一作ACL 2026Findings· Findings of ACL
C ode RM - NT : Reward Model for Code RL without Unit Tests
Xiao Xia · Dan Zhang · Tianrui Sun
摘要
Providing accurate reward signals for code generated by large language models (LLMs) is a significant challenge in applying reinforcement learning (RL) to code generation. Existing methods rely on unit tests to evaluate code correctness and provide rewards, which are hindered by the difficulty of...