CONF Mai_AACL_2022/IDIAP Bag-of-Vectors Autoencoders for Unsupervised Conditional Text Generation Mai, Florian Henderson, James https://publications.idiap.ch/index.php/publications/showcite/Mai_Idiap-RR-21-2021 Related documents Association for Computational Linguistics - Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) Online 2022 468–488 https://aclanthology.org/2022.aacl-main.36/ URL Text autoencoders are often used for unsupervised conditional text generation by applying mappings in the latent space to change attributes to the desired values. Recently, Mai et al. (2020) proposed Emb2Emb, a method to learn these mappings in the embedding space of an autoencoder. However, their method is restricted to autoencoders with a single-vector embedding, which limits how much information can be retained. We address this issue by extending their method to Bag-of-Vectors Autoencoders (BoV-AEs), which encode the text into a variable-size bag of vectors that grows with the size of the text, as in attention-based models. This allows to encode and reconstruct much longer texts than standard autoencoders. Analogous to conventional autoencoders, we propose regularization techniques that facilitate learning meaningful operations in the latent space. Finally, we adapt Emb2Emb for a training scheme that learns to map an input bag to an output bag, including a novel loss function and neural architecture. Our empirical evaluations on unsupervised sentiment transfer show that our method performs substantially better than a standard autoencoder. REPORT Mai_Idiap-RR-21-2021/IDIAP Bag-of-Vectors Autoencoders for Unsupervised Conditional Text Generation Mai, Florian Henderson, James autoencoders latent space learning Natural language processing variable-size Idiap-RR-21-2021 2021 Idiap December 2021 under review at ICLR 2022 Text autoencoders are often used for unsupervised conditional text generation by applying mappings in the latent space to change attributes to the desired values. Recently, Mai et al. (2020) proposed Emb2Emb, a method to learn these mappings in the embedding space of an autoencoder. However, their method is restricted to autoencoders with a single-vector embedding, which limits how much information can be retained. We address this issue by extending their method to Bag-of-Vectors Autoencoders (BoV-AEs), which encode the text into a variable-size bag of vectors that grows with the size of the text, as in attention-based models. This allows to encode and reconstruct much longer texts than standard autoencoders. Analogous to conventional autoencoders, we propose regularization techniques that facilitate learning meaningful operations in the latent space. Finally, we adapt Emb2Emb for a training scheme that learns to map an input bag to an output bag, including a novel loss function and neural architecture. Our experimental evaluations on unsupervised sentiment transfer and sentence summarization show that our method performs substantially better than a standard autoencoder.