Answer Selection on TrecQA (raw)
78MAPRank MP-CNN (Rao et al.)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Rank MP-CNN (Rao et al.)number of parameters=10.0M, training time=130s2017.07 | 78 | 83 | |
| HYPERQAnumber of parameters=90K, training time=12s2017.07 | 77 | 82.5 | |
| MP-CNN (He et al.)number of parameters=10.0M, training time=141s2017.07 | 76.2 | 82.2 | |
| aNMM (Yang et al.)2017.07 | 75 | 81.1 | |
| HD-LSTM (Tay et al.)2017.07 | 75 | 81.5 | |
| CNN (Severyn & Moschitti)2017.07 | 74.6 | 80.8 | |
| BLSTM + BM25 (Wang & Nyberg)2017.07 | 71.3 | 79.1 | |
| CNN (Yu et al)2017.07 | 71.1 | 78.5 | |
| Yih et al (2014)2017.07 | 70.9 | 77 | |
| Severyn and Moschitti (2013)2017.07 | 67.8 | 73.6 | |
| Yao (2013)2017.07 | 63.1 | 74.8 | |
| Heilman et al. (2010)2017.07 | 60.9 | 69.2 | |
| Wang et al. (2007)2017.07 | 60.3 | 68.5 | |
| Wang et al. (2010)2017.07 | 59.5 | 69.5 |