1
SWE-IF: Aligning Code Evaluation with Human Preference
超越功能正确性，提出对齐人类偏好的代码评估框架SWE-IF，让AI生成的代码通过“感觉”检验
arXiv:2510.07315v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have catalyzed vibe coding, where users leverage LLMs to genera…