Recent pre-trained language models have shown promising capabilities in generating fluent and realistic natural language text. However, generating multi-sentence text with global content planning has been a long-existing research question. Current approaches for controlled text generation can hardly address this issue, as they usually condition on single known control attributes. In this study, we propose a low-cost yet effective framework which explicitly models the global content plan of the generated text. Specifically, it optimizes the joint distribution of the natural language sequence and the global content plan in a plug-and-play manner. We conduct extensive experiments on the well-established Recipe1M+ benchmark. Both automatic and human evaluations verify that our model achieves the state-of-the-art performance on the task of recipe generation
@article{arxiv.2212.05093,
title = {Plug-and-Play Recipe Generation with Content Planning},
author = {Yinhong Liu and Yixuan Su and Ehsan Shareghi and Nigel Collier},
journal= {arXiv preprint arXiv:2212.05093},
year = {2022}
}