Hugging Face's logo

Datasets:
Video-Reason
/
VBVR-MultiStep-Bench

Tasks:
Image-to-Video
Text-to-Video
Languages:
English
Size:
n<1K
Tags:
video-reasoning
multi-step
long-horizon
image-to-video
evaluation
benchmark
License:
Dataset card Files
xet
 Community
VBVR-MultiStep-Bench / Multi-03_maze_marker_drawing_data-generator
1.94 MB
  • 1 contributor
 History: 1 commit
Mark7121983123's picture
Mark7121983123
Initial bench upload: 36 tasks, ~180 samples with GT for NeurIPS 2026 ED
75d9124 verified