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

QA performance by Gemini-2.5-Pro based on captions on World-Sense (test)

34World-Sense QA Score

ASID-Captioner

5.50412.90220.327.698Feb 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
34
2026.02
33.8
2026.02
32.3
2026.02
31
2026.02
25.7
2026.02
18.2
2026.02
14.1
2026.02
12.7
2026.02
11.2
2026.02
8.7
2026.02
8.6
2026.02
7.2
2026.02
6.6