Rethinking UMM Visual Generation: Masked Modeling for Efficient Image-Only Pre-training
Paper • 2603.16139 • Published • 29
Zhenglin Cheng (SII) PRO
kenshinn
AI & ML interests
SII is an institution dedicated to innovation in education and research in the field of Al. Zhenglin Cheng is a part of SII, focusing on unified multimodal models and few-step generation.
Recent Activity
upvoted a paper 19 days ago
Mode Seeking meets Mean Seeking for Fast Long Video Generation upvoted a paper 24 days ago
A Very Big Video Reasoning SuiteOrganizations
upvoted a paper 4 days ago
upvoted a paper 19 days ago
upvoted a paper 24 days ago
upvoted a paper about 1 month ago
upvoted a paper about 2 months ago
liked 2
models about 2 months ago
No COMFY
19
#3 opened 3 months ago
by 
BuBaLoM
liked a
model 3 months ago
authored 3
papers 3 months ago
Generative Modeling with Explicit Memory
Paper • 2412.08781 • Published • 1
G$^2$RPO-A: Guided Group Relative Policy Optimization with Adaptive Guidance
Paper • 2508.13023 • Published • 1
TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows
Paper • 2512.05150 • Published • 76
upvoted a collection 3 months ago
updated 2
models 3 months ago
liked a
model 3 months ago
published a
model 3 months ago
Support `TwinFlow` via `diffusers`
2
#6 opened 3 months ago
by 
alvarobartt
updated a
collection 3 months ago
Method for distillation
7
#2 opened 3 months ago
by 
diuibyang