Better Conversations by Modeling,Filtering,and Optimizing for Coherence and Diversity

TL;DR

This paper enhances open-domain conversational models by introducing coherence measures, filtering training data for topical relevance, and employing a variational autoencoder to improve response coherence and diversity, resulting in significant performance gains.

Contribution

The paper proposes a novel approach combining coherence measurement, data filtering, and a variational autoencoder to improve dialogue response quality.

Findings

Improved BLEU scores over baseline models

Enhanced coherence and diversity metrics

Effective use of coherence as a latent variable

Abstract

We present three enhancements to existing encoder-decoder models for open-domain conversational agents, aimed at effectively modeling coherence and promoting output diversity: (1) We introduce a measure of coherence as the GloVe embedding similarity between the dialogue context and the generated response, (2) we filter our training corpora based on the measure of coherence to obtain topically coherent and lexically diverse context-response pairs, (3) we then train a response generator using a conditional variational autoencoder model that incorporates the measure of coherence as a latent variable and uses a context gate to guarantee topical consistency with the context and promote lexical diversity. Experiments on the OpenSubtitles corpus show a substantial improvement over competitive neural models in terms of BLEU score as well as metrics of coherence and diversity.