EgoVITA: Learning to Plan and Verify for Egocentric Video Reasoning

Benchmark Model Rank Results
video-question-answering-on-tomatoEgoVITA (Qwen3-VL)#8Accuracy: 34.5
video-question-answering-on-mvbenchEgoVITA (InternVL-3.5)#17Avg.: 73.8
video-question-answering-on-lvbenchEgoVITA (Qwen3-VL)#21Accuracy: 56.5