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

Natural Language Reasoning on DROP

88.9Accuracy

GPT-4o + QuaSAR

48.23658.79369.3579.907Feb 18, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
88.9
2025.02
88
2025.02
84.2
2025.02
83.4
2025.02
81.4
2025.02
80.2
2025.02
69
2025.02
66.4
2025.02
64
2025.02
63.9
2025.02
60.2
2025.02
60
2025.02
59.3
2025.02
58.7
2025.02
58.4
2025.02
57.9
2025.02
57.3
2025.02
57.2
2025.02
56.8
2025.02
56
2025.02
55.3
2025.02
55
2025.02
54.6
2025.02
54.4
2025.02
54.2
2025.02
54.1
2025.02
53
2025.02
52.5
2025.02
52.5
2025.02
51.2
2025.02
50.8
2025.02
50.8
2025.02
49.8