通过GRPO实现多模态大语言模型推理的无监督后训练

Unsupervised Post-Training for Multi-Modal LLM Reasoning via GRPO

May 28, 2025

作者: Lai Wei, Yuting Li, Chen Wang, Yue Wang, Linghe Kong, Weiran Huang, Lichao Sun

cs.AI

摘要

提升多模态大语言模型（MLLMs）在训练后阶段的表现，通常依赖于监督微调（SFT）或强化学习（RL）。然而，这些监督方法需要昂贵且手动标注的多模态数据——这一资源最终难以持续。尽管近期研究探索了无监督的训练后优化，但其方法复杂且迭代困难。在本研究中，我们首次探讨了使用GRPO这一稳定且可扩展的在线RL算法，以实现无需外部监督的持续自我改进。我们提出了MM-UPT，一个简单而有效的框架，用于MLLMs的无监督训练后优化。MM-UPT基于GRPO构建，用基于多数投票的自我奖励机制替代了传统的奖励信号。实验表明，MM-UPT显著提升了Qwen2.5-VL-7B的推理能力（例如，在MathVista上从66.3%提升至72.9%，在We-Math上从62.9%提升至68.7%），且仅使用了无标注的标准数据集。MM-UPT不仅超越了先前的无监督基线，甚至接近了监督GRPO的效果。此外，我们发现，仅由MLLM自身生成的合成问题也能进一步提升性能，这为可扩展的自我改进指明了一条有前景的路径。总体而言，MM-UPT为在缺乏外部监督的情况下，实现MLLMs的持续自主优化提供了新范式。我们的代码已发布于https://github.com/waltonfuture/MM-UPT。

English

Improving Multi-modal Large Language Models (MLLMs) in the post-training stage typically relies on supervised fine-tuning (SFT) or reinforcement learning (RL). However, these supervised methods require expensive and manually annotated multi-modal data--an ultimately unsustainable resource. While recent efforts have explored unsupervised post-training, their methods are complex and difficult to iterate. In this work, we are the first to investigate the use of GRPO, a stable and scalable online RL algorithm, for enabling continual self-improvement without any external supervision. We propose MM-UPT, a simple yet effective framework for unsupervised post-training of MLLMs. MM-UPT builds upon GRPO, replacing traditional reward signals with a self-rewarding mechanism based on majority voting over multiple sampled responses. Our experiments demonstrate that MM-UPT significantly improves the reasoning ability of Qwen2.5-VL-7B (e.g., 66.3 %rightarrow72.9 % on MathVista, 62.9 %rightarrow68.7 % on We-Math), using standard dataset without ground truth labels. MM-UPT also outperforms prior unsupervised baselines and even approaches the results of supervised GRPO. Furthermore, we show that incorporating synthetic questions, generated solely by MLLM itself, can boost performance as well, highlighting a promising approach for scalable self-improvement. Overall, MM-UPT offers a new paradigm for continual, autonomous enhancement of MLLMs in the absence of external supervision. Our code is available at https://github.com/waltonfuture/MM-UPT.