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

Pronoun Resolution on WinoGrande

89.4Accuracy

UnifiedQA-T5-11B

46.65657.75368.8579.947May 26, 2021Mar 27, 2022Jan 26, 2023Nov 28, 2023Sep 28, 2024Jul 30, 2025Jun 1, 2026
Updated 22h ago

Evaluation Results

MethodLinks
2021.05
89.40.8571-
2021.05
84.60.7673-
2021.05
80.8--
2021.05
79.10.6641-
2021.05
77.9--
2026.03
75.33--
2026.02
75.06-0.32
2026.03
75--
2026.02
74.74--
2026.03
74.67--
2026.02
74.43--0.31
2026.03
74.33--
2026.03
74.33--
2026.02
74.11--
2026.03
74--
2026.02
73.72--1.02
2026.02
73.4--0.71
2021.05
73.30.6358-
2026.02
73.16--1.58
2026.02
72.93--
2026.02
72.45--1.66
2026.02
72.06--2.05
2026.02
71.87--1.06
2026.02
71.79--1.14
2026.02
71.43--1.5
2026.02
70.8--2.13
2026.02
70.48-1.42
2026.02
69.06--
2026.02
68.99--0.07
2026.02
68.39--0.67
2026.02
66.69--2.37
2021.05
650.6046-
2021.05
64.90.5289-
2026.05
63.06--
2026.05
61.88--
2026.05
60.46--
2021.05
59.20.6706-
2021.05
58.9--
2026.05
58.56--
2026.05
58.48--
2026.05
57.85--
2026.05
56.99--
2026.05
56.2--
2026.06
52.7--
2026.06
52.5--
2021.05
51.9--
2026.06
51.9--
2026.06
51.8--
2026.04
51.46--
2026.06
51.4--
2026.06
51.2--
2021.05
51--
2021.05
50.9--
2026.06
50.8--
2026.04
50.51--
2021.05
50--
2021.05
49.6--
2026.02
48.3--25.81