Multi-hop Reading Comprehension on HotpotQA distractor setting (dev)
70.57All ScoreDECOMPRC
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DECOMPRCTraining=Standard2019.06 | 70.57 | 72.53 | 62.78 | 84.31 | 58.74 | |
| BERTTraining=Standard2019.06 | 67.08 | 69.41 | 57.81 | 82.98 | 53.38 | |
| DECOMPRCTraining=1hop train (single-hop QA data only)2019.06 | 61.73 | 61.57 | 62.36 | 79.38 | 46.53 | |
| BiDAFTraining=Standard2019.06 | 58.28 | 59.09 | 55.05 | — | — | |
| BERTTraining=1hop train (single-hop QA data only)2019.06 | 56.27 | 62.77 | 30.4 | 87.21 | 29.64 |