AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Improved Masked Image Generation with Token-Critic

José Lezama, Huiwen Chang, Lu Jiang, Irfan Essa

ECCV 2022 · Image generation › Autoregressive and latent generation · approach: Accelerating autoregressive image generation · 35 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
add Token-Critic guided samplingMaskGIT · ImageNet 256x256FID: 6.56 -> 4.69Improved
add Token-Critic guided samplingMaskGIT · ImageNet 512x512FID: 8.48 -> 6.80Improved
add Token-Critic guided samplingMaskGIT · ImageNet 256x256IS: 203.6 -> 174.5Hurt
add Token-Critic guided samplingMaskGIT · ImageNet 512x512IS: 167.1 -> 182.1Improved
add Token-Critic with classifier rejection (20% acceptance)MaskGIT · ImageNet 256x256FID: 4.70 -> 3.75Improved
add Token-Critic with classifier rejection (20% acceptance)MaskGIT · ImageNet 512x512FID: 5.13 -> 4.03Improved
add Token-Critic with classifier rejection (20% acceptance)MaskGIT · ImageNet 512x512IS: 250.7 -> 305.2Improved
add classifier rejection samplingMaskGIT · ImageNet 256x256FID: 6.56 -> 4.70Improved
refine generated tokens with Token-CriticMaskGIT · ImageNet 256x256FID: 6.56 to 5.73Improved
refine generated tokens with Token-CriticMaskGIT · ImageNet 512x512FID: 8.48 to 7.64Improved
use Token-Critic instead of StyleGAN-XL (compare)MaskGIT · ImageNet 256x256FID: 3.26 vs 3.75Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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