A comprehensive framework designed to cultivate VLMs with human-like visuospatial abilities.
Ray Yang
rayruiyang
AI & ML interests
None yet
Recent Activity
upvoted a paper about 12 hours ago
ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training upvoted a paper about 1 month ago
HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone upvoted a paper about 2 months ago
Read It Back: Pretrained MLLMs Are Zero-Shot Reward Models for Text-to-Image GenerationOrganizations
None yet
models 6
rayruiyang/VST-3B-SFT
Image-Text-to-Text • 4B • Updated • 19
rayruiyang/VST-3B-RL
Image-Text-to-Text • 4B • Updated • 100 • 3
rayruiyang/VST-7B-RL
Image-Text-to-Text • 8B • Updated • 185
rayruiyang/VST-7B-SFT
Image-Text-to-Text • 8B • Updated • 44
rayruiyang/Haplo-8B-MI-672
8B • Updated • 4
rayruiyang/Haplo-8B-672
8B • Updated • 3 • 1