AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Co-evolving Representations in Joint Image-Feature Diffusion

Theodoros Kouzelis, Spyros Gidaris, Nikos Komodakis

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers

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 stop-gradient on the clean projected representation target (ablation)CoReDi-B/2 · ImageNet 256 B/2 200K ablation (DINOv2)FID: 50.8 -> 24.7Improved
add batch normalization after the learnable projection (ablation)CoReDi-B/2 · ImageNet 256 B/2 200K ablation (DINOv2)FID: 223.9 -> 24.7Improved
add feature-variance regularization to the learnable projection (ablation)CoReDi-B/2 · ImageNet 256 B/2 200K ablation (DINOv2)FID: 37.2 -> 24.7Improved
unregularized learned projection versus fixed PCA projection (ablation)ReDi · ImageNet 256 B/2 200K ablation (DINOv2)FID: 30.9 -> 37.2Hurt
use learned projection with feature variance regularization instead of fixed PCAReDi · ImageNet 256 B/2 200K ablation (DINOv2)FID: 30.9 -> 24.7Improved
feature variance versus orthogonality regularization (ablation)CoReDi-B/2 · ImageNet 256 B/2 200K ablation (DINOv2)FID: 25.6 -> 24.7Improved
feature variance versus covariance regularization (ablation)CoReDi-B/2 · ImageNet 256 B/2 200K ablation (DINOv2)FID: 25.9 -> 24.7Improved
raise feature variance regularization weight from 0.5 to 1.5 (ablation)CoReDi-B/2 · ImageNet 256 B/2 200K ablation (DINOv2)FID: 25.9 -> 24.3Improved
set pixel-space representation loss weight to 0.1 instead of larger values (ablation)CoReDi-L/16 · ImageNet 256 pixel CoReDi-L/16 100KFID: 33.5 -> 31.5Improved
use coevolving representation on pixel diffusionDeCo-L/16 · ImageNet 256 pixel diffusion (no CFG)FID: 31.3 -> 21.5Improved
use coevolving representation on pixel diffusion at equal iterationsDeCo-L/16 · ImageNet 256 pixel diffusion (no CFG)FID: 46.0 -> 31.5Improved
use CoReDi instead of ReDi with DINOv2 encoderReDi · ImageNet 256 B/2 200K, DINOv2 encoderFID: 30.9 -> 24.7Improved

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