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

Question Answering on MLLMU-Bench 1.0 (Retain)

58.6Accuracy

Vanilla

30.41637.73345.0552.367Dec 12, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
58.6
2025.12
57.6
2025.12
56.4
2025.12
55.7
2025.12
55.6
2025.12
54.4
2025.12
53.3
2025.12
49.4
2025.12
37.5
2025.12
36.1
2025.12
35.1
2025.12
32.2
2025.12
32.2
2025.12
32.1
2025.12
32
2025.12
31.5