TengyueWang TengyueWang

CODEMENV: Benchmarking Large Language Models on Code Migration

Large language models (LLMs) have demonstrated remarkable proficiency in handling a wide range of tasks within the software engineering domain, but their ability to perform code migration—adapting code to different environments—remains underexplored. In this work, we propose a novel benchmark, : Cod…