A Persona-Based Neural Conversation Model
This addresses the issue of inconsistent speaker behavior in conversational AI, which is an incremental improvement over existing sequence-to-sequence models.
The paper tackled the problem of speaker consistency in neural response generation by introducing persona-based models that encode individual characteristics and interaction properties, resulting in improved perplexity and BLEU scores over baseline models.
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.