Diffusion Model with Cross Attention as an Inductive Bias for Disentanglement
Keywords: diffusion models, disentangled representation
TL;DR: In this paper, we introduce a new perspective and framework, demonstrating that diffusion models with cross-attention can serve as a powerful inductive bias to facilitate the learning of disentangled representations.
Abstract: Disentangled representation learning strives to extract the intrinsic factors within observed data. Factorizing these representations in an unsupervised manner is notably challenging and usually requires tailored loss functions or specific structural designs. In this paper, we introduce a new perspective and framework, demonstrating that diffusion models with cross-attention can serve as a powerful inductive bias to facilitate the learning of disentangled representations. We propose to encode an image to a set of concept tokens and treat them as the condition of the latent diffusion for image reconstruction, where cross-attention over the concept tokens is used to bridge the interaction between the encoder and diffusion. Without any additional regularization, this framework achieves superior disentanglement performance on the benchmark datasets, surpassing all previous methods with intricate designs. We have conducted comprehensive ablation studies and visualization analysis, shedding light on the functioning of this model. We anticipate that our findings will inspire more investigation on exploring diffusion for disentangled representation learning towards more sophisticated data analysis and understanding.
Primary Area: Diffusion based models
Submission Number: 2944
Loading