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

Multi-task Generalization on Other tasks (test)

60.28Score

LM-Cocktail10

35.496841.930948.36554.7991Nov 22, 2023
Updated 1mo ago

Evaluation Results

MethodLinks
2023.11
60.28
2023.11
58.57
2023.11
58.52
2023.11
57.87
2023.11
56.31
2023.11
54.58
2023.11
54.09
2023.11
53.53
2023.11
52.19
2023.11
51.64
2023.11
50.88
2023.11
50.52
2023.11
49.48
2023.11
48.95
2023.11
48.32
2023.11
48.21
2023.11
48.21
2023.11
47.73
2023.11
47.49
2023.11
47.48
2023.11
47.33
2023.11
47.11
2023.11
47.1
2023.11
46.9
2023.11
46.8
2023.11
46.11
2023.11
46.1
2023.11
46.02
2023.11
45.74
2023.11
45.03
2023.11
42.77
2023.11
41.63
2023.11
39.56
2023.11
38.94
2023.11
38.58
2023.11
36.45