@inproceedings{wu-etal-2026-gam,
    title = "{GAM}: Hierarchical Graph-based Agentic Memory for {LLM} Agents",
    author = "Wu, Zhaofen  and
      Zhang, Hanrong  and
      Lin, Fulin  and
      Xu, Wujiang  and
      Xu, Xinran  and
      Chen, Yankai  and
      Zou, Henry Peng  and
      Chen, Shaowen  and
      Zhang, Weizhi  and
      Liu, Xue  and
      Yu, Philip S.  and
      Wang, Hongwei",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1600/",
    pages = "34647--34664",
    ISBN = "979-8-89176-390-6",
    abstract = "To sustain coherent long-term interactions, Large Language Model (LLM) agents must navigate the tension between acquiring new information and retaining prior knowledge. Current unified stream-based memory systems facilitate context updates but remain vulnerable to interference from transient noise. Conversely, discrete structured memory architectures provide robust knowledge retention but often struggle to adapt to fluid narrative evolution. To address this, we propose GAM, a hierarchical Graph-based Agentic Memory framework that explicitly decouples memory encoding from consolidation to effectively resolve the conflict between rapid context perception and stable knowledge retention. By isolating ongoing dialogue in a event progression graph and integrating it into a topic associative network only upon semantic shifts, our approach minimizes interference while preserving long-term consistency. Additionally, we introduce a Graph-guided, Multi-factor Retrieval strategy to enhance context precision. Experiments on LoCoMo and LongDialQA benchmarks indicate that our method consistently outperforms state-of-the-art baselines in both reasoning accuracy and computational efficiency."
}