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

Question Answering on TruthfulQA (test)

72Accuracy

ART-max

16.8831.1945.559.81Feb 11, 2026Feb 20, 2026Mar 1, 2026Mar 10, 2026Mar 19, 2026Mar 28, 2026Apr 7, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.04
72
2026.04
71.8
2026.04
71.6
2026.04
63.3
2026.04
63.1
2026.04
62.3
2026.04
54.6
2026.04
54
2026.04
51.2
2026.04
46.8
2026.04
46.5
2026.04
46.4
2026.04
46.3
2026.04
46.2
2026.04
46.1
2026.04
43.8
2026.04
43.2
2026.04
41.7
2026.02
30.45
2026.02
29.38
2026.02
28.76
2026.02
28.02
2026.04
21.7
2026.04
21.5
2026.04
19