LSTM based Conversation Models

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) LLM. The conversational model can function as a LLM 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 LLM perplexity and response ranking. Generated responses show characteristic differences between the two participant roles.