AnyTOD: A Programmable Task-Oriented Dialog System

TL;DR

AnyTOD introduces a zero-shot, programmable task-oriented dialog system that leverages a neuro-symbolic approach to adapt to unseen tasks without task-specific training, significantly reducing data needs.

Contribution

It presents a novel neuro-symbolic framework for TOD that generalizes to unseen schemas and tasks, enabling zero-shot transfer and reducing training requirements.

Findings

Achieves state-of-the-art results on STAR, ABCD, and SGD benchmarks.

Demonstrates strong zero-shot transfer in low-resource settings like MultiWOZ.

Provides STARv2 dataset with richer annotations for benchmarking.

Abstract

We propose AnyTOD, an end-to-end, zero-shot task-oriented dialog (TOD) system capable of handling unseen tasks without task-specific training. 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 occurring during a conversation and a symbolic program implementing the dialog policy is executed to recommend next 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, ABCD and SGD benchmarks. We also demonstrate strong zero-shot transfer ability in low-resource settings, such as zero-shot on MultiWOZ. In addition, we release STARv2, an updated version of the STAR dataset with richer annotations, for benchmarking zero-shot end-to-end TOD models.