| Benchmark | Model | Rank | Results |
|---|---|---|---|
| natural-language-inference-on-rte-glue-dev | ReFTA (RoBERTa-Large) | #4 | Accuracy: 87.1 |
| linguistic-acceptability-on-cola-glue-dev | ReFTA (RoBERTa-Large) | #6 | Matthews Correlation: 68.6 |
| natural-language-inference-on-qnli-glue-dev | ReFTA (RoBERTa-Large) | #6 | Accuracy: 94.8 |
| image-classification-on-cifar-100 | ReFTA (ViT-Large) | #13 | Percentage correct: 93.26 |