We present a novel response generation system that can be trained end to end on large quantities of unstructured Twitter conversations. A neural network architecture is used to address sparsity issues that arise when integrating contextual information into classic statistical models, allowing the system to take into account previous dialog utterances. Our dynamic-context generative models show consistent gains over both context-sensitive and non-context-sensitive Machine Translation and Information Retrieval baselines.
@article{arxiv.1506.06714,
title = {A Neural Network Approach to Context-Sensitive Generation of Conversational Responses},
author = {Alessandro Sordoni and Michel Galley and Michael Auli and Chris Brockett and Yangfeng Ji and Margaret Mitchell and Jian-Yun Nie and Jianfeng Gao and Bill Dolan},
journal= {arXiv preprint arXiv:1506.06714},
year = {2015}
}
Comments
A. Sordoni, M. Galley, M. Auli, C. Brockett, Y. Ji, M. Mitchell, J.-Y. Nie, J. Gao, B. Dolan. 2015. A Neural Network Approach to Context-Sensitive Generation of Conversational Responses. In Proc. of NAACL-HLT. Pages 196-205