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

Question Answering on NQ (Acc, CF, KF, MFS)

61.2Accuracy (Acc)

CR-DPO

21.57631.86342.1552.437May 11, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.05
61.279.473.576.4
2026.05
54.190.969.880.3
2026.05
52.17365.369.2
2026.05
51.793.736.765.2
2026.05
50.488.932.760.8
2026.05
49.666.759.262.9
2026.05
47.592.120.456.2
2026.05
47.593.930.262.1
2026.05
46.71004.152
2026.05
46.37353.163
2026.05
45.952.467.359.9
2026.05
43.81004.752.3
2026.05
43.856.167.461.8
2026.05
439111.451.2
2026.05
41.36.310053.2
2026.05
40.984.813.649.2
2026.05
40.156.155.855.9
2026.05
39.957.644.250.9
2026.05
36.47.610053.8
2026.05
33.918.858.838.8
2026.05
30.660.627.944.3
2026.05
23.110.945.728.3