Metraipolis › archive › dc_c7dc01a52cc5
CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory — arXiv:2609.36935
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.36935
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:23:18.352Z
- SHA-256 of the archived text
c54721bff02b70d0964d3fac1c37bea12029a7cf246fd6d117a1325127b037cf- Stable URL of this page
https://metraipolis.com/documento/dc_c7dc01a52cc5
How to cite
CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory — arXiv:2609.36935 — arXiv (en), https://arxiv.org/abs/2609.36935 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c7dc01a52cc5 (Metraipolis archive dc_c7dc01a52cc5, sha256 c54721bff02b70d0964d3fac1c37bea12029a7cf246fd6d117a1325127b037cf)
BibTeX
@misc{dcc7dc01a52cc5,
title = {CoEM: Empowering Long-Context Reasoning with Commit-on-Evidence Memory — arXiv:2609.36935},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.36935},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c7dc01a52cc5; sha256 c54721bff02b70d0964d3fac1c37bea12029a7cf246fd6d117a1325127b037cf},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36935. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-context reasoning is essential for complex and long-horizon tasks, yet the performance of large language models (LLMs) degrades as context length increases. Recent approaches address this by processing input chunk by chunk while maintaining a bounded textual memory in model context. However, premature information compression can discard critical details essential for subsequent reasoning. In this paper, we introduce Commit-on-Evidence Memory (CoEM), which learns when to convert source evidence into compact memory facts. Specifically, under a fixed context-memory budget, CoEM preserves potentially useful source excerpts verbatim in a pending set, allowing subsequent context to clarify their relevance before irreversible compression. As new context arrives, a learned policy revisits each pending excerpt and decides whether to promote it to the committed memory, retain it for further consideration, or discard it. A frozen verifier ensures proposed facts are accepted only if supported by retained excerpts and current context. To further guide effective memory management, we train this policy using reinforcement learning by combining fine-grained, step-level evidence rewards with final answer rewards. Extensive experiments demonstrate that CoEM consistently improves long-context reasoning. When evaluated on 6,400 documents long-context input, CoEM outperforms the strongest memory baseline by 10.4-11.4 F1 points on Qwen3.5-9B. Code repository: this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.36935 — 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 2026-10-07T15:54:53.569Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).