STVG-R1: Incentivizing Instance-Level Reasoning and Grounding in Videos via Reinforcement Learning
Open paper
Benchmark
Model
Rank
Results
spatio-temporal-video-grounding-on-hc-stvg2
STVG-R1
–
Val m_vIoU: 40.8
Val vIoU@0.3: 67.9
Val vIoU@0.5: 38.8
Rank counts only results with a code link.