CS-UM6P at SemEval-2021 Task 7: Deep Multi-Task Learning Model for Detecting and Rating Humor and Offense

25421

Knowledge Triples