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

Question Answering on NaturalQA (test)

67.4Accuracy

MemDefrag

-2.69615.50233.751.898Oct 1, 2025Nov 16, 2025Jan 2, 2026Feb 17, 2026Apr 5, 2026May 21, 2026Jul 7, 2026
Updated 17d ago

Evaluation Results

MethodLinks
2026.07
67.4-
2026.07
67.2-
2026.07
66.8-
2026.07
66-
2026.07
63.2-
2026.07
62.8-
2026.07
62.2-
2026.07
62-
2026.07
59.2-
2026.07
59.2-
2026.07
59-
2026.07
53.2-
2026.07
52.6-
2026.07
50-
2026.07
44.6-
2026.07
43-
2026.07
41.6-
2026.07
37.8-
2026.07
36.2-
2026.07
33.6-
2026.07
26.2-
2025.10
23.442.5
2026.07
23.4-
2025.10
22.356.2
2025.10
21.743.3
2026.07
21-
20.957.4
2026.07
19.6-
18.953.6
2026.07
18.4-
2026.07
18.2-
2026.07
18-
2026.07
17.6-
2026.07
17.4-
2026.07
17.2-
2025.10
14.754.8
0.153.6
2025.10
0-
2025.10
0-