Signals 4 · free daily AI digest

When Models Edit Too Much: On the Fidelity of Minimal Code Edits

Paper recorded by Signals 4 on 2026-09-03 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-03 on arXiv · recorded by Signals 4 on 2026-09-04

Category: cs.CL · 自然语言处理 · first seen 2026-09-04

Abstract

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

Read on arXiv →

#116 most recent of 186 cs.CL papers we have recorded · ↑ newer: CORE: Improving Compositional Reasoning in MLLM Embedding via Reranker · ↓ older: Translation as a Decision Space: A Multi-Agent Perspective on Low-Reso
Cite this page: When Models Edit Too Much: On the Fidelity of Minimal Code Edits: the #116 most recent of 186 cs.CL papers we have recorded (as of 2026-09-03). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/when-models-edit-too-much-on-the-fidelity-of-minimal-code-edits.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CL papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions