Question Answering on HotpotQA full wiki (dev)
80.9F1AISO
Evaluation results
| Method | Links | ||
|---|---|---|---|
| AISOPre-trained Model Scale=large2021.09 | 80.9 | 68.1 | |
| TPRRPre-trained Model Scale=large, Source=Official Leaderboard2021.09 | 80.1 | 67.3 | |
| HopRetriever-plusVariant=+2021.09 | 79.2 | 66.6 | |
| DDRQA2021.09 | 76.9 | 62.9 | |
| AISOPre-trained Model Scale=base2021.09 | 76.5 | 63.5 | |
| HopRetriever2021.09 | 75.2 | 62.2 | |
| MDRVariant=+2021.09 | 75.1 | 62.3 | |
| Recurrent Graph-based RetrievalReader=BERT wwm2019.11 | 73.3 | 60.5 | |
| GRRVariant=+2021.09 | 73.3 | 60.5 | |
| Recurrent Graph-based RetrievalReader=BERT base2019.11 | 65.8 | 52.7 | |
| Transformer-XH2019.11 | 62.4 | 50.2 | |
| Semantic Retrieval2019.11 | 58.8 | 46.5 | |
| Semantic Retrieval2021.09 | 58.8 | 46.5 | |
| GoldEn Retriever2019.11 | 49.8 | — | |
| Cognitive Graph2019.11 | 49.4 | 37.6 | |
| CogQA2021.09 | 49.4 | 37.6 | |
| DecompRC2019.11 | 43.3 | — | |
| MUPPET2019.11 | 40.4 | 31.1 | |
| MUPPET2021.09 | 40.4 | 31.1 | |
| Baseline2019.11 | 34.4 | 24.7 |