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

Question Answering on NQ (test) (Acc, ASR)

70.2Accuracy

Vanilla RAG

12.79227.69642.657.504May 21, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.05
70.2-
2026.05
67.86
2026.05
67.4-
2026.05
67-
2026.05
6613.4
2026.05
65.215
2026.05
65.25.6
2026.05
64.86.4
2026.05
64.811.4
2026.05
64.86.4
2026.05
63.47.8
2026.05
629.8
2026.05
61.6-
2026.05
60.22.2
2026.05
602.4
2026.05
57.4-
2026.05
57.23.8
2026.05
57.27.4
2026.05
5726.8
2026.05
55.66.8
2026.05
49.85.2
2026.05
38.623.6
2026.05
32.261
2026.05
31.21.8
2026.05
27.4-
2026.05
24.25.4
2026.05
234.4
2026.05
22.21.6
2026.05
15.876.4
2026.05
1583.6