VideoTG-R1: Boosting Video Temporal Grounding via Curriculum Reinforcement Learning on Reflected Boundary Annotations
Open paper
Benchmark
Model
Rank
Results
moment-retrieval-on-charades-sta
VideoTG-R1 (extra data)
#20
R@1 IoU=0.5: 64.8
R@1 IoU=0.7: 39.2
R@1 IoU=0.3: 79.5
mIoU: 55.9
Rank counts only results with a code link.