ANYTOD: A Programmable Task-Oriented Dialog System

Abstract

We propose ANYTOD, an end-to-end, zero-shot task-oriented dialog (TOD) system capable of zero-shot adaptation onto unseen tasks or domains. We view TOD as a program executed by a language model (LM), where program logic and ontology is provided by a designer as a schema. To enable generalization to unseen schemas and programs without prior training, ANYTOD adopts a neuro-symbolic approach. A neural LM keeps track of events that occur during a conversation, and a symbolic program implementing dialog policy is executed to recommend actions ANYTOD should take. This approach drastically reduces data annotation and model training requirements, addressing the enduring challenge of rapidly adapting a TOD system to unseen tasks and domains. We demonstrate state-of-the-art results on STAR (Mehri and Eskenazi, 2021), ABCD (Chen et al., 2021) and SGD (Rastogi et al., 2020) benchmarks. We also demonstrate strong zero-shot transfer ability in low-resource settings, such as zero-shot transfer onto MultiWOZ (Budzianowski et al., 2018a). In addition, we release STARV2, an updated version of the STAR dataset with richer annotations, for benchmarking zero-shot task transfer for end-to-end TOD models.