This work proposes Test Cases Scaling (TCS), a two-stage RL framework for effective test generation that improves both pass@1 and inference-time answer selection according to generated tests across TACO and LiveCodeBench.
Jiacheng Xu, Wen-Tao Zhang, Zhiyi Lyu et al.· 2 citations
Large language models (LLMs) have made substantial progress in code generation but still struggle with challenging programming tasks that require understanding rich natural language requirements. These requirements often specify problem goals, input/output formats, constraints, examples, and edge cases. Overlooking eve...
Yi-Xuan Li, Min Huang, Jia-Jing Wang et al.· 0 citations
Self-play methods that co-train a single language model as both coder and test author promise to move code-generation RL beyond fixed test suites, but they suffer from two coupled pathologies: permissiveness collapse, where pass-rate rewards are maximised by trivial, non-discriminative tests, and concentration bias, wh...
Large language models write code fluently, yet the programs they produce carry defect rates that ordinary functional evaluation seldom exposes, and detection systems that could catch these faults act too late to shape the generation that created them. This paper argues that generation and detection should be optimize...
Experimental results show that RL fine-tuning of Qwen3-32B via RobustTests achieves a 3% absolute gain on LiveCodeBench, demonstrating its effectiveness in advancing LLM code generation proficiency.
Yiwen Zhang, Xiaodong Yan, Zhen-Yu Huang et al.· 0 citations
This work evaluates Jev, a decision model that instead selects directly from a declared set of candidates and returns a probability for each, against six open-weight autoregressive models and a frontier proprietary model, and builds JevVibe, a diagnosis-guided repair agent that uses predicted CWE labels to repair code...
Arshak Rezvani, Sasha Behrouzi, Ahmad Sadeghi· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
