Question Answering On Reclor
Metrics
Accuracy
Accuracy (easy)
Accuracy (hard)
Results
Performance results of various models on this benchmark
Model Name | Accuracy | Accuracy (easy) | Accuracy (hard) | Paper Title | Repository |
|---|---|---|---|---|---|
| XLNet-large | 56.0 | 75.7 | 40.5 | ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning | |
| RoBERTa-large | 55.6 | 75.5 | 40.0 | ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning | |
| BERT-large | 49.8 | 72.0 | 32.3 | ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning |
0 of 3 row(s) selected.