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

Question Answering on 7 QA tasks

69.44Accuracy

Dense Model

30.85640.87350.8960.907Feb 6, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
69.44
2026.02
68.28
2026.02
66.96
2026.02
66.89
2026.02
66.76
2026.02
65.9
2026.02
65.74
2026.02
65.04
2026.02
64.71
2026.02
64.68
2026.02
64.64
2026.02
64.28
2026.02
64.24
2026.02
64.24
2026.02
63.8
2026.02
63.23
2026.02
63
2026.02
60.21
2026.02
59.85
2026.02
59.7
2026.02
59.61
2026.02
58.4
2026.02
58.12
2026.02
57.87
2026.02
55.02
2026.02
54.94
2026.02
53.97
2026.02
53.95
2026.02
53.74
2026.02
53.58
2026.02
52.37
2026.02
50.98
2026.02
49.81
2026.02
48.74
2026.02
47.59
2026.02
41.39
2026.02
40.61
2026.02
39.01
2026.02
38.04
2026.02
34.43
2026.02
32.34
2026.02
32.34