DialogFake

Since we have extensively used various closed-source models for data synthesis, we are currently collecting and organizing the legal credentials related to data usage. Once compiled, we will open-source all data and testing code.