A survey on XAI and natural language explanations

Erik Cambria, Lorenzo Malandri, Fabio Mercorio, Mario Mezzanzanica, Navid Nobani

Abstract: Highlights•Review of current XAI methods that provide explanations through natural language.•A novel set of hierarchical criteria (taxonomy) to compare the state of the art.•A multi-criteria decision-making model to compare XAI works according to user needs.•A ranking model, publicly available to the community to keep the survey updated.