Natural Language Inference on QNLI (dev)
0.945AccuracyEFL
Evaluation results
| Method | Links | |
|---|---|---|
| EFLTraining Setting=Full training dataset, Backbone=RoBERTa-large2021.04 | 0.945 | |
| Fine-tuningTraining Setting=Full training dataset, Backbone=RoBERTa-large2021.04 | 0.929 | |
| EFLTraining Setting=Few-shot with K=8, Backbone=RoBERTa-large2021.04 | 0.68 | |
| LM-BFFTraining Setting=Few-shot with K=8, Backbone=RoBERTa-large2021.04 | 0.618 | |
| Fine-tuningTraining Setting=Few-shot with K=8, Backbone=RoBERTa-large2021.04 | 0.527 | |
| MajorityTraining Setting=Full training dataset, Backbone=RoBERTa-large2021.04 | 0.505 |