Abstract:Vision-language models (VLMs) have achieved strong multimodal reasoning capabilities, but further improving them still relies heavily on large-scale human-constructed supervision for post-training. Such supervision is costly to obtain, especially for reasoning-intensive multimodal tasks where questions, answers, and feedback signals must be carefully designed. This motivates self-evolving learning, where a model improves itself through a dual-role closed loop: a questioner autonomously poses questions and a solver learns to solve them. However, we observe that current VLM self-evolving methods still face three major challenges: coarse-grained role alternation delays the interaction between question generation and solver adaptation; generated questions can progressively degrade in quality; and question types may collapse toward a narrow distribution. These issues limit the efficiency and reliability of self-evolution. Thus, we propose \textbf{RISE}, a reliable self-evolving framework for vision-language models. RISE is built on three complementary designs: fine-grained role alternation, which shortens the feedback loop between the questioner and the solver to improve efficiency; a quality supervisor, which improves question validity and pseudo-label reliability; and skill-aware dynamic balancing, which mitigates mode collapse and maintains broad skill coverage during evolution. Together, these components enable more reliable and effective self-evolution from unlabeled images. Experiments on two VLM backbones across seven benchmarks show that RISE consistently improves the base models, yielding broad and sustained gains. Our code is publicly available at this https URL.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2605.20914 [cs.CV] |
| (or arXiv:2605.20914v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.20914 arXiv-issued DOI via DataCite |
From: Chaoran Xu [view email]
[v1]
Wed, 20 May 2026 08:57:57 UTC (21,711 KB)
[v2]
Tue, 26 May 2026 05:21:21 UTC (21,712 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。