CONF Mai_EMNLP2020_2020/IDIAP Plug and Play Autoencoders for Conditional Text Generation Mai, Florian Pappas, Nikolaos Montero, Ivan Smith, Noah A. Henderson, James EXTERNAL https://publications.idiap.ch/attachments/papers/2020/Mai_EMNLP2020_2020.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/Mai_Idiap-RR-24-2020 Related documents Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing Online 2020 REPORT Mai_Idiap-RR-24-2020/IDIAP Plug and Play Autoencoders for Conditional Text Generation Mai, Florian Pappas, Nikolaos Montero, Ivan Smith, Noah A. Henderson, James autoencoders Natural language processing style transfer EXTERNAL https://publications.idiap.ch/attachments/reports/2020/Mai_Idiap-RR-24-2020.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/Mai_EMNLP2020_2020 Related documents Idiap-RR-24-2020 2020 Idiap October 2020 Accepted at EMNLP 2020 Text autoencoders are commonly used for conditional generation tasks such as style transfer. We propose methods which are plug and play, where any pretrained autoencoder can be used, and only require learning a mapping within the autoencoder's embedding space, training embedding-to-embedding (Emb2Emb). This reduces the need for labeled training data for the task and makes the training procedure more efficient. Crucial to the success of this method is a loss term for keeping the mapped embedding on the manifold of the autoencoder and a mapping which is trained to navigate the manifold by learning offset vectors. Evaluations on style transfer tasks both with and without sequence-to-sequence supervision show that our method performs better than or comparable to strong baselines while being up to four times faster.