Paraphrase Identification On Wikihop
Metriken
Accuracy
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| StructBERTRoBERTa ensemble | 90.7% | StructBERT: Incorporating Language Structures into Pre-training for Deep Language Understanding |
0 of 1 row(s) selected.