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

Text Classification on BoolQ

90.7Accuracy

Full Fine-Tuning (FT)

43.06855.43467.880.166Jul 14, 2021Apr 13, 2022Jan 11, 2023Oct 11, 2023Jul 10, 2024Apr 9, 2025Jan 8, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
90.7
2025.12
90.2
2025.12
89.6
2025.12
89.3
2025.12
88.6
2025.12
87.2
2025.12
86.7
2025.12
85.2
2025.05
83.4
2025.05
83.4
2025.05
78.2
2026.01
77.6
2024.04
77.1
2026.01
77.1
2026.01
77.1
2024.04
76.9
2024.04
76.8
2026.01
76.8
2026.01
76.6
2026.01
76.3
2026.01
76.1
2026.01
75.6
2022.05
75.55
2024.04
75.4
2026.01
75.3
2024.04
75.1
2024.04
75
2025.05
75
2026.01
74.9
2022.05
74.37
2022.05
74.12
2026.01
73.9
2026.01
73.8
2026.01
73.1
2026.01
73
2024.04
72.7
2024.04
72.6
2026.01
72.6
2026.01
72.5
2024.04
71.7
2026.01
70.5
2025.05
70.2
2025.05
69.6
2025.05
69.6
2025.05
69.2
2025.05
68.2
2026.01
67.6
2026.01
67.3
2024.04
67.2
2024.04
67.2
2026.01
66.9
2025.05
66.8
2025.05
66.4
2025.05
66.1
2025.05
66
2022.05
65.8
2024.04
65.8
2025.05
65
2026.01
64.8
2022.05
64.75
2025.05
64.6
2026.01
64.5
2022.05
64.4
2026.01
64.4
2026.01
63.4
2026.01
63
2026.01
62.5
2024.04
62.2
2024.04
62.2
2025.12
62.2
2024.04
62.1
2025.12
62
2025.12
61.8
2024.04
61.7
2021.07
60.5
2021.07
60.3
2025.05
59.5
2026.01
59.4
2025.05
59.1
2026.01
59
2021.07
58.9
2025.12
58.7
2021.07
55.3
2026.01
44.9