MineValiCoder: Reliable Code Generation with Test Case Quality Mining and Bipartite Graph-Based Mutual Validation

TL;DR - MineValiCoder is a closed-loop TDD framework that improves LLM code generation by filtering faulty generated tests and jointly validating code-test quality. It reports state-of-the-art performance across four coding benchmarks.

• Test Case Quality Mining uses self-validation to remove unreliable tests before refinement.
• Parallel TDD Refinement iteratively produces diverse code candidates using validated test feedback.
• Bipartite graph-based mutual validation models code-test interactions to select the most reliable solution.
• Reported Pass@1 reaches 96.34% on HumanEval, 87.40% on MBPP, 64.00% on APPS, and 51.33% on LiveCodeBench.