Transforming Slot Schema Induction with Generative Dialogue State Inference

TL;DR

This paper introduces a generative approach to Slot Schema Induction that automatically discovers dialogue state representations from unlabeled data, outperforming previous clustering methods on benchmark datasets.

Contribution

The paper proposes a novel generative method for SSI that does not rely on prior task knowledge and produces high-quality slot-value candidates aligned with human schemas.

Findings

Outperforms previous state-of-the-art on MultiWOZ and SGD datasets.

Generates high-quality slot-value candidates without prior task knowledge.

Effective clustering of generated candidates into human-aligned schemas.

Abstract

The challenge of defining a slot schema to represent the state of a task-oriented dialogue system is addressed by Slot Schema Induction (SSI), which aims to automatically induce slots from unlabeled dialogue data. Whereas previous approaches induce slots by clustering value spans extracted directly from the dialogue text, we demonstrate the power of discovering slots using a generative approach. By training a model to generate slot names and values that summarize key dialogue information with no prior task knowledge, our SSI method discovers high-quality candidate information for representing dialogue state. These discovered slot-value candidates can be easily clustered into unified slot schemas that align well with human-authored schemas. Experimental comparisons on the MultiWOZ and SGD datasets demonstrate that Generative Dialogue State Inference (GenDSI) outperforms the previous state-of-the-art on multiple aspects of the SSI task.