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

Factuality Evaluation on FActScore

84.5Pairwise Score

Grounded Decoding

-2.901619.789242.4865.1708Sep 28, 2025Nov 7, 2025Dec 18, 2025Jan 27, 2026Mar 9, 2026Apr 18, 2026May 29, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
84.50.89
2026.05
81.20.89
2026.05
79.50.87
2026.05
78.90.87
2026.05
78.10.86
2026.05
74.50.85
2026.05
73.20.86
2026.05
72.50.88
71.20.88
2026.01
69.3-
2026.01
50-
2026.01
48.9-
2025.09
0.612-
2025.09
0.576-
2025.09
0.557-
2025.09
0.549-
2025.09
0.548-
2025.09
0.539-
2025.09
0.53-
2025.09
0.527-
2025.09
0.52-
2025.09
0.487-
2025.09
0.468-
2025.09
0.46-