Single Sentence Classification on THUCNews (dev)
98.3AccuracyRoBERTa-wwm-ext
Evaluation Results
| Method | Links | |
|---|---|---|
| RoBERTa-wwm-extSize=Base, Whole Word Masking=True, extended=True2020.04 | 98.3 | |
| RoBERTa-wwm-extSize=Large, Whole Word Masking=True, extended=True2020.04 | 98.3 | |
| RoBERTaModel Scale=Base, Pre-train Data=Extended2021.06 | 98.3 | |
| RoBERTaModel Scale=Large, Pre-train Data=Extended2021.06 | 98.3 | |
| ChineseBERTModel Scale=Large, Pre-train Data=Standard2021.06 | 98.3 | |
| RoBERTa-wwmModel Scale=Base2019.06 | 98.3 | |
| RoBERTa-wwmModel Scale=Large2019.06 | 98.3 | |
| MacBERTSize=Base2020.04 | 98.2 | |
| ELECTRASize=Large2020.04 | 98.2 | |
| MacBERTModel Scale=Base, Pre-train Data=Standard2021.06 | 98.2 | |
| MacBERTModel Scale=Base2019.06 | 98.2 | |
| ELECTRAModel Scale=Large2019.06 | 98.2 | |
| ELECTRASize=Base2020.04 | 98.1 | |
| MacBERTSize=Large2020.04 | 98.1 | |
| ChineseBERTModel Scale=Base, Pre-train Data=Standard2021.06 | 98.1 | |
| MacBERTModel Scale=Large, Pre-train Data=Standard2021.06 | 98.1 | |
| ELECTRAModel Scale=Base2019.06 | 98.1 | |
| MacBERTModel Scale=Large2019.06 | 98.1 | |
| BERT-wwmWhole Word Masking=True2020.04 | 98 | |
| BERTModel Scale=Base, Pre-train Data=Standard2021.06 | 98 | |
| BERT-wwmModel Scale=Base2019.06 | 98 | |
| BERTSize=Base2020.04 | 97.7 | |
| BERT-wwm-extWhole Word Masking=True, extended=True2020.04 | 97.7 | |
| BERTModel Scale=Base, Pre-train Data=Extended2021.06 | 97.7 | |
| BERTModel Scale=Base2019.06 | 97.7 | |
| BERT-wwm-extModel Scale=Base2019.06 | 97.7 | |
| ERNIEModel Scale=Base, Pre-train Data=Standard2021.06 | 97.6 | |
| MacBERTModel Scale=Large2019.06 | 59 | |
| RoBERTa-wwmModel Scale=Large2019.06 | 58.8 | |
| RoBERTa-wwmModel Scale=Base2019.06 | 57.4 | |
| MacBERTModel Scale=Base2019.06 | 57.4 | |
| ELECTRAModel Scale=Large2019.06 | 57.2 | |
| BERT-wwm-extModel Scale=Base2019.06 | 57 | |
| BERT-wwmModel Scale=Base2019.06 | 56.5 | |
| BERTModel Scale=Base2019.06 | 56.3 | |
| ELECTRAModel Scale=Base2019.06 | 56.1 |