We develop a framework to find modular internal representations in generative models and manipulate then to generate counterfactual examples.
確定! 回上一頁