This is the T5-3B model for System 2 as described in our paper Just-DREAM-about-it: Figurative Language Understanding with DREAM-FLUTE, FigLang workshop @ EMNLP 2022 (Arxiv link: https://arxiv.org/abs/2210.16407)
System 2: Jointly predicting the type of figurative language
Using type of figurative language provided as part of the training set (Chakrabarty et al., 2022), one of our models jointly predicts the type of figurative language, together with the target label and explanation:
We provide a quick example of how you can try out System 2 in our paper with just a few lines of code:
>>> from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
>>> model = AutoModelForSeq2SeqLM.from_pretrained("allenai/System2_FigLang2022")
>>> tokenizer = AutoTokenizer.from_pretrained("t5-3b")
>>> input_string = "Premise: Yesterday two gangs were fighting just in front of my home. Hypothesis: Yesterday I saw two gangs fighting right in front of my house and it totally didn't make me scared at all. What is the type of figurative language involved? Is there a contradiction or entailment between the premise and hypothesis?"
>>> input_ids = tokenizer.encode(input_string, return_tensors="pt")
>>> output = model.generate(input_ids, max_length=200)
>>> tokenizer.batch_decode(output, skip_special_tokens=True)
['Answer : [Type] Sarcasm [Label] Contradiction. Explanation : Seeing two gangs of people fighting in public can be really dangerous and scary, so someone who claims that they were not scared at all is being sarcastic.']
More details about DREAM-FLUTE ...
For more details about DREAM-FLUTE, please refer to our:
This model is part of our DREAM-series of works. This is a line of research where we make use of scene elaboration for building a "mental model" of situation given in text. Check out our GitHub Repo for more!
