xiaohe Bo, Chen Ma, Rui Li, Xu Chen, Quanyu Dai, Jieming Zhu, Zhenhua Dong, and Ji-Rong Wen. 2024 d . A survey on the memory mechanism of large language model based agents. arXiv preprint arXiv:2404.13501

    2024

  143. [187]

    Andrew Zhao, Daniel Huang, Quentin Xu, Matthieu Lin, Yong-Jin Liu, and Gao Huang. 2024. Expel: Llm agents are experiential learners. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 19632--19642

    2024

  144. [188]

    Siyan Zhao, Mingyi Hong, Yang Liu, Devamanyu Hazarika, and Kaixiang Lin. 2025. Do llms recognize your preferences? evaluating personalized preference following in llms. arXiv preprint arXiv:2502.09597

    2025

  145. [189]

    Wanjun Zhong, Lianghong Guo, Qiqi Gao, He Ye, and Yanlin Wang. 2024. Memorybank: Enhancing large language models with long-term memory. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 19724--19731

    2024

  146. [190]

    Zijian Zhou, Ao Qu, Zhaoxuan Wu, Sunghwan Kim, Alok Prakash, Daniela Rus, Jinhua Zhao, Bryan Kian Hsiang Low, and Paul Pu Liang. 2025. Mem1: Learning to synergize memory and reasoning for efficient long-horizon agents. arXiv preprint arXiv:2506.15841

    2025

Pith tools

Reviewed May 9, 2026 · model on record in the stance chip above.