A Persona-Based Neural Conversation Model

TL;DR

This paper introduces persona-based neural conversation models that improve speaker consistency and response quality by encoding individual characteristics and interaction properties, showing better perplexity, BLEU scores, and human judgment.

Contribution

The paper proposes novel persona and dyadic models for neural response generation, enhancing speaker consistency and interaction modeling over standard sequence-to-sequence approaches.

Findings

Improved perplexity and BLEU scores over baselines.

Enhanced speaker consistency as rated by humans.

Qualitative performance gains in response quality.

Abstract

We present persona-based models for handling the issue of speaker consistency in neural response generation. A speaker model encodes personas in distributed embeddings that capture individual characteristics such as background information and speaking style. A dyadic speaker-addressee model captures properties of interactions between two interlocutors. Our models yield qualitative performance improvements in both perplexity and BLEU scores over baseline sequence-to-sequence models, with similar gains in speaker consistency as measured by human judges.