Disentangled Generative Model

A survey on disentangled representation learning [1]

Disentangled Diffusion Model

  • [2]: make adjustment based on text embedding, learn optimal combination coefficients for different time steps.
  • [3]: disentangled gradient field, predict gradients conditioned on latent factors
  • [4]: semantic subcode and stochastic details
  • [5]: predict the direction change in the latent h-space

References

[1] Xin Wang, Hong Chen, Siao Tang, Zihao Wu, and Wenwu Zhu. “Disentangled Representation Learning.”

[2] Wu, Qiucheng, et al. “Uncovering the disentanglement capability in text-to-image diffusion models.” CVPR, 2023.

[3] Yang, Tao, et al. “DisDiff: Unsupervised Disentanglement of Diffusion Probabilistic Models.” arXiv preprint arXiv:2301.13721 (2023).

[4] Preechakul, Konpat, et al. “Diffusion autoencoders: Toward a meaningful and decodable representation.” CVPR, 2022.

[5] Kwon, Mingi, Jaeseok Jeong, and Youngjung Uh. “Diffusion models already have a semantic latent space.” arXiv preprint arXiv:2210.10960 (2022).