When Models Edit Too Much: On the Fidelity of Minimal Code Edits
- 类型：arxiv
- 标识：2609.04061
- 链接：https://arxiv.org/abs/2609.04061
- 主分类：evaluation
- 形态：method
- TLDR：Large language models (LLMs) are increasingly used to edit existing code, but correctness alone is not enough: useful repairs should also be minimal, reviewable, and faithful to the original implementation. We study over-editing, the tendency of a model to rewrite code beyond what is required to fix a bug. We construct an evaluation framework from 400 BigCodeBench problems by injecting controlled AST-level corruptions into reference solutions, giving each repair task a known minimal patch. Across frontier LLMs, over-editing is widespread even among strong models like GPT-5.5: high Pass@1 can c
- 待LLM分类：否
- 来源文件：
- /inbox/tom/_candidates/2026-09-07-agent-rag-longcontext-candidates.json