{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deconvolutional-paragraph-representation","title":"Deconvolutional Paragraph Representation Learning","arxiv_id":"1708.04729","date":"2017-08-16","proceeding":"NeurIPS 2017 12","authors":["Yizhe Zhang","Dinghan Shen","Guoyin Wang","Zhe Gan","Ricardo Henao","Lawrence Carin"],"abstract":"Learning latent representations from long text sequences is an important\nfirst step in many natural language processing applications. Recurrent Neural\nNetworks (RNNs) have become a cornerstone for this challenging task. However,\nthe quality of sentences during RNN-based decoding (reconstruction) decreases\nwith the length of the text. We propose a sequence-to-sequence, purely\nconvolutional and deconvolutional autoencoding framework that is free of the\nabove issue, while also being computationally efficient. The proposed method is\nsimple, easy to implement and can be leveraged as a building block for many\napplications. We show empirically that compared to RNNs, our framework is\nbetter at reconstructing and correcting long paragraphs. Quantitative\nevaluation on semi-supervised text classification and summarization tasks\ndemonstrate the potential for better utilization of long unlabeled text data.","url_abs":"http://arxiv.org/abs/1708.04729v3","url_pdf":"http://arxiv.org/pdf/1708.04729v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deconvolutional-paragraph-representation","repo_url":"https://github.com/dreasysnail/deconv_paragraph_represention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"deconvolutional-paragraph-representation","repo_url":"https://github.com/dreasysnail/textGAN_public","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"deconvolutional-paragraph-representation","repo_url":"https://github.com/smalik169/recursive-convolutional-autoencoder","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deconvolutional-paragraph-representation","repo_url":"https://github.com/tuvuumass/SCoPE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"semi-supervised-text-classification-1","task_name":"Semi-Supervised Text Classification"},{"task_slug":"text-classification","task_name":"Text Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}