AI サービスが一時的に利用できないため、復旧後に翻訳を補完します。
[Submitted on 21 Oct 2026] Title:Edit, Don't Imitate: Minimal-Edit On-Policy Distillation for Multimodal Reasoning View a PDF of the paper titled Edit, Don't Imitate: Minimal-Edit On-Policy Distillation for Multimodal Reasoning, by Wen Ye and 10 other authors View PDF HTML (experimental) Abstract:Visual mathematical reasoning can fail before reasoning begins: when a vision-language model (VLM) misreads a diagram, subsequent reasoning proceeds from false premises. This perceptual bottleneck is especially pronounced in smaller models, while accurate textual descriptions can substantially narrow the gap to larger models. Because obtaining such descriptions at inference requires additional annotation or model calls, we propose minimal-edit corrective on-policy distillation. Using reference figure descriptions and verified reference solutions as training-only information, a teacher scores incorrect student rollouts under a minimal-edit objective, and only teacher-disfavored tokens contribute to the corrective loss. Correct rollouts receive task-reward reinforcement learning. This preserves on-policy supervision while avoiding full-response imitation. Across model scales, the method requires no descriptions at test time and meets or exceeds the corresponding base model supplied with a reference figure description at inference. It also improves perception-oriented performance, particularly low-level geometric perception, and transfers to external mathematics and perception benchmarks. Because the minimal-edit teacher evaluates corrections rather than generating its own solution, the method remains effective with a frozen same-size teacher, although its gains are smaller than with a larger teacher on most metrics. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2610.30231 [cs.LG] (or arXiv:2610.30231v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2610.30231 arXiv-issued DOI via DataCite (pending registration) Submission history From: Wen Ye [view email] [v1] Wed, 21 Oct 2026 17:59:28 UTC (1,384 KB) Full-text links: Access Paper: View a PDF of the paper titled Edit, Don't Imitate: Minimal-Edit On-Policy Distillation for Multimodal Reasoning, by Wen Ye and 10 other authors View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.LG new | recent | 2026-10 Change to browse by: cs cs.AI References & Citations SciX Google Scholar Semantic Scholar Loading... Data provided by: Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)