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

Multiple-Choice Video QA on EgoSchema latest (test)

72.2Accuracy

GPT4-O

37.04846.17455.364.426Jun 11, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.06
72.2
2024.06
63.9
2024.06
63.2
2024.06
62
2024.06
61.5
2024.06
60.9
2024.06
58
2024.06
55.7
2024.06
55.6
2024.06
53.3
2024.06
53.1
2024.06
51.7
2024.06
50.5
2024.06
43.9
2024.06
42.2
2024.06
38.5
2024.06
38.4