AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

A Hidden Semantic Bottleneck in Conditional Embeddings of Diffusion Transformers

Trung X. Pham, Kang Zhang, Ji Woo Hong, Chang D. Yoo

ICLR 2026 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers · 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
prune low-magnitude tail dimensions (tau=0.01, ~39%) of the conditional embedding only at the first step (ablation)REPA · ImageNet-1K class-conditionalFID: 7.1694 -> 7.1690No clear effect
prune tail dimensions (tau=0.01) during the last steps (ablation)REPA · ImageNet-1K class-conditionalFID: 7.1694 -> 7.1598Improved
prune tail dimensions (tau=0.01) at every step (ablation)REPA · ImageNet-1K class-conditionalFID: 7.1694 -> 7.2143Hurt
prune 66% of tail dimensions (tau=0.02) (ablation)REPA · ImageNet-1K class-conditionalFID: 7.1694 -> 9.2202Hurt
prune 96% of tail dimensions (tau=0.05) (ablation)REPA · ImageNet-1K class-conditionalFID: 7.1694 -> 56.2308Hurt
remove 8 high-magnitude head dimensions (ablation)REPA · ImageNet-1K class-conditionalFID: 7.1694 -> 523.7637Hurt
prune tail dimensions (tau=0.01, every step) on LightningDiT (ablation)LightningDiT · ImageNet-1K class-conditionalFID: 7.0802 -> 7.0130 FID; IS 169.8574 -> 166.0569Mixed
prune tail dimensions (tau=0.01) at t0 on Model-Guide (ablation)Model-Guide · ImageNet-1K class-conditionalFID: 7.2478 -> 7.2466No clear effect
prune ~40% dimensions (tau=0.1) of conditional embedding (ablation)X-MDPT · DeepFashion pose-guided generationFID: 18.6372 -> 18.6692No 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