AI RESEARCH

Improving LLM Code Reasoning via Semantic Equivalence Self-Play with Formal Verification

arXiv CS.LG

ArXi:2604.17010v1 Announce Type: cross