Abstract: To evaluate the repository-level code generation capabilities of Large Language Models (LLMs) in complex real-world software development scenarios, many evaluation methods have been ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results