Metraipolis › archive › dc_2ed2eaf8f7e3

SkillMem: Compact External Skill Memory for Latent Multi-Agent Reasoning — arXiv:2610.37935

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.37935
License (as archived)
CC0-1.0
Archived
2026-10-31T06:08:42.078Z
SHA-256 of the archived text
9fbfa180dd2af2c326b048e9cfb8d910baa9249f2061618374e70de31b64e54a
Stable URL of this page
https://metraipolis.com/documento/dc_2ed2eaf8f7e3

How to cite

SkillMem: Compact External Skill Memory for Latent Multi-Agent Reasoning — arXiv:2610.37935 — arXiv (en), https://arxiv.org/abs/2610.37935 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2ed2eaf8f7e3 (Metraipolis archive dc_2ed2eaf8f7e3, sha256 9fbfa180dd2af2c326b048e9cfb8d910baa9249f2061618374e70de31b64e54a)
BibTeX
@misc{dc2ed2eaf8f7e3,
  title = {SkillMem: Compact External Skill Memory for Latent Multi-Agent Reasoning — arXiv:2610.37935},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.37935},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2ed2eaf8f7e3; sha256 9fbfa180dd2af2c326b048e9cfb8d910baa9249f2061618374e70de31b64e54a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.37935. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Latent multi-agent systems reduce inference cost by keeping intermediate reasoning and inter-agent communication in continuous representations.\ Reusable skills, however, are commonly supplied as text or stored as layer-wise Key-Value states or skill-specific model parameters, introducing repeated processing or substantial per-skill storage.\ We introduce \textbf{SkillMem}, which represents each reusable skill as a small external latent memory and makes it directly available during latent multi-agent reasoning.\ A role-conditioned cross-attention reader allows the same memory to guide different agents according to their roles and current hidden states.\ We train the skill memories and shared reader through behavioral distillation, matching the change in token predictions induced by a textual skill while keeping the backbone frozen.\ For our experiments, we obtain textual skills from paired successful and failed multi-agent trajectories, although existing skills can also be compiled directly.\ Across six benchmarks and Qwen3 models ranging from 4B to 14B parameters, SkillMem provides strong gains for smaller backbones while preserving efficient latent multi-agent inference across model scales.\ We use the same $8\times256$ memory configuration across all evaluated model scales, requiring 9.6\,KB of skill-specific storage per skill, including its retrieval embedding.\ For a library of 100 skills, SkillMem reduces total storage by more than $38\times$ compared with alternative skill representations and achieves up to a $4.23\times$ end-to-end inference speedup.\ These results show that compact procedural memories can be reused without sacrificing the efficiency of latent multi-agent reasoning.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.37935 — 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 2026-10-31T06:07:08.444Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).