Official paper, fitted lens, and interactive full-reference perceptual distance demo.
Shreshth
shreshthsaini
AI & ML interests
Generative Models, Image/Video Gen, Post-training, RL for Media Gen, Inverse Problems, Computer vision, video processing, ML
Recent Activity
liked a model 1 day ago
migtissera/Tess-4.5-27B-GGUF new activity 1 day ago
chaofengc/IQA-PyTorch-Weights:Add JLD lens weights (DINOv2-S/14, block 1, k=64) upvoted a collection 20 days ago
MEND: RL for Flow Models via Proximal Velocity MatchingOrganizations
models 30
shreshthsaini/Rectified-CFGpp
Text-to-Image • Updated
shreshthsaini/CachedSearch
Text-to-Video • Updated • 1
shreshthsaini/JLD
Other • Updated • 38 • 1
shreshthsaini/MEND-Z-Image-Turbo-PickScore
Text-to-Image • Updated • 104 • • 1
shreshthsaini/MEND-Z-Image-Turbo-HPSv2.1
Text-to-Image • Updated • 89 •
shreshthsaini/MEND-SD3M-PickScore-Seed2
Text-to-Image • Updated • 39
shreshthsaini/MEND-SD3M-PickScore-Seed1
Text-to-Image • Updated • 54
shreshthsaini/MEND-SD3.5M-ImageReward
Text-to-Image • Updated • 42
shreshthsaini/MEND-SD3.5M-HPSv2.1
Text-to-Image • Updated • 52
shreshthsaini/MEND-SD3.5M-CLIPScore
Text-to-Image • Updated • 88