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

Reading Comprehension Accuracy and Calibration on RACE

74.93Accuracy

CORAL

22.118835.829449.5463.2506Feb 5, 2026Feb 9, 2026Feb 13, 2026Feb 17, 2026Feb 21, 2026Feb 25, 2026Mar 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
74.933.083.1855.0615.63
2026.02
73.062.623.0857.6916.06
2026.02
71.043.463.3554.416.88
2026.02
61.4511.977.5468.4317.77
2026.02
61.433.183.7257.5919.72
2026.02
60.562.453.2957.819.77
2026.02
58.385.116.4455.2215.55
2026.02
58.3710.66.4464.8421.48
2026.02
58.3415.613.0670.4519.36
2026.02
55.674.44.1661.5921.18
2026.02
55.6316.568.4473.1819.09
2026.02
54.934.2560.3620.84
2026.02
50.8310.668.5270.1525.1
2026.02
45.6921.4510.87104.6827.11
2026.02
40.834.674.78122.2726.52
2026.03
39----
2026.03
38.5----
2026.03
37.5----
2026.03
37.1----
2026.03
36.9----
2026.03
36.7----
2026.03
36.5----
2026.03
36.4----
2026.03
36.4----
2026.03
36.3----
2026.03
36.3----
2026.02
36.2----
2026.03
36.2----
2026.03
35.9----
2026.03
35.5----
2026.03
35.4----
2026.03
35.2----
2026.03
35.2----
2026.03
35.1----
2026.02
35----
2026.03
34.9----
2026.02
34.7----
2026.03
34.6----
2026.03
34.6----
2026.03
34.6----
2026.03
34.3----
2026.03
34.1----
2026.03
33.5----
2026.03
33.2----
2026.03
31.2----
2026.03
30.9----
2026.03
30.5----
2026.03
30.4----
2026.03
30.3----
2026.03
30.3----
2026.03
30.2----
2026.03
30.1----
2026.03
29.9----
2026.02
29.3----
2026.03
29----
2026.02
28.5----
2026.02
28.35----
2026.02
28.2----
2026.02
28.12----
2026.02
27.9----
2026.02
27.2----
2026.02
26.95----
2026.02
26.8----
2026.02
26.55----
2026.02
26.1----
2026.02
24.95----
2026.02
24.7----
2026.02
24.4----
2026.02
24.2----
2026.02
24.15----