Task2Dial: A Novel Task and Dataset for Commonsense enhanced Task-based Dialogue Grounded in Documents

TL;DR

This paper introduces Task2Dial, a new dataset and task for document-grounded task-based dialogue that incorporates commonsense knowledge, lexical richness, and planning, aiming to improve naturalness and complexity in dialogue systems.

Contribution

The paper presents Task2Dial, a novel dataset and task that challenge models with more natural, varied dialogues requiring commonsense reasoning and planning, advancing research in document-grounded dialogue systems.

Findings

Task2Dial dataset contains longer, more complex dialogues.

Generating responses requires paraphrasing and commonsense knowledge.

The dataset promotes development of more natural dialogue systems.

Abstract

This paper proposes a novel task on commonsense-enhanced task-based dialogue grounded in documents and describes the Task2Dial dataset, a novel dataset of document-grounded task-based dialogues, where an Information Giver (IG) provides instructions (by consulting a document) to an Information Follower (IF), so that the latter can successfully complete the task. In this unique setting, the IF can ask clarification questions which may not be grounded in the underlying document and require commonsense knowledge to be answered. The Task2Dial dataset poses new challenges: (1) its human reference texts show more lexical richness and variation than other document-grounded dialogue datasets; (2) generating from this set requires paraphrasing as instructional responses might have been modified from the underlying document; (3) requires commonsense knowledge, since questions might not necessarily be grounded in the document; (4) generating requires planning based on context, as task steps need to be provided in order. The Task2Dial dataset contains dialogues with an average $18.15$ number of turns and 19.79 tokens per turn, as compared to 12.94 and 12 respectively in existing datasets. As such, learning from this dataset promises more natural, varied and less template-like system utterances.