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

Question Answering on OpenBookQA (OBQA)

94.95Accuracy

Direct Fine-tuning

63.17871.426579.67587.9235Nov 15, 2022Jun 4, 2023Dec 22, 2023Jul 10, 2024Jan 27, 2025Aug 16, 2025Mar 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
94.95
2026.02
94.95
2026.02
94.83
2026.02
94.71
2026.02
94.71
2026.02
94.71
2026.02
94.59
2026.02
94.47
2022.11
89.2
2026.02
89.06
2026.02
88.7
2026.02
88.22
2026.02
88.1
2026.01
87.9
2026.02
87.74
2026.02
87.74
2026.02
87.5
2026.01
87.5
2026.02
87.26
2022.11
87.2
2026.01
86.6
2025.12
86
2025.12
86
2025.12
86
2026.01
85.1
2025.12
84
2025.12
84
2025.12
84
2025.12
84
2025.12
84
2025.12
84
2026.03
84
2026.03
84
2026.03
83.6
2026.01
83.5
2026.03
83.4
2026.03
83.2
2026.03
83.2
2025.12
83
2025.12
83
2026.03
83
2026.03
83
2026.01
82.9
2026.03
82.6
2026.03
82.4
2026.01
82.1
2025.12
82
2026.01
81.8
2026.01
81.7
2026.01
81.7
2026.01
81.6
2026.01
81.4
2026.01
81.2
2026.01
81.2
2026.01
81.1
2026.01
80.2
2026.01
79.5
2026.01
79.2
2025.12
79
2026.01
79
2026.01
78.4
2026.01
78.1
2026.01
77.81
2026.01
77.79
2026.01
77.77
2026.01
77.2
2025.12
77
2025.12
77
2026.01
76.5
2026.01
76.3
2026.01
76.1
2025.12
76
2024.04
75.8
2024.04
75.4
2026.01
74.49
2026.01
74.01
2026.03
73.8
2026.03
73.2
2026.03
73.2
2025.12
73
2026.01
72.29
2025.12
72
2025.12
72
2026.01
71.96
2026.01
71.47
2024.04
71.4
2026.03
71.2
2026.03
71
2024.04
70.6
2024.04
70.4
2026.01
70.01
2024.04
70
2026.01
70
2026.03
67
2026.01
66.29
2025.12
66
2026.03
66
2026.03
65
2024.04
64.8
2026.03
64.4
Showing 100 of 300 rows