6 papers
Evaluating the Representation Space of Diffusion Models via Self-Supervised Principles
Xiao Li, Yixuan Jia, Zekai Zhang +6
Diffusion models have demonstrated remarkable generative capabilities and have also emerged as powerful self-supervised representation learners, yet the connection between these tw…
MCLR: Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives
Xiang Li, Yixuan Jia, Xiao Li +3
Diffusion models achieve strong generative performance but often rely on classifier-free guidance (CFG), an inference-time intervention that modifies the sampling trajectory. Since…
Generalization of Diffusion Models Arises with a Balanced Representation Space
Zekai Zhang, Xiao Li, Xiang Li +4
Diffusion models excel at generating high-quality, diverse samples, yet they risk memorizing training data when overfit to the training objective. We analyze the distinctions betwe…
Towards Understanding the Mechanisms of Classifier-Free Guidance
Xiang Li, Rongrong Wang, Qing Qu
Classifier-free guidance (CFG) is a core technique powering state-of-the-art image generation systems, yet its underlying mechanisms remain poorly understood. In this work, we begi…
Understanding Representation Dynamics of Diffusion Models via Low-Dimensional Modeling
Xiao Li, Zekai Zhang, Xiang Li +4
Diffusion models, though originally designed for generative tasks, have demonstrated impressive self-supervised representation learning capabilities. A particularly intriguing phen…
Understanding Generalizability of Diffusion Models Requires Rethinking the Hidden Gaussian Structure
Xiang Li, Yixiang Dai, Qing Qu
In this work, we study the generalizability of diffusion models by looking into the hidden properties of the learned score functions, which are essentially a series of deep denoise…