Metraipolis › archive › dc_0d446fe1b89e
Context Language Models — arXiv:2609.37725
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37725
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:24:07.021Z
- SHA-256 of the archived text
385a5ea8827b1aeb5e52512f8fba16d607e6232cc594b396ff4b8b5ae4f27e50- Stable URL of this page
https://metraipolis.com/documento/dc_0d446fe1b89e
How to cite
Context Language Models — arXiv:2609.37725 — arXiv (en), https://arxiv.org/abs/2609.37725 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0d446fe1b89e (Metraipolis archive dc_0d446fe1b89e, sha256 385a5ea8827b1aeb5e52512f8fba16d607e6232cc594b396ff4b8b5ae4f27e50)
BibTeX
@misc{dc0d446fe1b89e,
title = {Context Language Models — arXiv:2609.37725},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37725},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0d446fe1b89e; sha256 385a5ea8827b1aeb5e52512f8fba16d607e6232cc594b396ff4b8b5ae4f27e50},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37725. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We introduce Context Language Models (CLMs), language models that natively manage their own context. We implement this by treating the context as a file and allowing the model to make unrestricted updates to this file. This allows the model to learn what is most important to maintain in context, and naturally extends to multi-agent systems where multiple agent contexts coexist as files. Building CLMs zero-shot with existing models outperforms SOTA context management strategies across a variety of tasks: 11.4% higher accuracy with 21.5% fewer FLOPs on BrowseComp-Plus, 5% higher scores with 59% fewer FLOPs on 12-hour EdgeBench, and 65% greater improvement with the same compute on a 24-hour multi-repository agent-swarm task. Moreover, by shifting context management from external harness control to intrinsic model behavior, CLMs naturally enable both in-context and parametric learning of context-management strategies. We show that CLMs can be steered with natural-language instructions evolved through a standard skill-optimization loop, improving held-out accuracy by up to 35.9 points on a context-management task while reducing compute. We also introduce an online reinforcement learning method for CLMs, improving Qwen3.5-9B performance on BrowseComp-Plus by 47.6% while using 12% fewer FLOPs. Finally, we co-design Suffix Cache Reuse for CLM serving, further reducing server-side compute by 35% relative to standard SGLang at matched performance. 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); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.37725 — 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 2026-10-07T15:54:53.652Z 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).