无需标注数据，让多模态模型通过自我反馈持续提升推理能力。
EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewards
- 用同一模型生成问题并自解，通过内部一致性实现持续自奖励。
- 在多个多模态数学推理任务上提升约3%，仅用原始图像训练。
- 适合想研究无监督自进化模型的研究者，方法简洁易复现。
近期大型多模态模型（LMMs）在推理与感知能力上取得显著进展，但多数训练流程仍依赖人工标注数据或外部验证的奖励模型，限制了其自主性与可扩展性。本文提出一种名为EvoLMM的自进化框架，从单一主干模型中构建两个协作智能体：生成多样化、图像相关的提问的提案者（Proposer），以及通过内部一致性求解问题的求解者（Solver）。学习过程基于连续自奖励机制，无需真实答案或人类判断，推动问题生成与结构化推理的协同优化。以Qwen2.5-VL为基座模型，在仅使用原始训练图像的前提下，EvoLMM在ChartQA、MathVista和MathVision等多模态数学推理基准上实现约3%的稳定提升。我们希望该简单而有效的方法能成为未来全无监督自进化多模态模型研究的坚实基线。代码与模型已开源。
原文摘要 · Abstract (English)
Recent advances in large multimodal models (LMMs) have enabled impressive reasoning and perception abilities, yet most existing training pipelines still depend on human-curated data or externally verified reward models, limiting their autonomy and scalability. In this work, we strive to improve LMM reasoning capabilities in a purely unsupervised fashion (without any annotated data or reward distillation). To this end, we propose a self-evolving framework, named EvoLMM, that instantiates two cooperative agents from a single backbone model: a Proposer, which generates diverse, image-grounded questions, and a Solver, which solves them through internal consistency, where learning proceeds through a continuous self-rewarding process. This dynamic feedback encourages both the generation of informative queries and the refinement of structured reasoning without relying on ground-truth or human judgments. When using the popular Qwen2.5-VL as the base model, our EvoLMM yields consistent gains upto $\sim$3\% on multimodal math-reasoning benchmarks, including ChartQA, MathVista, and MathVision, using only raw training images. We hope our simple yet effective approach will serve as a solid baseline easing future research in self-improving LMMs in a fully-unsupervised fashion. Our code and models are available at https://github.com/mbzuai-oryx/EvoLMM.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。