Machine Reading Comprehension on CJRC (dev)
66.5EMChineseBERT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ChineseBERTModel scale=Large2021.06 | 66.5 | 77.9 | |
| RoBERTaModel scale=Large, Pre-trained on extended data=true2021.06 | 65.6 | 77.5 | |
| ChineseBERTModel scale=Base2021.06 | 65.2 | 77.8 | |
| RoBERTaModel scale=Base, Pre-trained on extended data=true2021.06 | 62.9 | 76.6 | |
| MacBERT-largeSize=Large2020.04 | 62.4 | 82.3 | |
| RoBERTa-wwm-ext-LSize=Large2020.04 | 62.1 | 82.4 | |
| ELECTRA-largeSize=Large2020.04 | 61.9 | 82.1 | |
| BERTModel scale=Base, Pre-trained on extended data=true2021.06 | 60.8 | 74 | |
| MacBERT-baseSize=Base2020.04 | 60.4 | 80.3 | |
| BERTModel scale=Base, Pre-trained on extended data=false2021.06 | 59.8 | 73 | |
| ELECTRA-baseSize=Base2020.04 | 59 | 79.4 | |
| RoBERTa-wwm-extSize=Base2020.04 | 58.7 | 79.1 | |
| BERT-wwm-extSize=Base2020.04 | 55.6 | 76 | |
| BERT-wwmSize=Base2020.04 | 54.7 | 75.2 | |
| BERTSize=Base2020.04 | 54.6 | 75.4 |