Two-Stage Reinforcement Learning for Sound and Adversarial Test Generation in Code LLMs

September 03, 2026 ยท Grace Period ยท ๐Ÿ› EMNLP 2026

โณ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Jiacheng Xu, Wentao Zhang, Zhiyi Lyu, Fuxiang Zhang, Chaojie Wang, Yang Liu, Bo An arXiv ID 2609.03955 Category cs.CL: Computation & Language Cross-listed cs.LG Citations 0 Venue EMNLP 2026
Abstract
Reinforcement learning (RL) has substantially advanced code generation with large language models (LLMs) through executable feedback. The feedback for coding problems mainly comes from specific test cases, where high-quality test cases are often scarce since they should be both sound and discriminative. We thus turn to study the auto-generation of test cases using the learned model. We find this is naturally an adversarial RL problem: the model is expected to generate effective test cases as counterexamples, depending on the solver's current failure modes. We propose Test Cases Scaling (TCS), a two-stage RL framework for effective test generation. Both stages train a test generator from a rolling policy-aligned buffer: Stage 1 generates tests consistent with the reference solution, and Stage 2 restricts the buffer to current failure modes and learns counterexample tests. Across TACO and LiveCodeBench, TCS improves both pass@1 and inference-time answer selection according to generated tests. We find the learned test generator also enables effective selection among other LLM outputs.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Computation & Language

๐ŸŒ… ๐ŸŒ… Old Age

Attention Is All You Need

Ashish Vaswani, Noam Shazeer, ... (+6 more)

cs.CL ๐Ÿ› NeurIPS ๐Ÿ“š 166.0K cites 9 years ago