Answer Selection on TREC-QA (test)
92.57MAPASR
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ASRBackbone=RoBERTa Base, k=32021.07 | 92.57 | 97.06 | 98.16 | |
| MASR-FBackbone=RoBERTa Base, k=32021.07 | 92.22 | 94.12 | 97.06 | |
| MASRBackbone=RoBERTa Base, k=32021.07 | 92 | 92.65 | 96.32 | |
| Joint Model PairwiseBackbone=RoBERTa Base, k=32021.07 | 91.96 | 95.59 | 97.43 | |
| MASR-FPBackbone=RoBERTa Base, k=32021.07 | 91.91 | 95.59 | 97.43 | |
| KGATBackbone=RoBERTa Base, k=22021.07 | 91.55 | 94.12 | 96.45 | |
| Reranker by Garg et al., 2020Backbone=RoBERTa Base2021.07 | 91.4 | — | 95.2 | |
| Joint Model Multi-classifierBackbone=RoBERTa Base, k=52021.07 | 90.52 | 89.71 | 94.24 | |
| Our RerankerBackbone=RoBERTa Base2021.07 | 90.43 | 91.18 | 94.98 |