Commonsense-Focused Dialogues for Response Generation: An Empirical Study

TL;DR

This paper empirically investigates how dialogue response generation can incorporate commonsense reasoning by creating new datasets and proposing evaluation methods, leading to more socially aware and sensible conversational models.

Contribution

It introduces a new dataset of commonsense dialogues and evaluates models trained on these datasets, demonstrating improved commonsense response generation.

Findings

Models trained on combined datasets produce more commonsense responses.

The proposed automatic evaluation correlates well with human judgments.

Collected datasets facilitate research on social commonsense in dialogue.

Abstract

Smooth and effective communication requires the ability to perform latent or explicit commonsense inference. Prior commonsense reasoning benchmarks (such as SocialIQA and CommonsenseQA) mainly focus on the discriminative task of choosing the right answer from a set of candidates, and do not involve interactive language generation as in dialogue. Moreover, existing dialogue datasets do not explicitly focus on exhibiting commonsense as a facet. In this paper, we present an empirical study of commonsense in dialogue response generation. We first auto-extract commonsensical dialogues from existing dialogue datasets by leveraging ConceptNet, a commonsense knowledge graph. Furthermore, building on social contexts/situations in SocialIQA, we collect a new dialogue dataset with 25K dialogues aimed at exhibiting social commonsense in an interactive setting. We evaluate response generation models trained using these datasets and find that models trained on both extracted and our collected data produce responses that consistently exhibit more commonsense than baselines. Finally we propose an approach for automatic evaluation of commonsense that relies on features derived from ConceptNet and pre-trained language and dialog models, and show reasonable correlation with human evaluation of responses' commonsense quality. We are releasing a subset of our collected data, Commonsense-Dialogues, containing about 11K dialogs.