A Syntactically Constrained Bidirectional-Asynchronous Approach for Emotional Conversation Generation

TL;DR

This paper introduces a novel bidirectional-asynchronous model for emotional conversation generation that incorporates emotion and topic keywords to produce more diverse, logical, and emotionally expressive replies.

Contribution

It proposes a syntactically constrained bidirectional-asynchronous decoding method that integrates emotion and topic keywords, improving response diversity and emotional expression.

Findings

Enhanced reply diversity over baseline models

Improved logical coherence in generated responses

Greater emotional expressiveness in outputs

Abstract

Traditional neural language models tend to generate generic replies with poor logic and no emotion. In this paper, a syntactically constrained bidirectional-asynchronous approach for emotional conversation generation (E-SCBA) is proposed to address this issue. In our model, pre-generated emotion keywords and topic keywords are asynchronously introduced into the process of decoding. It is much different from most existing methods which generate replies from the first word to the last. Through experiments, the results indicate that our approach not only improves the diversity of replies, but gains a boost on both logic and emotion compared with baselines.