Metraipolis › archive › dc_f9290bbf2f23
Terminal-RSI: Recursive Self-Improving Terminal Agents via Recursive Self-Rewrite and Policy-Refine — arXiv:2610.22192
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22192
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T09:05:36.201Z
- SHA-256 of the archived text
71997fc3db42608003579aacdf6117e7bcd09cfab1e8f75ba259abe3a87dd7a3- Stable URL of this page
https://metraipolis.com/documento/dc_f9290bbf2f23
How to cite
Terminal-RSI: Recursive Self-Improving Terminal Agents via Recursive Self-Rewrite and Policy-Refine — arXiv:2610.22192 — arXiv (en), https://arxiv.org/abs/2610.22192 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f9290bbf2f23 (Metraipolis archive dc_f9290bbf2f23, sha256 71997fc3db42608003579aacdf6117e7bcd09cfab1e8f75ba259abe3a87dd7a3)
BibTeX
@misc{dcf9290bbf2f23,
title = {Terminal-RSI: Recursive Self-Improving Terminal Agents via Recursive Self-Rewrite and Policy-Refine — arXiv:2610.22192},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22192},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f9290bbf2f23; sha256 71997fc3db42608003579aacdf6117e7bcd09cfab1e8f75ba259abe3a87dd7a3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22192. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Converting verifier outcomes into useful policy updates and partial progress into complete task resolution is crucial for terminal agents. We present Terminal-RSI, which alternates Recursive Self-Rewrite (RSR) of verified trajectories with critic-inheriting Recursive Policy-Refine (RPR). We formulate Turn-Boundary Skip-Observation GAE (TB-GAE) to retain token-specific baselines while applying decay across interaction turns. Family Dense Reward (FDR) calibrates semantic progress across task families; Success-Prioritized Reward (SPR) separates strict success from bounded partial credit. A Warm-up Critic Model, recursive critic handoff, and stage-specific learning rates coordinate the two refinement phases. The five-stage sequence from Qwen3.8-27B base model through RSR-1, RPR-1, RSR-2, and RPR-2 yields Terminal-RSI-27B: 19.84% on Terminal-Bench 4.0, 86.52% on Terminal-Bench 2.1, 16.95% on Terminal-Bench 3.0, and 79.00% on Terminal-Bench Hard. On LHTB, process reward rises from 0.21 to 0.37 and strict completion from 0/46 to 5/46; SWR100 success doubles from 3.00% to 6.00%. By recursively coupling verified trajectory rewriting with critic-inheriting policy refinement, each retained stage consolidates discovered behaviors and progressively advances the agent from partial local repairs to strict terminal verification. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.22192 — 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 2026-10-16T08:44:17.821Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).