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

Multi-modal Evaluation on SEED-Bench all

14.54Performance Gain Sum

Task Arithmetic

-29.0152-17.7076-6.44.9076Mar 31, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
14.54
2025.03
12.02
10.58
2025.03
-4.02
2025.03
-6.96
-27.34