统一奖励模型：面向多模态理解与生成

Unified Reward Model for Multimodal Understanding and Generation

March 7, 2025

作者: Yibin Wang, Yuhang Zang, Hao Li, Cheng Jin, Jiaqi Wang

cs.AI

摘要

人类偏好对齐领域的最新进展显著提升了多模态生成与理解能力。其中，训练奖励模型以指导偏好优化是关键方法。然而，现有模型往往局限于特定任务，限制了其在多样化视觉应用中的适应性。我们认为，联合学习评估多项任务可能产生协同效应：图像理解的提升有助于图像生成评估的改进，而精细化的图像评估则通过更优的帧分析促进视频评估。为此，本文提出了UnifiedReward，首个用于多模态理解与生成评估的统一奖励模型，支持成对排序与逐点评分，可应用于视觉模型的偏好对齐。具体而言：（1）我们首先在构建的大规模人类偏好数据集上开发UnifiedReward，涵盖图像与视频的生成/理解任务；（2）随后，基于视觉模型自动构建高质量偏好对数据，通过成对排序与逐点筛选逐步精炼其输出；（3）最后，利用这些数据通过直接偏好优化（DPO）进行偏好对齐。实验结果表明，联合学习评估多样视觉任务能带来显著的相互增益，我们将此流程应用于图像与视频的理解/生成任务，显著提升了各领域的性能。

English

Recent advances in human preference alignment have significantly enhanced multimodal generation and understanding. A key approach is training reward models to guide preference optimization. However, existing models are often task-specific, limiting their adaptability across diverse visual applications. We also argue that jointly learning to assess multiple tasks may foster a synergistic effect, where improved image understanding enhances image generation assessment, and refined image evaluation benefits video assessment through better frame analysis. To this end, this paper proposes UnifiedReward, the first unified reward model for multimodal understanding and generation assessment, enabling both pairwise ranking and pointwise scoring, which can be employed for vision model preference alignment. Specifically, (1) we first develop UnifiedReward on our constructed large-scale human preference dataset, including both image and video generation/understanding tasks. (2) Then, it is utilized to automatically construct high-quality preference pair data based on the vision models, fine-gradually filtering their outputs through pair ranking and point sifting. (3) Finally, these data are used for their preference alignment through Direct Preference Optimization (DPO). Experimental results demonstrate that joint learning to assess diverse visual tasks can lead to substantial mutual benefits and we apply our pipeline to both image and video understanding/generation tasks, significantly improving the performance in each domain.