MIST-RL: Mutation-based Incremental Suite Testing via Reinforcement Learning
This addresses the problem of test redundancy and diminishing returns in fault detection for developers using LLMs for code generation, representing a novel method rather than an incremental improvement.
The paper tackles the problem of inefficient test generation for verifying LLM-generated code by proposing MIST-RL, which shifts from brute-force scaling to utility-based scaling, resulting in a +28.5% higher mutation score while reducing test cases by 19.3% and improving code reranking accuracy by 3.05%.
Large Language Models (LLMs) often fail to generate correct code on the first attempt, which requires using generated unit tests as verifiers to validate the solutions. Despite the success of recent verification methods, they remain constrained by a "scaling-by-quantity" paradigm. This brute-force approach suffers from a critical limitation: it yields diminishing returns in fault detection while causing severe test redundancy. To address this, we propose MIST-RL (Mutation-based Incremental Suite Testing via Reinforcement Learning), a framework that shifts the focus to "scaling-by-utility". We formulate test generation as a sequential decision process optimized via Group Relative Policy Optimization (GRPO). Specifically, we introduce a novel incremental mutation reward combined with dynamic penalties, which incentivizes the model to discover new faults while it suppresses functionally equivalent assertions. Experiments on HumanEval+ and MBPP+ demonstrate that MIST-RL outperforms state-of-the-art baselines. It achieves a +28.5% higher mutation score while reducing the number of test cases by 19.3%. Furthermore, we show that these compact, high-utility tests serve as superior verifiers, which improves downstream code reranking accuracy on HumanEval+ by 3.05% over the SOTA baseline with 10 candidate samples. The source code and data are provided in the supplementary material.