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

Question Answering on OpenBookQA published (test)

65.4Accuracy

CORAL

31.49640.29849.157.902Feb 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
65.47.217.4466.1120.92
2026.02
63.210.666.6569.5720.7
2026.02
62.811.478.1981.2523.19
2026.02
5228.928.192.0932.97
2026.02
50.411.48.870.1824.5
2026.02
50.2311.47.9160.522.48
2026.02
49.817.6410.3774.5425.4
2026.02
49.817.649.3264.2623.31
2026.02
4817.1211.5585.1124.87
2026.02
47.615.1310.4672.626.53
2026.02
47.619.5711.4380.4728.73
2026.02
46.4817.1210.3873.3722.82
2026.02
34.4417.4111.2569.0330.03
2026.02
34.0326.9513.2673.3231.13
2026.02
32.826.1514.7683.7130.48