AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

InfoDiffusion: Representation Learning Using Information Maximizing Diffusion Models

Yingheng Wang, Yair Schiff, Aaron Gokaslan, Weishen Pan, Fei Wang, Christopher De Sa, Volodymyr Kuleshov

ICML 2023 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use InfoDiffusion instead of DiffAEDiffusion representation learning · FashionMNISTLatent Qual.: 0.822 \pm 0.006 -> 0.856 \pm 0.004Improved
Use InfoDiffusion instead of DiffAEDiffusion representation learning · CIFAR10Latent Qual.: 0.442 \pm 0.004 -> 0.461 \pm 0.003Improved
Use InfoDiffusion instead of DiffAEDiffusion representation learning · FFHQLatent Qual.: 0.565 \pm 0.001 -> 0.628 \pm 0.002Improved
Sample z from a learned latent prior instead of the standard priorInfoDiffusion · FashionMNISTFID: 22.5 -> 8.1Improved
Sample z from a learned latent prior instead of the standard priorInfoDiffusion · CIFAR10FID: 36.2 -> 25.7Improved
Sample z from a learned latent prior instead of the standard priorInfoDiffusion · FFHQFID: 19.8 -> 13.1Improved
Sample z from a learned latent prior instead of the standard priorInfoDiffusion · CelebAFID: 12.1 -> 7.6Improved
Add guidance on z with w=2 (ablation)InfoDiffusion with learned latent sampling · FashionMNISTFID: 8.1 -> 6.8Improved
Add guidance on z with w=2 (ablation)InfoDiffusion with learned latent sampling · CIFAR10FID: 25.7 -> 21.3Improved
Add guidance on z with w=2 (ablation)InfoDiffusion with learned latent sampling · FFHQFID: 13.1 -> 9.8Improved
Add guidance on z with w=2 (ablation)InfoDiffusion with learned latent sampling · CelebAFID: 7.6 -> 6.3Improved
Use the regularization setting (lambda,zeta)=(0.1,0.9) instead of zeta=1 with small lambda (ablation)InfoDiffusion · CelebAattribute information in latentsImproved

Later papers that compared with this method

1 later accepted paper reports InfoDiffusion as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy