Think Before You Speak: Explicitly Generating Implicit Commonsense Knowledge for Response Generation

TL;DR

This paper introduces Think-Before-Speaking, a novel response generation method that explicitly externalizes and utilizes implicit commonsense knowledge, leading to more informative, relevant, and explainable conversational responses.

Contribution

The paper proposes a new generative approach that externalizes implicit knowledge before response generation, improving informativeness and explainability over existing models.

Findings

TBS models outperform baselines on automatic metrics.

Generated responses are more informative and commonsense-aware.

Knowledge relevance is about 85% according to human evaluation.

Abstract

Implicit knowledge, such as common sense, is key to fluid human conversations. Current neural response generation (RG) models are trained to generate responses directly, omitting unstated implicit knowledge. In this paper, we present Think-Before-Speaking (TBS), a generative approach to first externalize implicit commonsense knowledge (think) and use this knowledge to generate responses (speak). We expect that externalizing implicit knowledge allows more efficient learning, produces more informative responses, and enables more explainable models. We analyze different choices to collect knowledge-aligned dialogues, represent implicit knowledge, and transition between knowledge and dialogues. Empirical results show TBS models outperform end-to-end and knowledge-augmented RG baselines on most automatic metrics and generate more informative, specific, and commonsense-following responses, as evaluated by human annotators. TBS also generates knowledge that makes sense and is relevant to the dialogue around 85\% of the time.