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

Commonsense Reasoning on aNLI

87.3Accuracy

RAINBOW

30.630445.342760.05574.7673Oct 17, 2022May 12, 2023Dec 6, 2023Jul 1, 2024Jan 25, 2025Aug 21, 2025Mar 17, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2023.05
87.3
2023.05
85.6
2022.10
84
2022.10
83.3
2022.10
83
2022.10
82.7
2023.05
71.56
2023.05
70.8
2023.05
70.5
2023.05
70
2023.05
67.54
2023.05
65.5
2023.06
64.37
2023.06
63.2
2023.06
63.18
2023.06
63.04
2023.06
61.88
2023.06
61.57
2023.06
61.42
2023.06
61.1
2023.06
60.94
2023.06
60.15
2023.06
60.14
2023.06
59.92
2023.06
57.63
2023.05
57.47
2023.05
57.4
2023.05
50.96
2026.03
34.24
2026.03
33.86
2026.03
33.35
2026.03
33.29
2026.03
33.25
2026.03
33.13
2026.03
32.81