AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Product-Quantised Image Representation for High-Quality Image Synthesis

Denis Zavadski, Nikita Philip Tatsch, Carsten Rother

ICLR 2026 · Image generation › Autoregressive and latent generation · approach: Hierarchical VQ-VAE generation · 0 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 vector quantisation with product quantisation (S>1) as latent dim grows (ablation)VQGAN autoencoder · ImageNet 256FIDImproved
increase number of subspaces S up to d/2 (ablation)PQGAN · ImageNet 256FIDImproved
increase codebook size K (ablation)PQGAN · ImageNet 256FIDNo clear effect
PQGAN (F=8, K=512) vs best baseline VQGAN-LC (K=100000)latent autoencoder · ImageNet 256 valPSNR (dB): 27.0 -> 37.4Improved
PQGAN F=16 vs continuous SDv2.1 VAE F=8latent autoencoder · ImageNet 256 valrFID: 0.75 -> 0.41Improved
train latent diffusion from scratch on PQGAN latents instead of SDv2.1 VAELDM 400M · ImageNet 256gFID: 27.21 -> 22.66Improved
adapt SD to PQ latent with F=16 and 48x48 latent resolutionSDv2.1 (PQSD-Quick) · 768x768 generationsamples/s: 0.116 -> 0.465Improved
PQGAN F=16 on 1536px generation vs SDv2.1 768pxSDv2.1 (PQSD-HR) · generation costsamples/s: 0.116 -> 0.112No clear effect

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