README.md

---
widget:
 - text: "I almost forgot to eat lunch. </s></s> I didn't forget to eat lunch."
 - text: "I believe I will get into UW. </s></s> I will get into UW."
---

This is an off-the-shelf roberta-large model finetuned on WANLI, the Worker-AI Collaborative NLI dataset ([Liu et al., 2022](https://arxiv.org/abs/2201.05955)). It outperforms the `roberta-large-mnli` model on seven out-of-domain test sets, including by 11% on HANS and 9% on Adversarial NLI.

### How to use
```python
from transformers import RobertaTokenizer, RobertaForSequenceClassification

model = RobertaForSequenceClassification.from_pretrained('alisawuffles/roberta-large-wanli')
tokenizer = RobertaTokenizer.from_pretrained('alisawuffles/roberta-large-wanli')

x = tokenizer("I believe I will get into UW.", "I will get into UW.", hypothesis, return_tensors='pt', max_length=128, truncation=True)
logits = model(**x).logits
probs = logits.softmax(dim=1).squeeze(0)
label_id = torch.argmax(probs).item()
prediction = model.config.id2label[label_id]
```

### Citation
```
@misc{liu-etal-2022-wanli,
    title = "WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creation",
    author = "Liu, Alisa  and
      Swayamdipta, Swabha  and
      Smith, Noah A.  and
      Choi, Yejin",
    month = jan,
    year = "2022",
    url = "https://arxiv.org/pdf/2201.05955",
}
```