Metraipolis › archive › dc_49250b06e477
RIPPLE: Recovery Internalization through Process-Level Policy Learning and Effect-Aware Credit for Long-Horizon Agentic Coding — arXiv:2610.31276
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.31276
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-25T08:33:30.931Z
- SHA-256 of the archived text
712952ca86cd0a83c8bc231ed93b3bea627680284b976f1a0c0c6415e71afd7d- Stable URL of this page
https://metraipolis.com/documento/dc_49250b06e477
How to cite
RIPPLE: Recovery Internalization through Process-Level Policy Learning and Effect-Aware Credit for Long-Horizon Agentic Coding — arXiv:2610.31276 — arXiv (en), https://arxiv.org/abs/2610.31276 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_49250b06e477 (Metraipolis archive dc_49250b06e477, sha256 712952ca86cd0a83c8bc231ed93b3bea627680284b976f1a0c0c6415e71afd7d)
BibTeX
@misc{dc49250b06e477,
title = {RIPPLE: Recovery Internalization through Process-Level Policy Learning and Effect-Aware Credit for Long-Horizon Agentic Coding — arXiv:2610.31276},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.31276},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_49250b06e477; sha256 712952ca86cd0a83c8bc231ed93b3bea627680284b976f1a0c0c6415e71afd7d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31276. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-horizon coding agents can observe a failure and generate a plausible critique yet continue along the same incorrect trajectory. We call this the Critique-to-Action Handoff Gap. We introduce RIPPLE, a recovery-oriented training framework that internalizes process-level guidance into a single self-recovering policy. A prefix-only teacher provides evidence-grounded recovery supervision without access to future actions or terminal outcomes. After lightweight distillation, reinforcement learning assigns stage-specific credit to reflection triggering, diagnosis, and corrective action. Local validity signals gate trajectory advantages so that intermediate recovery decisions do not indiscriminately inherit terminal outcomes. Across four repository-level software-engineering benchmarks, RIPPLE improves the base policy by an arithmetic mean of 4.8 score points, with no auxiliary critic at deployment. Controlled ablations isolate the benefits of local, stage-aligned credit, and matched-prefix interventions show that critique semantics affect subsequent correction. Recovery gains concentrate where the prefix contains sufficient diagnostic evidence and the repository state remains reversible within the remaining budget. 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/2610.31276 — 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 2026-10-25T08:31:13.081Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).