cardiffnlp/twitter-roberta-large-intimacy-latest

This is a RoBERTa-large model trained on 154M tweets until the end of December 2022 and finetuned for intimacy analysis (regression on a single text) on the TweetIntimacy dataset of SuperTweetEval. The original Twitter-large RoBERTa model can be found here.

Example

from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch.nn.functional as F

model_name = "cardiffnlp/twitter-roberta-large-intimacy-latest"
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

text= '@user Furthermore, harassment is ILLEGAL in any form!'

# with pipeline
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)
pipe(text)
>> [{'label': 'LABEL_0', 'score': 0.5246055722236633}]

# alternatively
logits = model(**tokenizer(text, return_tensors="pt"))
prob = F.sigmoid(logits.logits).item()
>> 0.5246055722236633

Citation Information

Please cite the reference paper if you use this model.

@inproceedings{antypas2023supertweeteval,
  title={SuperTweetEval: A Challenging, Unified and Heterogeneous Benchmark for Social Media NLP Research},
  author={Dimosthenis Antypas and Asahi Ushio and Francesco Barbieri and Leonardo Neves and Kiamehr Rezaee and Luis Espinosa-Anke and Jiaxin Pei and Jose Camacho-Collados},
  booktitle={Findings of the Association for Computational Linguistics: EMNLP 2023},
  year={2023}
}