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

Multi-task Generalization on Multi-task Overall Average

40.5Accuracy

Single Best

32.734.72536.7538.775May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.05
40.5
2026.05
38
2026.05
37.5
2026.05
37.2
2026.05
37.2
2026.05
36.4
2026.05
36.4
2026.05
35.6
2026.05
35.2
2026.05
35
2026.05
33