Metraipolis › archive › dc_7b595ce6db97

Certified Long-Horizon Code Agent Evolution via Validation-Gated Skill Optimization — arXiv:2609.32990

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32990
License (as archived)
CC0-1.0
Archived
2026-10-07T20:56:19.809Z
SHA-256 of the archived text
7433cd6b10d28606d0d39467ae2de5a4c647d8dce909d8d632d97ae5cc334ecc
Stable URL of this page
https://metraipolis.com/documento/dc_7b595ce6db97

How to cite

Certified Long-Horizon Code Agent Evolution via Validation-Gated Skill Optimization — arXiv:2609.32990 — arXiv (en), https://arxiv.org/abs/2609.32990 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7b595ce6db97 (Metraipolis archive dc_7b595ce6db97, sha256 7433cd6b10d28606d0d39467ae2de5a4c647d8dce909d8d632d97ae5cc334ecc)
BibTeX
@misc{dc7b595ce6db97,
  title = {Certified Long-Horizon Code Agent Evolution via Validation-Gated Skill Optimization — arXiv:2609.32990},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32990},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7b595ce6db97; sha256 7433cd6b10d28606d0d39467ae2de5a4c647d8dce909d8d632d97ae5cc334ecc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32990. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long horizon agent self-evolution without model weight updates is essential for enabling deployed agents to accumulate reusable skills and improve over time. Prior self-evolution work has focused primarily on short-horizon tasks, while repository-level software engineering remains unexplored despite being an ideal testbed for long-horizon adaptation. In this setting, agents are required to solve streams of sequential tasks, navigate complex dependencies with evolving repositories and persistently store and reuse experience. Text-based skill optimization offers an efficient, non-parametric approach for such adaptation. However, existing methods often suffer from unstable updates, performance drawdown, and agent collapse over extended deployments. In this paper, we formalize the concept of in-context self-evolution and introduce VALVE, a validated-gated framework for long-horizon skill optimization. We establish finite convergence, provide theoretical guarantees for future-task gain and drawdown, and derive the validation and evaluation holdout sizes required for a prescribed tolerance, with leading-order scaling Empirically, our pipeline, VALVE achieves stable self-improvement over evolution horizon spanning more than 1,000 SWE tasks, with average final and peak gains of $14.9$ and $16.5$ points across three frontier models (GPT-5.5, Claude-4.6 and MiniMax-M2.7). The validation gate reduces average drawdown by 75% and produces an 11x more compact skill bank than ungated evolution. We further present extensive ablations identifying the design choices most critical to long-horizon skill evolution.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.32990 — 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 2026-10-07T20:29:01.122Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).