Amarsaikhan Batjargal, Noriko Tomuro · 2026 · arXiv
DOI: 10.48550/arXiv.2610.26416 · open-access PDF
O Reinforcement Learning with Verifiable Rewards (RLVR) é particularmente adequado para a geração de código porque os programas gerados podem ser avaliados objetivamente através da execução. A execução também fornece informações comportamentais que podem ser reutilizadas como supervisão adicional. O Synergistic Policy and Reward Co-Evolving Framework (SPARK), originalmente proposto para tarefas de raciocínio, explora esta idéia por reciclar os lançamentos de treinamento e seus resultados de avaliação para treinar um modelo não apenas para gerar soluções, mas também para avaliá-las e melhorá-las.
Machine translation (Meta M2M100); the English original follows.
Reinforcement learning with verifiable rewards (RLVR) is particularly well suited to code generation because generated programs can be objectively evaluated through execution. Execution also provides behavioral information that can be reused as additional supervision. The Synergistic Policy and Reward Co-Evolving Framework (SPARK), originally proposed for reasoning tasks, exploits this idea by recycling training rollouts and their evaluation results to train a model not only to generate solutions but also to evaluate and improve them. We investigate whether this policy-reward co-evolution appr
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26416, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange