activity
20242026
most citedTowards Uncovering How Large Language Model Works: An Explainability Perspective

8 citations · 10 across the 4 of their papers we have counts for

collaborators
Showing cs.CVShow all

1 paper · 1 filter

cs.CV2026

LoopREG: Representation Supervision Enables Test-Time Recurrent Depth in Diffusion Transformers

Zipeng Wang, Anping Xiong, Jing Zhang +17

Reusing Transformer blocks offers a way to increase diffusion-model computation at test time, but naive recurrence can degrade generation beyond its training depth. We introduce Lo…