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

Open-Ended Generation on MLLMU-Bench (test)

34.5ROUGE-L

Base

14.32419.56224.830.038May 26, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
34.5
2026.05
33.8
2026.05
33.8
2026.05
33.6
2026.05
32.4
2026.05
32.4
2026.05
31.1
2026.05
26.2
2026.05
25.1
2026.05
24.2
2026.05
23.7
2026.05
23.7
2026.05
23.7
2026.05
23.6
2026.05
23.3
2026.05
23.1
2026.05
22.2
2026.05
21.7
2026.05
21.6
2026.05
21.3
2026.05
21.2
2026.05
20.3
2026.05
18.9
2026.05
18.7
2026.05
17.6
2026.05
17.3
2026.05
16.9
2026.05
16.2
2026.05
15.9
2026.05
15.1