AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

MAGE: MAsked Generative Encoder To Unify Representation Learning and Image Synthesis

Tianhong Li, Huiwen Chang, Shlok Mishra, Han Zhang, Dina Katabi, Dilip Krishnan

CVPR 2023 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling · 115 citations (OpenAlex)

This paper tried a common change: Add a contrastive loss

Add contrastive loss to reconstructive loss · ImageNet-1K linear probing · top-1 accuracy: 73.3 -> 77.1 Improved

What 1,317 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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 variable rather than fixed masking ratios (ablation)MAGE · ImageNet-1K class-unconditional generationFID: N/A -> 12.5Improved
Use variable rather than fixed masking ratios (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 71.5 -> 72.2Improved
Use quantized tokens instead of unquantized features (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 49.5 -> 74.7Improved
Add image-specific [C] padding instead of universal [MASK] padding (ablation)MAGE · ImageNet-1K class-unconditional generationFID: 12.4 -> 11.6Improved
Add image-specific [C] padding instead of universal [MASK] padding (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 72.5 -> 73.3Improved
Use weak rather than strong augmentation (ablation)MAGE · ImageNet-1K class-unconditional generationFID: 11.1 -> 8.67Improved
Use weak rather than strong augmentation (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 74.7 -> 70.5Hurt
Train for more pre-training epochs (ablation)MAGE · ImageNet-1K class-unconditional generationFID: 12.2 -> 11.6 -> 11.1Improved
Train for more pre-training epochs (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 72.2 -> 73.3 -> 74.7Improved
Add contrastive loss to reconstructive loss (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 73.3 -> 77.1Improved
Remove reconstructive loss, leaving contrastive loss only (ablation)MAGE · ImageNet-1K linear probingtop-1 accuracy: 77.1 -> 72.9Hurt
Reduce maximum masking ratio (ablation)MAGE-C · ImageNet-1K class-unconditional generationFID: 14.1 -> 23.5 -> 27.0Hurt

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