Metraipolis › archive › dc_b089c347b7fe

ActiveMem: Dynamic Latent Memory Trees for Long-Horizon Agents — arXiv:2609.33244

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33244
License (as archived)
CC0-1.0
Archived
2026-10-07T20:56:33.121Z
SHA-256 of the archived text
e0707c19c8a6d265e0315f615038f668eccfc13ba63156f44228b7ff16f399bd
Stable URL of this page
https://metraipolis.com/documento/dc_b089c347b7fe

How to cite

ActiveMem: Dynamic Latent Memory Trees for Long-Horizon Agents — arXiv:2609.33244 — arXiv (en), https://arxiv.org/abs/2609.33244 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b089c347b7fe (Metraipolis archive dc_b089c347b7fe, sha256 e0707c19c8a6d265e0315f615038f668eccfc13ba63156f44228b7ff16f399bd)
BibTeX
@misc{dcb089c347b7fe,
  title = {ActiveMem: Dynamic Latent Memory Trees for Long-Horizon Agents — arXiv:2609.33244},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33244},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b089c347b7fe; sha256 e0707c19c8a6d265e0315f615038f668eccfc13ba63156f44228b7ff16f399bd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33244. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Model (LLM) agents increasingly rely on external memory to support long-horizon reasoning and decision making. Existing memory systems typically retrieve historical trajectories or summaries as independent context fragments, overlooking the procedural dependencies underlying multi-step execution. As memory scales, such flat retrieval introduces context fragmentation and cross-task interference, leading to structurally inconsistent reasoning trajectories. We propose ActiveMem, a hierarchical memory framework that recursively organizes agent experiences into dependency-aware latent execution trees. ActiveMem abstracts trajectories into reusable subtask nodes while explicitly preserving execution transitions, enabling coherent reasoning-path retrieval conditioned on the current execution state. To support continual adaptation, ActiveMem further learns dynamic memory expansion, retrieval, and pruning policies through reinforcement learning. Experiments across various agent benchmarks demonstrate that ActiveMem consistently improves task completion, reasoning stability, and memory efficiency over existing memory-based agents. Moreover, ActiveMem enables compact open-weight models to achieve competitive performance with substantially larger proprietary systems.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.33244 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T20:29:01.179Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).