Steering Output Style and Topic in Neural Response Generation
This work addresses the need for conversational systems to produce language in specific styles and topics, but it is incremental as it builds on existing neural encoder-decoder frameworks.
The authors tackled the problem of controlling output style and topic in neural response generation by proposing simple training and decoding methods, achieving results that restrict style and topic without degrading output quality in conversational tasks.
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.