Metraipolis › archive › dc_f4be7983b175

Let the Fixer Decide: Training a Code Localizer from the Fixer's Likelihood — arXiv:2611.12164

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.12164
License (as archived)
CC0-1.0
Archived
2026-11-10T18:41:46.198Z
SHA-256 of the archived text
2c88fa6c278e41b5cb2c4a3f7409131557a5a52d69a6385ce5be479449f3f3b1
Stable URL of this page
https://metraipolis.com/documento/dc_f4be7983b175

How to cite

Let the Fixer Decide: Training a Code Localizer from the Fixer's Likelihood — arXiv:2611.12164 — arXiv (en), https://arxiv.org/abs/2611.12164 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f4be7983b175 (Metraipolis archive dc_f4be7983b175, sha256 2c88fa6c278e41b5cb2c4a3f7409131557a5a52d69a6385ce5be479449f3f3b1)
BibTeX
@misc{dcf4be7983b175,
  title = {Let the Fixer Decide: Training a Code Localizer from the Fixer's Likelihood — arXiv:2611.12164},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.12164},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f4be7983b175; sha256 2c88fa6c278e41b5cb2c4a3f7409131557a5a52d69a6385ce5be479449f3f3b1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.12164. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Code localizers are usually trained to predict the files a reference patch edits, but the code a fixer needs to read is not the same as the code it changes. We propose Fixer-Guided Localization (FGL): the localizer proposes several localizations, a frozen fixer scores each by how likely it makes a verified patch, and the localizer is trained on these scores with weighted supervised fine-tuning, without running tests and without reinforcement learning. We build the full pipeline on Gemma 4 E4B for the Kaggle Gemma-4 Developer Agent competition, from 1,629 test-verified Kimi K3 trajectories. The FGL target helps the fixer find the bug as often as the teacher's localization while using 27.5% less code. On the development set, one EM round moves the agent in the expected direction compared with a matched control: more tasks reach a patch (53 vs. 46), fewer tasks get stuck reading files that do not exist (43 vs. 54), and more tasks are solved (8 vs. 6). Training also changes how the agent works: trained agents explore about three times more than the base model and write far fewer wrong patches, and the single agent solves up to 11 of 78 tasks, against 8 for the base model. An analysis of 13 evaluation runs identifies what limits a 4B agent under an 8-minute budget (time, repetition loops and edit quality) and gives practical lessons for training small coding agents. All data, code and adapters are publicly available.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2611.12164 — 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 2026-11-10T18:37:37.541Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).