THINK: A Novel Conversation Model for Generating Grammatically Correct and Coherent Responses

TL;DR

The paper introduces THINK, a conversation model that enhances response coherence and grammatical correctness by simplifying context vectors and strengthening the decoder, outperforming existing models in automatic and human evaluations.

Contribution

The novel THINK model improves response coherence and grammaticality by modifying the encoder-decoder framework with a teamwork generation approach and semantics extractor.

Findings

THINK produces more coherent responses than baselines.

Automatic and human evaluations favor THINK's responses.

The model reduces duplicated and self-contradicting responses.

Abstract

Many existing conversation models that are based on the encoder-decoder framework have focused on ways to make the encoder more complicated to enrich the context vectors so as to increase the diversity and informativeness of generated responses. However, these approaches face two problems. First, the decoder is too simple to effectively utilize the previously generated information and tends to generate duplicated and self-contradicting responses. Second, the complex encoder tends to generate diverse but incoherent responses because the complex context vectors may deviate from the original semantics of context. In this work, we proposed a conversation model named "THINK" (Teamwork generation Hover around Impressive Noticeable Keywords) to make the decoder more complicated and avoid generating duplicated and self-contradicting responses. The model simplifies the context vectors and increases the coherence of generated responses in a reasonable way. For this model, we propose Teamwork generation framework and Semantics Extractor. Compared with other baselines, both automatic and human evaluation showed the advantages of our model.