Metraipolis › archive › dc_76aec6c7d581
Madeleine: Learning Involuntary Recall for Conversational Memory from Simulated Lives — arXiv:2610.01118
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.01118
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:12:48.850Z
- SHA-256 of the archived text
11eff4d6af22e2a53ee956fd0504d05e07a49fa100fbc2366c85c5a25913bc10- Stable URL of this page
https://metraipolis.com/documento/dc_76aec6c7d581
How to cite
Madeleine: Learning Involuntary Recall for Conversational Memory from Simulated Lives — arXiv:2610.01118 — arXiv (en), https://arxiv.org/abs/2610.01118 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_76aec6c7d581 (Metraipolis archive dc_76aec6c7d581, sha256 11eff4d6af22e2a53ee956fd0504d05e07a49fa100fbc2366c85c5a25913bc10)
BibTeX
@misc{dc76aec6c7d581,
title = {Madeleine: Learning Involuntary Recall for Conversational Memory from Simulated Lives — arXiv:2610.01118},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.01118},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_76aec6c7d581; sha256 11eff4d6af22e2a53ee956fd0504d05e07a49fa100fbc2366c85c5a25913bc10},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01118. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A long-term conversational assistant must recall the right memory at the right moment, yet the memory that matters most is often not similar to what the user says now. Current systems recover such associations by letting an LLM reason at write or read time, at a cost of hundreds to over a thousand LLM calls per memory bank and up to several thousand context tokens per query. We argue that association is a learnable relevance: the pointwise mutual information of memories under how human lives unfold. We introduce Madeleine, which learns amortized association: offline, an LLM life simulator writes simulated lives, whose cue-trigger pairs teach a query encoder a residual association on top of frozen similarity; online, it calls no LLM and plugs into any vector memory by replacing only the query encoder. On LoCoMo-Plus under the official protocol, Madeleine (I) reaches 66.6 when plugged into HyperMem, the highest among all systems evaluated under this protocol; (II) used alone, reaches the score of HyperMem as released (52.4 vs. 52.9) with zero LLM calls and about 1/21 of its answer context; and (III) lifts T-Mem by 26.2 points, significantly outperforms the same untrained backbone inside both systems, and leaves ordinary QA intact on the 4B backbone. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Information Retrieval (cs.IR). Pagina del lavoro: https://arxiv.org/abs/2610.01118 — 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 2026-10-06T10:21:28.155Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).