Sparrow is a data augmentation method that enriches the instruction diversity of video data. You can find related data and weights here.
Shukang Yin
xjtupanda
AI & ML interests
Computer Vision, Multimodal learning
Recent Activity
upvoted a paper about 12 hours ago
Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding updated a dataset 6 months ago
xjtupanda/VisualProbe_MediumOrganizations
None yet
datasets 6
xjtupanda/VisualProbe_Medium
Viewer • Updated • 268 • 24
xjtupanda/VisualProbe_Easy
Viewer • Updated • 141 • 18
xjtupanda/VisualProbe_Hard
Viewer • Updated • 106 • 35
xjtupanda/VLMsAreBlind
Viewer • Updated • 8.02k • 26
xjtupanda/VStar_Bench
Viewer • Updated • 191 • 342
xjtupanda/Sparrow-Synthetic
Preview • Updated • 137