LSTM based Conversation Models

TL;DR

This paper introduces an LSTM-based conversational model that incorporates context and participant roles, improving response quality and capturing role-specific characteristics in two-party dialogues.

Contribution

The paper proposes novel architectures for integrating participant roles and context into LSTM models, enhancing conversational modeling capabilities.

Findings

Outperforms traditional LSTM in perplexity and response ranking

Captures role-specific differences in generated responses

Effectively models multi-turn interactions in dialogues

Abstract

In this paper, we present a conversational model that incorporates both context and participant role for two-party conversations. Different architectures are explored for integrating participant role and context information into a Long Short-term Memory (LSTM) language model. The conversational model can function as a language model or a language generation model. Experiments on the Ubuntu Dialog Corpus show that our model can capture multiple turn interaction between participants. The proposed method outperforms a traditional LSTM model as measured by language model perplexity and response ranking. Generated responses show characteristic differences between the two participant roles.