Elamur Research⌕
☰
home / papers / 2026 / GAM: Hierarchical Graph-based Agentic Memory for LLM Agents

GAM: Hierarchical Graph-based Agentic Memory for LLM Agents

Paper · 2026 · Zhaofen Wu, Hanrong Zhang, Fulin Lin et al.

Summary

LLM agents in long conversations must absorb new information without letting passing chatter corrupt what they already know. Stream-based memories such as MemGPT or Mem0 update readily but suffer memory loss and topic conflation. Rigid graph-based stores such as GraphRAG are stable but cannot follow how a dialogue develops. GAM resolves this by separating the writing of memory from its consolidation. Incoming turns go into a local Event Progression Graph, which acts as a buffer. Only when a semantic divergence indicator detects a topic shift is the buffered episode summarized and merged into a global Topic Associative Network. The authors cite sleep-dependent memory consolidation as the inspiration. At query time, retrieval works top-down: it first finds the relevant topics, then descends to event-level details, ranking candidates by temporal, confidence and speaker-role signals. On the LoCoMo and LongDialQA benchmarks, GAM outperforms six memory baselines across several backbone LLMs while using fewer tokens. This makes it a practical option for long-horizon assistants. The explicit graph structure also gives a natural point for privacy filtering and selective deletion.

Key points

  • Two-level graph: local Event Progression Graphs act as a write-isolated buffer, and a global Topic Associative Network holds consolidated knowledge. The system switches between an Episodic Buffering State and a Semantic Consolidation State.
  • Setup: baselines were ReadAgent, MemoryBank, MemGPT, A-Mem, MemoryOS and Mem0, tested with GPT-4o-mini, Qwen 2.5-7B and Qwen 2.5-14B. Metrics were F1 and BLEU-1.
  • Efficiency on LoCoMo:
    • GAM reaches an F1 of 40.00 versus 35.38 for Mem0.
    • It uses about 1,370 tokens per query, roughly 11% fewer than Mem0.
    • Its latency is 0.80 s, slightly slower than Mem0's 0.51 s.
    • A-Mem needs about 4,221 tokens per query, and MemoryOS takes about 154 s.
  • Segmentation: splitting dialogue at semantic boundaries beats fixed token windows (256/512), fixed turn counts (k=3/5) and session-based splitting.
  • Ablations and scope: removing any of the four components (TAN, MFR, SSM, EPG) lowers scores. The system is currently text-only, and multimodal memory is left for future work.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Zhejiang UniversityHangzhou · China224 works · 13,587 citations🎓McGill UniversityMontreal · Canada91 works · 11,178 citations🎓University of Illinois ChicagoChicago · United States70 works · 20,363 citations🎓Mohamed bin Zayed University of Artificial IntelligenceAbu Dhabi · United Arab Emirates138 works · 6,590 citations🎓Rutgers, The State University of New JerseyNew Brunswick · United States40 works · 12,486 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

Cite

Cite as Zhaofen Wu, Hanrong Zhang, Fulin Lin, Wujiang Xu, Xinran Xu, Yankai Chen, Henry Peng Zou, Shaowen Chen, Weizhi Zhang, Xue Liu, Philip S. Yu and Hongwei Wang. 2026. GAM: Hierarchical Graph-based Agentic Memory for LLM Agents. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.1600

@inproceedings{wu2026gam,
  title = {{GAM: Hierarchical Graph-based Agentic Memory for LLM Agents}},
  author = {Zhaofen Wu, Hanrong Zhang, Fulin Lin, Wujiang Xu, Xinran Xu, Yankai Chen, Henry Peng Zou, Shaowen Chen, Weizhi Zhang, Xue Liu, Philip S. Yu and Hongwei Wang},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1600},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1600}
}

References

34 references, 10 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Petr Anokhin, Nikita Semenov, Artyom Sorokin, Dmitry Evseev, Andrey Kravchenko, Mikhail Burtsev, and Evgeny Burnaev. 2025. Arigraph: learning knowledge graph world models with episodic memory for llm agents. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligenc… find
  2. Abrar Anwar, John Welsh, Joydeep Biswas, Soha Pouya, and Yan Chang. 2025. Remembr: Building and reasoning over long-horizon spatio-temporal memory for robot navigation. In 2025 IEEE International Conference on Robotics and Automation (ICRA), pages 2838–2845. find
  3. Hongyu Chang, Wenbo Tang, Annabella M Wulf, Thokozile Nyasulu, Madison E Wolf, Antonio Fernandez-Ruiz, and Azahara Oliva. 2025. Sleep microstructure organizes memory replay. Nature, 637(8048):1161–1169. find
  4. Ding Chen, Simin Niu, Kehang Li, Peng Liu, Xiangping Zheng, Bo Tang, Xinchi Li, Feiyu Xiong, and Zhiyu Li. 2026. Halumem: Evaluating hallucinations in memory systems of agents. Preprint, arXiv:2511.03506. — [arXiv:2511.03506](https://arxiv.org/abs/2511.03506) find
  5. Prateek Chhikara, Dev Khant, Saket Aryan, Taranjeet Singh, and Deshraj Yadav. 2025. Mem0: Building production-ready ai agents with scalable long-term memory. Preprint, arXiv:2504.19413.
  6. Aniket Didolkar, Anirudh Goyal, Nan Rosemary Ke, Siyuan Guo, Michal Valko, Timothy Lillicrap, Danilo Rezende, Yoshua Bengio, Michael Mozer, and Sanjeev Arora. 2024. Metacognitive capabilities of llms: An exploration in mathematical problem solving. In Advances in Neural Information Processing System… find
  7. Yufeng Du, Minyang Tian, Srikanth Ronanki, Subendhu Rongali, Sravan Bodapati, Aram Galstyan, Azton Wells, Roy Schwartz, Eliu A Huerta, and Hao Peng. 2025. Context length alone hurts llm performance despite perfect retrieval. Preprint, arXiv:2510.05381. find
  8. Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. 2025. From local to global: A graph rag approach to query-focused summarization. Preprint, arXiv:2404.16130. find
  9. Zirui Guo, Lianghao Xia, Yanhua Yu, Tu Ao, and Chao Huang. 2025. Lightrag: Simple and fast retrieval-augmented generation. Preprint, arXiv:2410.05779.
  10. 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. 2026. Memory in the age of ai agents. Preprint, arXi… — [arXiv:2512.13564](https://arxiv.org/abs/2512.13564) find
  11. Wei-Chieh Huang, Weizhi Zhang, Yueqing Liang, Yuanchen Bei, Yankai Chen, Tao Feng, Xinyu Pan, Zhen Tan, Yu Wang, Tianxin Wei, Shanglin Wu, Ruiyao Xu, Liangwei Yang, Rui Yang, Wooseong Yang, Chin-Yuan Yeh, Hanrong Zhang, Haozhen Zhang, Siqi Zhu, and 41 others. 2026. Rethinking memory mechanisms of fo… — [arXiv:2602.06052](https://arxiv.org/abs/2602.06052) find
  12. Runsong Jia, Bowen Zhang, Sergio José Rodríguez Méndez, and Pouya G. Omran. 2025a. Structrag: Structure-aware rag framework with scholarly knowledge graph for diverse question answering. In Companion Proceedings of the ACM on Web Conference 2025, pages 2567–2573. Association for Computing Machinery. find
  13. Zixi Jia, Qinghua Liu, Hexiao Li, Yuyan Chen, and Jiqiang Liu. 2025b. Evaluating the long-term memory of large language models. In Findings of the Association for Computational Linguistics: ACL 2025, pages 19759–19777. find
  14. Jiazheng Kang, Mingming Ji, Zhe Zhao, and Ting Bai. 2025. Memory OS of AI agent. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 25961–25970.
  15. Jiho Kim, Woosog Chay, Hyeonji Hwang, Daeun Kyung, Hyunseung Chung, Eunbyeol Cho, Yeonsu Kwon, Yohan Jo, and Edward Choi. 2024. DialSim: A dialogue simulator for evaluating long-term multi-party dialogue understanding of conversational agents. Preprint, arXiv:2406.13144. — [arXiv:2406.13144](https://arxiv.org/abs/2406.13144) find
  16. 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. In Proceedings of the 41st International Conference on Machine Learning, volume 235, pages 26396–26415.
  17. Yangning Li, Weizhi Zhang, Yuyao Yang, Wei-Chieh Huang, Yaozu Wu, Junyu Luo, Yuanchen Bei, Henry Peng Zou, Xiao Luo, Yusheng Zhao, Chunkit Chan, Yankai Chen, Zhongfen Deng, Yinghui Li, Hai-Tao Zheng, Dongyuan Li, Renhe Jiang, Ming Zhang, Yangqiu Song, and Philip S. Yu. 2025. Towards agentic rag with…
  18. Chuangtao Ma, Yongrui Chen, Tianxing Wu, Arijit Khan, and Haofen Wang. 2025. Large language models meet knowledge graphs for question answering: Synthesis and opportunities. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 24578–24597. find
  19. Adyasha Maharana, Dong-Ho Lee, Sergey Tulyakov, Mohit Bansal, Francesco Barbieri, and Yuwei Fang. 2024. Evaluating very long-term conversational memory of llm agents. Preprint, arXiv:2402.17753. find
  20. Charles Packer, Sarah Wooders, Kevin Lin, Vivian Fang, Shishir G. Patil, Ion Stoica, and Joseph E. Gonzalez. 2024. Memgpt: Towards llms as operating systems. Preprint, arXiv:2310.08560. find
  21. Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S. Bernstein. 2023. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology. find
  22. Preston Rasmussen, Pavlo Paliychuk, Travis Beauvais, Jack Ryan, and Daniel Chalef. 2025. Zep: A temporal knowledge graph architecture for agent memory. Preprint, arXiv:2501.13956.
  23. Miao Su, Yucan Guo, Zhongni Hou, Long Bai, Zixuan Li, Yufei Zhang, Guojun Yin, Wei Lin, Xiaolong Jin, Jiafeng Guo, and Xueqi Cheng. 2026. Beyond dialogue time: Temporal semantic memory for personalized llm agents. Preprint, arXiv:2601.07468.
  24. Zhen Tan, Jun Yan, I-Hung Hsu, Rujun Han, Zifeng Wang, Long Le, Yiwen Song, Yanfei Chen, Hamid Palangi, George Lee, Anand Rajan Iyer, Tianlong Chen, Huan Liu, Chen-Yu Lee, and Tomas Pfister. 2025. In prospect and retrospect: Reflective memory management for long-term personalized dialogue agents. In…
  25. Wujiang Xu, Zujie Liang, Kai Mei, Hang Gao, Juntao Tan, and Yongfeng Zhang. 2025. A-mem: Agentic memory for LLM agents. In The Thirty-ninth Annual Conference on Neural Information Processing Systems.
  26. Enneng Yang, Anke Tang, Li Shen, Guibing Guo, Xingwei Wang, Xiaochun Cao, and Jie Zhang. 2025a. Continual model merging without data: Dual projections for balancing stability and plasticity. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. find
  27. Shiyi Yang, Zhibo Hu, Xinshu Li, Chen Wang, Tong Yu, Xiwei Xu, Liming Zhu, and Lina Yao. 2025b. Drunkagent: Stealthy memory corruption in llm-powered recommender agents. Preprint, arXiv:2503.23804. — [arXiv:2503.23804](https://arxiv.org/abs/2503.23804) find
  28. Guibin Zhang, Muxin Fu, Kun Wang, Guancheng Wan, Miao Yu, and Shuicheng YAN. 2025a. G-memory: Tracing hierarchical memory for multi-agent systems. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. find
  29. Hanrong Zhang, Shicheng Fan, Henry Peng Zou, Yankai Chen, Zhenting Wang, Jiayu Zhou, Chengze Li, Wei-Chieh Huang, Yifei Yao, Kening Zheng, Xue Liu, Xiaoxiao Li, and Philip S. Coevoskills: Self-evolving agent Yu. 2026a. skills via co-evolutionary verification. Preprint, arXiv:2604.01687. — [arXiv:2604.01687](https://arxiv.org/abs/2604.01687) find
  30. Weizhi Zhang, Yangning Li, Yuanchen Bei, Junyu Luo, Guancheng Wan, Liangwei Yang, Chenxuan Xie, Yuyao Yang, Wei-Chieh Huang, Chunyu Miao, Henry Peng Zou, Xiao Luo, Yusheng Zhao, Yankai Chen, Chunkit Chan, Peilin Zhou, Xinyang Zhang, Chenwei Zhang, Jingbo Shang, and 4 others. 2025b. From web search t… — [arXiv:2506.18959](https://arxiv.org/abs/2506.18959) find
  31. Wenyuan Zhang, Xinghua Zhang, Haiyang Yu, Shuaiyi Nie, Bingli Wu, Juwei Yue, Tingwen Liu, and Expseek: Self-triggered Yongbin Li. 2026b. experience seeking for web agents. Preprint, arXiv:2601.08605. — [arXiv:2601.08605](https://arxiv.org/abs/2601.08605) find
  32. Wanjun Zhong, Lianghong Guo, Qiqi Gao, He Ye, and Yanlin Wang. 2024. MemoryBank: Enhancing large language models with long-term memory. Proceedings of the AAAI Conference on Artificial Intelligence, 38(17):19724–19731. find
  33. Henry Peng Zou, Wei-Chieh Huang, Yaozu Wu, Yankai Chen, Chunyu Miao, Hoang Nguyen, Yue Zhou, Weizhi Zhang, Liancheng Fang, Langzhou He, Yangning Li, Dongyuan Li, Renhe Jiang, Xue Liu, and Philip S. Yu. 2025. Llm-based human-agent collaboration and interaction systems: A survey. Preprint, arXiv:2505.…
  34. Henry Peng Zou, Chunyu Miao, Wei-Chieh Huang, Yankai Chen, Yue Zhou, Hanrong Zhang, Yaozu Wu, Liancheng Fang, Zhengyao Gu, Zhen Zhang, Kening Zheng, Fangxin Wang, Yi Nian, Shanghao Li, Wenzhe Fan, Langzhou He, Weizhi Zhang, Xue Liu, and Philip S. Yu. 2026. When users change their mind: Evaluating in… — [arXiv:2604.00892](https://arxiv.org/abs/2604.00892) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wu-2026-gam-hierarchical-graph-based-agentic/