CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning
This addresses the challenge of post-verification in code generation for developers, offering a more efficient and effective solution, though it is incremental as it builds on existing RL methods with novel reward shaping.
The paper tackles the problem of inefficient and unreliable code verification for LLM-generated code by proposing CVeDRL, a reinforcement learning-based verifier that uses difficulty-aware rewards, achieving up to 28.97% higher pass rates and over 20x faster inference than GPT-3.5.
Code verifiers play a critical role in post-verification for LLM-based code generation, yet existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency. While reinforcement learning (RL) offers a promising alternative by optimizing models through execution-driven rewards without labeled supervision, our preliminary results show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples. We first theoretically analyze showing that branch coverage, sample difficulty, syntactic and functional correctness can be jointly modeled as RL rewards, where optimizing these signals can improve the reliability of unit-test-based verification. Guided by this analysis, we design syntax- and functionality-aware rewards and further propose branch- and sample-difficulty--aware RL using exponential reward shaping and static analysis metrics. With this formulation, CVeDRL achieves state-of-the-art performance with only 0.6B parameters, yielding up to 28.97% higher pass rate and 15.08% higher branch coverage than GPT-3.5, while delivering over $20\times$ faster inference than competitive baselines. Code is available at https://github.com/LIGHTCHASER1/CVeDRL.git