Newest papers

Co-Evolving Policy and Reward for Code Generation: A Mechanistic Study of Single-Adapter Drift

Amarsaikhan Batjargal, Noriko Tomuro · 2026 · arXiv

DOI: 10.48550/arXiv.2610.26416 · open-access PDF

Spanish

El aprendizaje reforzado con recompensas verificables (RLVR) es particularmente adecuado para la generación de código porque los programas generados pueden ser evaluados objetivamente a través de la ejecución. La ejecución también proporciona información de comportamiento que puede ser reutilizada como supervisión adicional. El Marco de Coevolución de Políticas sinérgicas y recompensas (SPARK), propuesto originalmente para tareas de razonamiento, explota esta idea reciclando los desarrollos de capacitación y sus resultados de evaluación para formar un modelo no solo para generar soluciones, sino también para evaluar y mejorarlas.

Machine translation (Meta M2M100); the English original follows.

Abstract

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

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: 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.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange