Conversational Response Selection On Polyai
Metriken
1-of-100 Accuracy
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| Multi-context ConveRT | 71.8% | ConveRT: Efficient and Accurate Conversational Representations from Transformers |
| ConveRT | 68.3% | ConveRT: Efficient and Accurate Conversational Representations from Transformers |
| PolyAI Encoder | 61.3% | A Repository of Conversational Datasets |
| USE | 47.7% | Universal Sentence Encoder |
| ELMO | 19.3% | Deep contextualized word representations |
0 of 5 row(s) selected.