Question Answering on SQuAD hidden 1.1 (test)
82.3EMHuman
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HumanModel Type=Ensemble2017.11 | 82.3 | 91.2 | |
| FusionNetModel Type=Ensemble2017.11 | 78.8 | 85.9 | |
| DCN+Model Type=Ensemble2017.11 | 78.7 | 85.6 | |
| R-netModel Type=Ensemble2017.11 | 78.2 | 85.2 | |
| R. Mnemonic ReaderModel Type=Ensemble2017.11 | 77.7 | 84.9 | |
| FusionNetModel Type=Single2017.11 | 76 | 83.9 | |
| R-netModel Type=Single2017.11 | 75.7 | 83.5 | |
| MEMENModel Type=Ensemble2017.11 | 75.4 | 82.7 | |
| ReasoNetModel Type=Ensemble2017.11 | 75 | 82.3 | |
| DCN+Model Type=Single2017.11 | 74.9 | 82.8 | |
| R. Mnemonic ReaderModel Type=Single2017.11 | 73.2 | 81.8 | |
| RaSoRModel Type=Single2017.11 | 70.8 | 78.7 | |
| DrQAModel Type=Single2017.11 | 70.7 | 79.4 | |
| ReasoNetModel Type=Single2017.11 | 70.6 | 79.4 | |
| SEDTModel Type=Single2017.11 | 68.2 | 77.5 | |
| BiDAFModel Type=Single2017.11 | 68 | 77.3 | |
| Match-LSTMModel Type=Single2017.11 | 64.7 | 73.7 | |
| LR BaselineModel Type=Single2017.11 | 40.4 | 51 |