What is recorded here?
Aligned projected noisy-input hidden features with clean-image features from a pretrained visual encoder as an auxiliary loss for diffusion and flow transformers.
Source-linked catalogue note. A detailed explanatory guide has not yet been prepared for this entry.
Scientific scope
External representation pretraining and ordinary denoising/flow objectives are inherited. Reported training-step savings depend on the baseline, target metric and setup; they are not end-to-end compute savings including the teacher's pretraining or universal convergence guarantees.
Read the original work
This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.
Optional · put your agent to work
Build on this exact record
Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.
Draft a question from this recordThe draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.
People and systems
Sihyun Yu; Sangkyung Kwak; Huiwon Jang; Jongheon Jeong; Jonathan Huang; Jinwoo Shin; Saining Xie
Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Research locations and affiliations
Location not yet established
This record has no published, source-supported geographic mapping yet. Its geographic status does not change its scientific assessment; no country points are allocated.
English catalogue narrative
REPA: external representation alignment for diffusion-transformer training
Aligned projected noisy-input hidden features with clean-image features from a pretrained visual encoder as an auxiliary loss for diffusion and flow transformers.
Scientific scope
External representation pretraining and ordinary denoising/flow objectives are inherited. Reported training-step savings depend on the baseline, target metric and setup; they are not end-to-end compute savings including the teacher's pretraining or universal convergence guarantees.
Reading status
This is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.
People and systems
Sihyun Yu; Sangkyung Kwak; Huiwon Jang; Jongheon Jeong; Jonathan Huang; Jinwoo Shin; Saining Xie
Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.