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

Question Answering on QA Suite Zero-shot (PIQA, ARC-E, ARC-C, BoolQ, HellaSwag, WinoGrande)

80.85PIQA Accuracy

FP16

48.55856.941565.32573.7085Jun 3, 2024Sep 22, 2024Jan 12, 2025May 4, 2025Aug 24, 2025Dec 14, 2025Apr 5, 2026
Updated 11d ago

Evaluation Results

MethodLinks
2026.04
80.8577.7853.4181.2879.1672.8474.22
2026.04
80.8577.7853.4181.2879.1672.8474.22
2024.06
80.7958.7146.2482.2980.7277.571.04
2026.04
80.2577.4452.7380.478.9973.2473.84
2026.04
80.277.2752.0580.1279.1472.7773.59
2024.06
80.0858.9245.4768.4479.2172.5367.44
2024.06
79.9858.2944.877.8979.2272.2168.73
2024.06
79.7157.9545.0579.8278.6672.2968.91
2026.04
79.1657.7444.4567.8376.4569.7765.9
2024.06
79.159.8944.4568.0176.2170.3166.33
2025.04
79.0559.8944.7168.4776.2370.2466.43
2026.04
79.0557.9144.269.0276.669.6966.08
2026.04
79.0557.9144.269.0276.669.6966.08
2025.04
7957.9544.2869.0276.5869.6966.09
2026.04
78.9475.8849.4977.5877.3970.871.68
2026.04
78.973.9547.3574.9576.7770.5670.41
2026.04
78.7858.2943.77-75.168.4364.87
2024.06
78.7356.5243.1768.8477.5370.9665.96
2026.04
78.6256.9443.4368.3576.1969.2265.46
2024.06
78.5656.9942.3266.7376.769.6165.15
2026.04
78.5156.1441.9868.4173.7865.9864.13
2026.04
78.2957.4143.8669.575.0266.9365.17
2026.04
78.2457.5843.8671.175.5268.3565.78
2026.04
78.2455.4741.368.274.0767.6464.15
2025.04
77.6458.3841.8168.573.7565.9864.64
2024.06
77.6457.3241.2166.7974.1265.9863.84
2026.04
77.4852.9939.33-71.3865.9861.43
2024.06
77.4752.4841.4673.087367.0764.09
2025.04
77.3756.7342.3267.6274.0768.2765.46
2026.04
77.3155.641.5566.6173.6866.0663.47
2024.06
77.2658.0441.5567.5573.6266.6964.12
2026.04
77.0452.9740.6170.472.7167.8863.6
2026.04
76.9952.9940.8770.472.4967.3263.51
2025.04
76.9352.140.768.172.762.5162.99
2026.04
76.9370.4544.9775.273.3666.1467.84
2026.04
76.8853.5440.5371.1372.9667.2563.72
2026.04
76.8853.5440.5371.1372.9667.2563.72
2025.04
76.6155.341.6467.0973.0265.5964.24
2026.04
76.5553.8340.9668.7555.8965.5960.26
2024.06
76.4450.0438.9970.9869.3964.7261.76
2024.06
76.2250.0438.3170.0969.8262.5961.18
2026.04
76.1269.5342.9271.9374.1367.0166.94
2026.04
75.8452.8239.5169.5770.0663.6161.9
2026.04
75.6868.4841.8171.9973.0766.2266.21
2026.04
75.5753.6239.9369.5471.7666.1462.76
2026.04
75.2451.8936.7767.8669.5462.1260.57
2026.04
74.6551.1838.6568.5969.6561.7260.74
2024.06
74.4851.640.6173.7673.7862.1262.73
2024.06
73.8350.6738.4-67.9158.5657.87
2024.06
73.5652.0639.68-70.9462.959.83
2025.04
72.0948.5734.1362.5462.6364.8857.47
2025.04
71.9849.9236.2665.960.5261.857.73
2024.06
71.9849.0740.0266.8570.4558.6459.5
2024.06
71.8148.0235.9273.2766.8159.5159.22
2024.06
71.4447.7435.4967.7163.8955.0156.88
2024.06
71.3847.634.3-63.759.4355.28
2024.06
71.3849.0736.6964.536858.5658.04
2024.06
71.2149.4534.4765.3364.6559.1956.63
2024.06
70.8449.4137.1270.1265.5358.6458.61
2026.04
70.4648.4834.39-62.855.4154.31
2026.04
69.847.2233.7965.4759.3455.4955.19
2024.06
69.6947.3933.162.8458.9655.854.37
2024.06
69.3742.5531.9163.7357.6555.3353.42
2024.06
68.7745.231.14-57.4356.6751.84
2024.06
68.0643.9835.3262.7550.7354.352.52
2026.04
67.6844.430.89-58.4556.5951.6
2024.06
67.6346.1734.460.754.3252.6452.62
2024.06
66.3243.929.6164.156.8854.752.58
2024.06
66.1545.231.1463.5156.4453.4352.65
2026.04
65.6144.2830.3862.6653.5151.8551.38
2024.06
64.4740.4429.8259.3839.952.2447.71
2025.04
64.4241.528.7563.3648.4956.6750.53
2024.06
6340.3230.3860.3453.6151.5449.86
2024.06
62.7339.9830.2960.2144.3952.9648.43
2026.04
62.340.2830.7260.4942.2449.9647.67
2025.04
61.1540.3629.5261.5645.6251.2248.24
2024.06
61.0439.1830.861.852.2951.0649.36
2026.04
60.1735.2327.1357.9237.0849.5744.52
2025.04
59.6836.6229.0158.5644.8452.0146.79
2025.04
59.0934.624.2362.1135.0351.3844.41
2024.06
58.6535.5327.7360.4235.5648.0644.83
2026.04
54.5731.924.2352.7231.2651.1440.97
2026.04
50.2226.9424.5737.9826.5550.236.08
2024.06
49.830.425.849.127.44838.41