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

MultiRC

Benchmarks

Task NameDataset NameSOTA ResultTrend
ClassificationMultiRC
Accuracy71.1
29
Reading ComprehensionMultiRC
MultiRC Accuracy72.9
25
Reading ComprehensionMultiRC
F1 Score88.2
17
Multi-Sentence Reading ComprehensionMultiRC
F182.72
16
Explanation EvaluationMultiRC (test)
Sufficiency13.19
16
Question AnsweringMultiRC
F1 Score87.82
14
Machine Reading ComprehensionMultiRC (dev)
F1 Score77.5
10
Reading ComprehensionMultiRC
Total Communication Time13,150
9
Reading ComprehensionMultiRC
Accuracy (0-shot)10.3
6
ClassificationMultiRC Dir alpha=0.1
Generalized Accuracy72.53
5
Reading ComprehensionMultiRC Dir alpha=0.1 Standard
Personalized Accuracy (Acc_p)75.21
5
Machine Reading ComprehensionMultiRC SuperGLUE (test)
EM27.2
5
Text ClassificationMultiRC ERASER (test)
Weighted Avg F1 (MultiRC)67
5
Question AnsweringMultiRC SuperGLUE (dev)
Accuracy68.67
4
Showing 14 of 14 rows