Response Selection on Ubuntu v2 (test)
91.9MRRCross-encoder
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Cross-encoderInitialization=This paper, Pre-training corpus=Reddit2019.04 | 91.9 | 86.5 | 99.1 | — | |
| Cross-encoderPre-training Data=Reddit, Pre-training Source=Our pre-training2019.04 | 91.9 | — | — | 86.5 | |
| Poly-encoderm=64, Mechanism=First-m, Initialization=This paper, Pre-training corpus=Reddit2019.04 | 91.5 | 85.9 | 99.1 | — | |
| Poly-encoder 16Pre-training Data=Reddit, Pre-training Source=Our pre-training, m context vectors=162019.04 | 91.5 | — | — | 86 | |
| Poly-encoder 64Pre-training Data=Reddit, Pre-training Source=Our pre-training, m context vectors=642019.04 | 91.5 | — | — | 85.9 | |
| Poly-encoder 360Pre-training Data=Reddit, Pre-training Source=Our pre-training, m context vectors=3602019.04 | 91.5 | — | — | 85.9 | |
| Bi-encoderInitialization=This paper, Pre-training corpus=Reddit2019.04 | 90.1 | 83.6 | 98.8 | — | |
| Poly-encoder 360Pre-training Data=Toronto Books + Wikipedia, Pre-training Source=Our pre-training, m context vectors=3602019.04 | 90.1 | — | — | 83.7 | |
| Bi-encoderPre-training Data=Reddit, Pre-training Source=Our pre-training2019.04 | 90.1 | — | — | 83.6 | |
| Poly-encoder 16Pre-training Data=Toronto Books + Wikipedia, Pre-training Source=Our pre-training, m context vectors=162019.04 | 89.9 | — | — | 83.4 | |
| Poly-encoder 64Pre-training Data=Toronto Books + Wikipedia, Pre-training Source=Our pre-training, m context vectors=642019.04 | 89.9 | — | — | 83.4 | |
| Cross-encoderPre-training Data=Toronto Books + Wikipedia, Pre-training Source=Our pre-training2019.04 | 89.7 | — | — | 83.1 | |
| Cross-encoderPre-training Data=Toronto Books + Wikipedia, Pre-training Source=Devlin et al., 20192019.04 | 89.4 | — | — | 82.8 | |
| Poly-encoder 64Pre-training Data=Toronto Books + Wikipedia, Pre-training Source=Devlin et al., 2019, m context vectors=642019.04 | 88.4 | — | — | 81.3 | |
| Poly-encoder 16Pre-training Data=Toronto Books + Wikipedia, Pre-training Source=Devlin et al., 2019, m context vectors=162019.04 | 88.3 | — | — | 81.2 | |
| Bi-encoderPre-training Data=Toronto Books + Wikipedia, Pre-training Source=Our pre-training2019.04 | 88.2 | — | — | 80.8 | |
| Poly-encoder 360Pre-training Data=Toronto Books + Wikipedia, Pre-training Source=Devlin et al., 2019, m context vectors=3602019.04 | 88.1 | — | — | 80.9 | |
| Bi-encoderPre-training Data=Toronto Books + Wikipedia, Pre-training Source=Devlin et al., 20192019.04 | 88 | — | — | 80.6 | |
| Dong & Huang2019.04 | 84.8 | 75.9 | 97.3 | — | |
| Dong & Huang2019.04 | 84.8 | — | — | 75.9 | |
| Yoon et al.2019.04 | — | — | — | 65.2 |