Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Video Question Answering on VideoMME (Length Breakdown, ER, CS Metrics)

76.7Accuracy (Short Length)

Qwen2.5-VL-7B

60.37264.61168.8573.089Mar 5, 2026Mar 13, 2026Mar 21, 2026Mar 29, 2026Apr 6, 2026Apr 14, 2026Apr 22, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
76.768.256.967.3100166.3-
2026.03
74.361.352.862.893.35.233-
2026.03
73.862.852.463.193.75.430.7-
2026.04
71.2256.2248.61-----
2026.04
67.8956.7848----1.31
2026.04
67.8955.6748.89----0.95
2026.04
67.6754.2249.11----1.18
2026.04
67.3354.8949.33----1.6
2026.03
6759.451.659.388.1---
2026.04
66.7854.5647.22----2.32
2026.04
66.6755.2248.89----2.02
2026.04
65.8953.5648.11----3.15
2026.04
65.6753.7847.67----3.34
2026.04
65.2254.2248.22----2.9
2026.04
65.1154.5648.11----3.23
2026.04
64.8954.2247.67----3.17
2026.04
64.7854.4447.78----3.21
2026.04
62.5552.6746.67----5.12
2026.04
6152.8945.78----7.36