SOURCE-LINKED INTELLIGENCE

Robust Code RL via Faulty-Code-Driven Test case Synthesis and Dense Reward Shaping

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Reinforcement Learning from Verifiable Rewards (RLVR) is pivotal for enhancing LLM code generation, yet its efficacy is often hindered by insufficient test case coverage, leading to reward hacking and policy degradation. To address this, we propose RobustTests, a framework featuring a faulty-code-driven test case synthesis strategy. By leveraging "near-correct" faulty codes, RobustTests captures latent logical discrepancies and employs validator agents with behavioral feature clustering to filter invalid or redundant test cases. Additionally, a stepwise dense reward function based on pass rate

Read original source ↗ Open in workspace

recordType

paper

region

Global