Predictions of BERT-large-cased. “Subject” and “Object” are from T-REx (Elsahar et al., 2018). “Pattern #i” / “Pred #i”: three different patterns from our resource and their predictions. The predictions are colored in blue if the model predicted correctly (out of the candidate list), and in red otherwise. If there is more than a single erroneous prediction, it is colored by a different red.
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.