We propose simple and flexible training and decoding methods for influencing output style and topic in neural encoder-decoder based language generation. This capability is desirable in a variety of applications, including conversational systems, where successful agents need to produce language in a specific style and generate responses steered by a human puppeteer or external knowledge. We decompose the neural generation process into empirically easier sub-problems: a faithfulness model and a decoding method based on selective-sampling. We also describe training and sampling algorithms that bias the generation process with a specific language style restriction, or a topic restriction. Human evaluation results show that our proposed methods are able to restrict style and topic without degrading output quality in conversational tasks.
@article{arxiv.1709.03010,
title = {Steering Output Style and Topic in Neural Response Generation},
author = {Di Wang and Nebojsa Jojic and Chris Brockett and Eric Nyberg},
journal= {arXiv preprint arXiv:1709.03010},
year = {2017}
}