#artificial intelligence
Apr 2026
RefineRL: Advancing Competitive Programming with Self-Refinement Reinforcement Learning
RefineRL introduces Skeptical-Agent, an iterative self-refinement agent equipped with local execution tools to validate generated solutions against public test cases of CP problems and highlights self-refinement as a promising direction for scaling LLMs' general reasoning ability.
Shaopeng Fu, Xingxing Zhang, Li Dong et al.
· arXiv.org · 1 citation