发表机构
Fudan University; School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University; The University of Hong Kong（复旦大学; 西北工业大学光学与电子信息人工智能学院（iOPEN）; 香港大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本综述聚焦应用场景，提出三层渐进框架，梳理扩散与基于流的模型中表示学习的双向关系，分类下游任务方法，明确研究挑战并指明未来方向，为相关应用研究提供参考。
AI 中文摘要
扩散模型和基于流的模型近年来已成为生成建模领域的主导范式，这很大程度上得益于它们通过大规模训练学习丰富的多层次视觉表示的能力。这在生成模型与表示学习之间形成了双向关系：改进表示学习可提升生成质量，而学习到的表示可被用于更广泛的理解任务。本综述系统探讨了这种相互作用，重点关注应用场景。我们提出了一个三层渐进框架，从三个视角组织现有研究：利用表示学习提升生成能力、利用生成模型提取用于感知任务的表示，以及最终走向通用统一应用。我们系统地对广泛下游任务中的代表性方法进行分类，包括图像分类、密集视觉预测、实例级感知和标注稀缺场景。通过提供统一分类法并确定关键挑战，本综述旨在阐明当前研究的底层逻辑，并为未来探索提出有前景的方向。我们希望这项工作能为有兴趣利用生成模型的表示能力开展生成以外应用的研究人员提供有价值的参考。
英文摘要
Diffusion models and flow-based models have recently become the dominant paradigms in generative modeling, largely due to their ability to learn rich, multi-level visual representations through large-scale training. This creates a bidirectional relationship between generative models and representation learning: improving representation learning enhances generation quality, while the learned representations can be leveraged for broader understanding tasks. This survey systematically explores this interplay with a focus on applications. We propose a three-tier progressive framework that organizes existing works from three perspectives: using representation learning to improve generative capabilities, exploiting generative models to extract representations for perception tasks, and ultimately moving toward general-purpose unified applications. We systematically categorize representative methods across a wide range of downstream tasks, including image classification, dense visual prediction, instance-level perception, and annotation-scarce scenarios. By providing a unified taxonomy and identifying key challenges, this survey aims to clarify the underlying logic of current research and suggest promising directions for future exploration. We hope this work can serve as a valuable reference for researchers interested in harnessing the representation power of generative models for applications beyond generation.
CommentsAccepted by Vicinagearth
