Conversational Semantic Parsing

TL;DR

This paper introduces a new semantic representation and dataset for session-based conversational parsing, enabling better understanding of co-reference and context in task-oriented dialogue systems.

Contribution

It proposes a novel session-based semantic representation, releases a large dataset, and develops Seq2Seq models that outperform existing methods on multiple benchmarks.

Findings

Improved state-of-the-art results on DSTC2 by up to 5 points.

Achieved comparable performance on ATIS, SNIPS, TOP datasets.

Demonstrated effectiveness of the new semantic representation and models.

Abstract

The structured representation for semantic parsing in task-oriented assistant systems is geared towards simple understanding of one-turn queries. Due to the limitations of the representation, the session-based properties such as co-reference resolution and context carryover are processed downstream in a pipelined system. In this paper, we propose a semantic representation for such task-oriented conversational systems that can represent concepts such as co-reference and context carryover, enabling comprehensive understanding of queries in a session. We release a new session-based, compositional task-oriented parsing dataset of 20k sessions consisting of 60k utterances. Unlike Dialog State Tracking Challenges, the queries in the dataset have compositional forms. We propose a new family of Seq2Seq models for the session-based parsing above, which achieve better or comparable performance to the current state-of-the-art on ATIS, SNIPS, TOP and DSTC2. Notably, we improve the best known results on DSTC2 by up to 5 points for slot-carryover.