I am using Semantic Role Labeling to get the predicate and argument for the given sentence using BERT pre-trained model. But along with that I also need sense of the predicates used in sentence. This is similar to the problem of ‘Word Sense disambiguation’, a same word can have different meaning based on the structure of sentence. According to the paper used for this SRL module a Bert based model , it mentioned that its predicting and using the sense. While going through the your GitHub repository, we are not able to extract word sense using SRL.
Is it feasible to extract the word sense from other modules/model of Allennlp? Or we have missed something that we can try?