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

Open-Ended Generation on MLLMU-Bench (Retain Set)

53.1ROUGE-L

PO

29.28435.46741.6547.833May 26, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
53.1
2026.05
52.9
2026.05
52.6
2026.05
52.5
2026.05
52.4
2026.05
52.3
2026.05
50.9
2026.05
50.6
2026.05
50.5
2026.05
50.5
2026.05
49.6
2026.05
49.3
2026.05
49.3
2026.05
49.3
2026.05
49
2026.05
48.1
2026.05
48.1
2026.05
47.4
2026.05
45.7
2026.05
44.1
2026.05
43.8
2026.05
41.6
2026.05
41.4
2026.05
41.3
2026.05
34.4
2026.05
32.9
2026.05
32.3
2026.05
31.3
2026.05
30.4
2026.05
30.2