metadata
widget:
- 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). 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
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 = nli_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",
}