Extractive Question Answering on QUOREF (dev)
74.94EMCorefRoBERTa_LARGE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CorefRoBERTa_LARGEbackbone=RoBERTa-Large2020.04 | 74.94 | 81.71 | |
| RoBERTa_LARGEdescription=stronger baseline with reasoning framework2020.04 | 74.15 | 81.05 | |
| RoBERTa-MTtraining=multi-task pre-training, source=official leaderboard2020.04 | 74.11 | 81.51 | |
| CorefBERT_LARGEbackbone=BERT-Large2020.04 | 70.89 | 76.56 | |
| BERT_LARGEdescription=stronger baseline with reasoning framework2020.04 | 67.91 | 73.82 | |
| CorefBERT_Basebackbone=BERT-Base2020.04 | 66.87 | 72.27 | |
| BERT_BASEdescription=stronger baseline with reasoning framework2020.04 | 61.29 | 67.25 | |
| BERT_BASEdescription=simple fine-tuning2020.04 | 58.44 | 64.95 | |
| QANet+BERT_BASEsource=Dasigi et al. (2019)2020.04 | 43.09 | 47.38 | |
| QANetsource=Dasigi et al. (2019)2020.04 | 34.41 | 38.26 |