AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Importance of Reading Completion Models
The Dr. QA paper also shows the possibility to try reading comprehension models based on only the question answering, question answer pairs by using the law. I think that's actually a key idea to get things working for other data sets. So then from there, would you say that like it was kind of these large-scale BERT type models that just generally improved performance and then going from the kind of simple TFIDF to a neural model as well?