List your company

Video Understanding on Charades-STA (accuracy)

40.2AccuracyCapRL-Video-178K
13.78420.64227.534.358Jun 8, 2026

Evaluation results

MethodLinks
CapRL-Video-178KBackbone=Molmo2-8B2026.0640.2
CapRL-Video-178KBackbone=Molmo2-4B2026.0631.4
Tarsier2-7B-Video-178KBackbone=Molmo2-8B2026.0624
ShareGPT4Video-8B-Video-178KBackbone=Molmo2-8B2026.0623.3
GPT-4o-Video-178KBackbone=Molmo2-8B2026.0622.9
ShareGPT4Video-8B-Video-178KBackbone=Molmo2-4B2026.0619.3
Tarsier2-7B-Video-178KBackbone=Molmo2-4B2026.0616.5
GPT-4o-Video-178KBackbone=Molmo2-4B2026.0614.8