Metraipolis › archive › dc_943bf3ce0925

Hy-Memory-30B-A3B: Attributed Reinforcement Learning for Long-Term Memory Writing — arXiv:2610.22196

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.22196
License (as archived)
CC0-1.0
Archived
2026-10-16T10:02:02.714Z
SHA-256 of the archived text
1c392f59dbf0c1bcd59771832a2e2fbc6328d25d982dbf0c9c492ee3a5bcb923
Stable URL of this page
https://metraipolis.com/documento/dc_943bf3ce0925

How to cite

Hy-Memory-30B-A3B: Attributed Reinforcement Learning for Long-Term Memory Writing — arXiv:2610.22196 — arXiv (en), https://arxiv.org/abs/2610.22196 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_943bf3ce0925 (Metraipolis archive dc_943bf3ce0925, sha256 1c392f59dbf0c1bcd59771832a2e2fbc6328d25d982dbf0c9c492ee3a5bcb923)
BibTeX
@misc{dc943bf3ce0925,
  title = {Hy-Memory-30B-A3B: Attributed Reinforcement Learning for Long-Term Memory Writing — arXiv:2610.22196},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22196},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_943bf3ce0925; sha256 1c392f59dbf0c1bcd59771832a2e2fbc6328d25d982dbf0c9c492ee3a5bcb923},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22196. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

External-memory systems extend language-model applications beyond a context window, but their quality depends critically on the model executing the write path: deciding what to retain, how precisely to represent it, and how new evidence should update persistent state. Existing frameworks typically place a general-purpose large language model behind framework-specific prompts. This raises serving cost, and a writer specialized for one protocol does not transfer directly to another; a unified model that can operate across established memory frameworks remains largely unexplored. We present Hy-Memory-30B-A3B, a compact long-term-memory writer trained across Mem0, MemOS, and Hy-Memory. Memory On-Policy Distillation teaches framework-specific write protocols on student-generated states; Memory RL then optimizes downstream answerability using dense per-chunk probes, attributed cross-chunk rewards, over-storage controls, and chunk-level GRPO. Carrying policy-written banks across windows keeps training aligned with deployment. On fixed 20% benchmark subsamples, Hy-Memory-30B-A3B ranks first on all four evaluated harnesses, including held-out EverMemOS, with a 55.49% macro average versus 54.62% for Claude Opus 5.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.22196 — 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 2026-10-16T08:44:47.701Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).