微软论文提出 CABRA：编码智能体的瓶颈在理解代码而非编辑代码
微软研究人员提出 CABRA，可生成合成编码任务并每次只提升一种难度，对 8 个 LLM 和 6 个智能体在 6,840 个任务上运行，并将每次工具调用标注为阅读、分析、搜索、编辑或测试。
New Microsoft paper finds that coding agents trip up when they have to understand a lot of code, not when they have to edit a lot of it, so test them on reading and comparing code instead of diff size.
Microsoft researchers built CABRA, which generates synthetic coding tasks and raises 1 kind of difficulty at a time. They ran 8 LLMs and 6 agents on 6,840 tasks and labeled each tool call as reading, analyzing, searching, editing, or testing.
Plain LLMs got worse as tasks grew, but agents stayed near-perfect by using tools like grep. On SWE-bench Verified, the count of reading and analysis calls tracked agent failures better than lines edited, with correlations of -0.200 versus -0.159.
来源：rohanpaul_ai · x.com