| Benchmark | Model | Rank | Results |
|---|---|---|---|
| natural-language-inference-on-snli | RoBERTa-large+Self-Explaining | #3 | % Test Accuracy: 92.3Parameters: 340 |
| natural-language-inference-on-snli | RoBERTa-large + self-explaining layer | #4 | % Test Accuracy: 92.3% Train Accuracy: ?Parameters: 355m+ |
| paraphrase-identification-on-quora-question | 1-3[0.8pt/2pt] Random | #22 | Accuracy: 80 |
| sentiment-analysis-on-sst-5-fine-grained | RoBERTa-large+Self-Explaining | #3 | Accuracy: 59.1 |