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

Question Answering on Reasoning Suite (ArcC, ArcE, PiQA, Wino) Zero-shot

50.43Arc Challenge (C) Accuracy

Dense

16.390825.227934.06542.9021May 16, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
50.4380.0979.7172.6170.71
2025.05
45.7376.9377.4770.3267.61
2025.05
45.5676.7678.0171.968.06
2025.05
44.4575.3376.9870.0866.71
2025.05
44.275.277.670.766.93
2025.05
42.3275.4678.4571.6766.98
2025.05
42.373.676.570.365.68
2025.05
41.0474.4576.8267.9565.07
2025.05
4071.57669.564.25
2025.05
39.272.975.668.263.98
2025.05
35.5868.6874.165.4360.95
2025.05
33.6167.5973.9964.6459.96
2025.05
32.6865.7872.6364.458.87
2025.05
25.0856.2267.4658.851.89
2025.05
21.6750.2565.1756.4348.38
2025.05
17.73656.15140.2