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

Question Answering on NQ-Open (val)

49.62Accuracy

NWCAD

4.1215.932527.74539.5575Dec 14, 2025Jan 3, 2026Jan 24, 2026Feb 14, 2026Mar 6, 2026Mar 27, 2026Apr 17, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
49.62--
2026.04
49.5--
2026.04
48.6--
2026.04
45--
2026.04
45--
2026.04
44.6--
2026.04
43.8--
2026.04
43--
2026.04
42.6--
2026.04
42.25--
2026.04
38.4--
2026.04
37.6--
2026.04
35.2--
2026.04
35.2--
2025.12
30.7--
2026.02
290.6856.95
2025.12
26.5--
2025.12
26--
2026.04
23.25--
2026.02
22.60.8462.44
2025.12
22.1--
2025.12
22.1--
2026.02
21.910.5363.46
2025.12
21.2--
2025.12
20.7--
2026.02
19.470.1954.46
2025.12
18.9--
2025.12
18.3--
2026.02
17.950.7974.21
2025.12
17.9--
2025.12
17.6--
2026.04
16.62--
2026.04
16.4--
2025.12
16.2--
2025.12
15.5--
2025.12
14.9--
2025.12
13.5--
2025.12
12.9--
2025.12
12.6--
2026.02
11.860.3482.77
2025.12
11.8--
2025.12
11.1--
2026.04
11--
2025.12
10.8--
2025.12
7.8--
2025.12
5.87--