Metraipolis › archive › dc_681f293bfde8

When Models Edit Too Much: On the Fidelity of Minimal Code Edits — arXiv:2609.04061

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.04061
License (as archived)
CC0-1.0
Archived
2026-10-17T01:12:15.287Z
SHA-256 of the archived text
a7b4fd447351ce1d1ceaf5dc520dc93d37086112d4bee3a17499288c0710d5b5
Stable URL of this page
https://metraipolis.com/documento/dc_681f293bfde8

How to cite

When Models Edit Too Much: On the Fidelity of Minimal Code Edits — arXiv:2609.04061 — arXiv (en), https://arxiv.org/abs/2609.04061 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_681f293bfde8 (Metraipolis archive dc_681f293bfde8, sha256 a7b4fd447351ce1d1ceaf5dc520dc93d37086112d4bee3a17499288c0710d5b5)
BibTeX
@misc{dc681f293bfde8,
  title = {When Models Edit Too Much: On the Fidelity of Minimal Code Edits — arXiv:2609.04061},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04061},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_681f293bfde8; sha256 a7b4fd447351ce1d1ceaf5dc520dc93d37086112d4bee3a17499288c0710d5b5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04061. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI).
Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.04061 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T00:28:03.026Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).