PrimeQA/

tydiqa-boolean-answer-classifier

An answer classification model for boolean questions based on XLM-RoBERTa.

The answer classifier takes as input a boolean question and a passage, and returns a label (yes, no-answer, no).

The model was initialized with xlm-roberta-large and fine-tuned on the boolean questions from TyDiQA, as well as BoolQ-X.

You can use the raw model for question classification. Biases associated with the pre-existing language model, xlm-roberta-large, may be present in our fine-tuned model, tydiqa-boolean-answer-classifier.

You can use this model directly in the the PrimeQA framework for supporting boolean questions in reading comprehension: examples.