Improving Generalizability in Implicitly Abusive Language Detection with Concept Activation Vectors

Isar Nejadgholi, Kathleen C. Fraser, Svetlana Kiritchenko