Question Answering On Quora Question Pairs
평가 지표
Accuracy
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Accuracy |
|---|---|
| exploring-the-limits-of-transfer-learning | 90.4% |
| how-to-train-bert-with-an-academic-budget | 70.7 |
| clear-contrastive-learning-for-sentence | 90.3% |
| electra-pre-training-text-encoders-as-1 | 90.1% |
| roberta-a-robustly-optimized-bert-pretraining | 90.2% |
| exploring-the-limits-of-transfer-learning | 88.0% |
| ernie-20-a-continual-pre-training-framework | 90.1% |
| big-bird-transformers-for-longer-sequences | 88.6% |
| exploring-the-limits-of-transfer-learning | 89.4% |
| simple-and-effective-text-matching-with-1 | 89.2 % |
| squeezebert-what-can-computer-vision-teach | 80.3% |
| deberta-decoding-enhanced-bert-with | 92.3% |
| albert-a-lite-bert-for-self-supervised | 90.5% |
| xlnet-generalized-autoregressive-pretraining | 92.3% |
| baseline-needs-more-love-on-simple-word | 83.03% |
| exploring-the-limits-of-transfer-learning | 89.7% |
| exploring-the-limits-of-transfer-learning | 89.9% |
| distilbert-a-distilled-version-of-bert | 89.2% |
| ernie-20-a-continual-pre-training-framework | 89.8% |