AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

GIVT: Generative Infinite-Vocabulary Transformers

Michael Tschannen, Cian Eastwood, Fabian Mentzer

ECCV 2024 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling · 46 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
replace discrete VQ tokens with continuous latents modeled by a Gaussian mixture (GIVT-MaskGIT vs MaskGIT)MaskGIT · ImageNet 512x512FID: 7.8007 -> 4.8635Improved
replace discrete VQ tokens with continuous latents (GIVT-MaskGIT vs MaskGIT)MaskGIT · ImageNet 256x256FID: 4.916 -> 4.64Improved
add distribution-based classifier-free guidance (DB-CFG) (ablation)GIVT-Causal · ImageNet 256x256FID: 5.6729 -> 3.35Improved
add adapter (invertible flow) and larger model (ablation)GIVT-Causal · ImageNet 256x256FID: 3.35 -> 2.5933 (GIVT-Causal vs Causal-L+A with DB-CFG)Improved
increase GMM mixtures k from 1 to 16 (ablation)GIVT-Causal · ImageNet 256x256sampling FIDImproved
increase VAE KL weight beta (ablation)beta-VAE with GIVT · ImageNet 256x256sampling FID vs reconstruction FIDMixed
model full covariance Gaussian instead of diagonal GMM (ablation)GIVT-MaskGIT · ImageNetFID: about 3% gainNo clear effect
replace VQ-VAE with beta-VAE and GIVT in UViMUViM · COCO Panoptic and NYU Depth v2PQ and RMSE: PQ 39.0 -> 40.2; RMSE 0.459 -> 0.474Mixed
GIVT-Causal vs ViT-VQGAN with larger model and 4x longer sequenceunconditional generative model · ImageNet linear probingaccuracy: 65.1 vs 65.1No clear effect

Later papers that compared with this method

3 later accepted papers report GIVT (1.67B) as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

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