@inproceedings{li-etal-2026-timem,
    title = "{T}i{M}em: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents",
    author = "Li, Kai  and
      Yu, Xuanqing  and
      Ni, Ziyi  and
      Zeng, Yi  and
      Xu, Yao  and
      Zhang, Zheqing  and
      Li, Xin  and
      Sang, Jitao  and
      Duan, Xiaogang  and
      Wang, Xuelei  and
      Liu, Chengbao  and
      Tan, Jie",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1091/",
    pages = "21700--21720",
    ISBN = "979-8-89176-395-1",
    abstract = "Long-horizon conversational agents have to manage ever-growing interaction histories that quickly exceed the finite context windows of large language models (LLMs). Existing memory frameworks provide limited support for temporally structured information across hierarchical levels, often leading to fragmented memories and unstable long-horizon personalization. We present TiMem, a temporal{--}hierarchical memory framework that organizes conversations through a Temporal Memory Tree (TMT), enabling systematic memory consolidation from raw conversational observations to progressively abstracted persona representations. TiMem is characterized by three core properties: (1) temporal{--}hierarchical organization through TMT; (2) semantic-guided consolidation that enables memory integration across hierarchical levels without fine-tuning; and (3) complexity-aware memory recall that balances precision and efficiency across queries of varying complexity. Under a consistent evaluation setup, TiMem achieves state-of-the-art accuracy on both benchmarks, reaching 75.30{\%} on LoCoMo and 76.88{\%} on LongMemEval-S. It outperforms all evaluated baselines while reducing the recalled memory length by 52.20{\%} on LoCoMo. Manifold analysis indicates clear persona separation on LoCoMo and reduced dispersion on LongMemEval-S. Overall, TiMem treats temporal continuity as a first-class organizing principle for long-horizon memory in conversational agents. The code is available at https://github.com/TiMEM-AI/timem."
}