Yongjie Wang

Exploring why AI model explanations are essential for trust and reliability.

PairCFR improves training models using counterfactual data for better performance.

A look into methods and challenges of generating counterfactuals in NLP.