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

Natural Language Inference on ANLI (Accuracy and Calibration)

74.02Accuracy

Zero-shot-EI

-2.960817.024637.0156.9954Oct 31, 2022May 24, 2023Dec 16, 2023Jul 8, 2024Jan 30, 2025Aug 23, 2025Mar 17, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.02
74.02-----
2026.02
71-----
2026.02
65.1-----
2026.02
64.83-----
2026.02
64.45-----
2026.02
64.42-----
2026.02
64.18-----
2026.02
64-----
2026.02
63.95-----
2026.02
63.89-----
2026.02
63.58-----
2026.02
62.73-----
2026.02
61.67-----
2026.02
56.34-----
2026.02
55.75-----
2022.12
52.59-----
2026.02
52.18-----
52-----
2022.12
51.51-----
2022.12
51.48-----
2024.03
51.2-----
2024.03
49.58-----
2026.03
47.58-----4.32
2026.02
47.45-----
2026.02
47.36-----
2025.07
47.31-----
2025.07
47.03-----
2025.07
46.35-----
2025.07
45.82-----
2026.02
45.42-----
2026.02
44.23-----
2025.07
43.96-----
2024.03
43.58-----
2024.03
43.5-----
2026.02
43.07-----
2026.02
42.72-----
2026.02
41.22-----
2025.09
39.06-----
2025.09
38.28-----
2025.09
38.28-----
2025.09
38.28-----
2026.02
37.92-----
2026.03
37.61-----6.89
2025.07
34.7-----
2025.09
34.38-----
2025.07
33.22-----
2025.07
33.17-----
2025.07
31.52-----
2022.10
31.4185.4954.0916.4986.4-
2022.10
31.4140.9238.365.4343.82-
2022.10
31.3175.4844.1726.8790.3-
2022.10
31.3185.5854.2716.2286.41-
2022.10
31.3181.6550.7418.3981.66-
2022.10
31.3141.6738.6865.8445.11-
2025.09
31.25-----
2025.09
31.25-----
2022.10
30.6937.0228.3768.8439.62-
2022.10
30.6638.6546.0668.441.76-
2022.10
30.577.7147.2123.7778.36-
2022.10
30.3487.4557.1113.8688.03-
2025.07
28.01-----
2025.09
27.34-----
2025.07
0-----
2025.07
0-----
2025.07
0-----
2024.03
-42.7211.12---
-35.96.94---