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

• URL: http://arxiv.org/abs/2609.04061v1
• Date: Thu, 03 Sep 2026 16:36:05 GMT
• Title: When Models Edit Too Much: On the Fidelity of Minimal Code Edits
• Abstract summary: Large language models (LLMs) are increasingly used to edit existing code, but correctness alone is not enough.<n>We study over-editing, the tendency of a model to rewrite code beyond what is required to fix a bug.<n>We construct an evaluation framework from 400 BigCodeBench problems by injecting controlled AST-level corruptions into reference solutions.
• Score: 21.17848461238914
• License: http://creativecommons.org/licenses/by/4.0/
• 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 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 coexist with unnecessarily large edits and added cognitive complexity. A preservation instruction substantially reduces this behavior, lowering average excess Levenshtein distance from 0.195 to 0.131, reducing added cognitive complexity by 26.6%, and increasing Pass@1 by 2.3 points. However, these gains do not simply follow from a larger reasoning budget or larger models. We next ask whether minimal editing can be learned directly during post-training. We observe that supervised fine-tuning overfits to seen corruption patterns, whereas reinforcement learning gives the best out-of-domain edit-fidelity and performance-retention trade-off. These results position edit fidelity as a distinct axis of code-repair quality and show that it can be measured and learned.

Related papers

• REVES: REvision and VErification--Augmented Training for Test-Time Scaling [53.197756110943395]
  We propose a two-stage iterative framework that alternates between online data/prompt augmentation and policy optimization.<n>We observe gains of +6.5 points over the RL baseline and +4.0 points over standard multi-turn training.
  arXiv  Detail & Related papers  (2026-06-17T10:37:23Z)
• Imperfect Visual Verification for Code Edition : A Case Study on TikZ [3.6213861298681826]
  We use TikZ as a focused case study that isolates the core difficulties of the problem.<n>We define visual code customization as an iterative editing problem with an imperfect oracle.<n>Our findings show that even imperfect verifiers can determine with moderate accuracy whether visual instructions are applied to code, achieving F1-scores up to 0.815.
  arXiv  Detail & Related papers  (2026-04-09T12:21:52Z)
• QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization [50.722636027998]
  Large Language Models (LLMs) achieve strong program repair performance but often suffer from over-editing.<n>We propose PRepair, a framework that mitigates over-editing and improves repair accuracy.
  arXiv  Detail & Related papers  (2026-04-07T14:56:38Z)
• Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updates [69.6610686845008]
  CoRSA is a parameter-efficient, holistic approach for knowledge editing with multiple updates.<n>It tackles multiple challenges simultaneously: it improves generalization to different input forms and enhances stability across multiple updates.<n>CoRSA also generalizes to the code domain, outperforming the strongest baseline by 5.48% Pass@5 in update efficacy.
  arXiv  Detail & Related papers  (2026-02-03T16:18:06Z)
• Understanding Robustness of Model Editing in Code LLMs: An Empirical Study [1.5624785508022727]
  We present a systematic study of five state-of-the-art model editing methods.<n>We apply these methods to three leading open-source code LLMs, CodeLlama, CodeQwen1.5, and DeepSeek-Coder.<n>Instant edits consistently degrade model performance, with syntactic validity dropping by up to 86 percentage points and functional correctness declining by 45 points even in the best-performing setting.
  arXiv  Detail & Related papers  (2025-11-05T04:58:13Z)
• MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs [76.28901550926021]
  Existing methods for lifelong model editing compromise generalization, interfere with past edits, or fail to scale to long editing sequences.<n>We propose MEMOIR, a novel scalable framework that injects knowledge through a residual memory, while preserving the core capabilities of the pre-trained model.<n>MeMOIR achieves state-of-the-art performance across reliability, generalization, and locality metrics, scaling to thousands of sequential edits with minimal forgetting.
  arXiv  Detail & Related papers  (2025-06-09T16:16:42Z)
• The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse [58.0132400208411]
  Even a single edit can trigger model collapse, manifesting as significant performance degradation in various benchmark tasks. benchmarking Large Language Models after each edit is impractically time-consuming and resource-intensive. We have utilized GPT-3.5 to develop a new dataset, HardEdit, based on hard cases.
  arXiv  Detail & Related papers  (2024-02-15T01:50:38Z)
• Converge to the Truth: Factual Error Correction via Iterative Constrained Editing [30.740281040892086]
  We propose VENCE, a novel method for factual error correction (FEC) with minimal edits. VENCE formulates the FEC problem as iterative sampling editing actions with respect to a target density function. Experiments on a public dataset show that VENCE improves the well-adopted SARI metric by 5.3 (or a relative improvement of 11.8%) over the previous best distantly-supervised methods.
  arXiv  Detail & Related papers  (2022-11-22T10:03:13Z)
• Memory-Based Model Editing at Scale [102.28475739907498]
  Existing model editors struggle to accurately model an edit's intended scope. We propose Semi-Parametric Editing with a Retrieval-Augmented Counterfactual Model (SERAC) SERAC stores edits in an explicit memory and learns to reason over them to modulate the base model's predictions as needed.
  arXiv  Detail & Related papers  (2022-06-13T23:40:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.