Metraipolis › archive › dc_ff4a53561d84

Route to Remember: Adaptive Memory Selection for LLM Agents — arXiv:2611.05636

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.05636
License (as archived)
CC0-1.0
Archived
2026-11-05T18:16:41.439Z
SHA-256 of the archived text
26752e704d79554da883ecdec8cb74056d2f38fc46c5ef91b4fcec88ab47ac1e
Stable URL of this page
https://metraipolis.com/documento/dc_ff4a53561d84

How to cite

Route to Remember: Adaptive Memory Selection for LLM Agents — arXiv:2611.05636 — arXiv (en), https://arxiv.org/abs/2611.05636 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff4a53561d84 (Metraipolis archive dc_ff4a53561d84, sha256 26752e704d79554da883ecdec8cb74056d2f38fc46c5ef91b4fcec88ab47ac1e)
BibTeX
@misc{dcff4a53561d84,
  title = {Route to Remember: Adaptive Memory Selection for LLM Agents — arXiv:2611.05636},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.05636},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff4a53561d84; sha256 26752e704d79554da883ecdec8cb74056d2f38fc46c5ef91b4fcec88ab47ac1e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05636. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The memory of production agents is increasingly becoming a bottleneck and is built almost exclusively from text: key points are extracted from out-of-limit context and retrieved into context at query time. Therefore, every new query incurs an additional cost that grows with the accumulated history. Weight memory is a complementary alternative that internalizes knowledge into model parameters and answers with no context at all, and has recently proven effective in self-evolving agents for internalizing skills and behaviors. However, its application in episodic and semantic memory, such as user-specific facts, remains comparatively unexplored, and a gap remains between earlier attempts at weight memory and text-based systems regarding recall performance. We study how to make weight memory competitive as a long-term factual store by applying on-policy self-distillation (OPSD). It trains a context-free student from a teacher, which is the same model conditioned on the user's memory, together with a verify-then-regenerate router that lets weight and text memory each cover the queries they handle best. On the HaluMem benchmark, OPSD matches or surpasses strong text-memory baselines, including full-context prompting and retrieval systems, while reading fewer tokens per query. We further conduct an extensive study that characterizes the behavior of weight memory and analyzes how different training recipes affect its performance. As agent harnesses move toward persistent, multi-session deployments, these results position weight memory as a practical, low-cost component of agent memory and provide insights into feasible training and evaluation protocols for combining it with retrieval.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.05636 — 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 2026-11-05T18:12:48.992Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).