{"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/style-transfer-in-text-exploration-and","title":"Style Transfer in Text: Exploration and Evaluation","arxiv_id":"1711.06861","date":"2017-11-18","proceeding":null,"authors":["Zhenxin Fu","Xiaoye Tan","Nanyun Peng","Dongyan Zhao","Rui Yan"],"abstract":"Style transfer is an important problem in natural language processing (NLP).\nHowever, the progress in language style transfer is lagged behind other\ndomains, such as computer vision, mainly because of the lack of parallel data\nand principle evaluation metrics. In this paper, we propose to learn style\ntransfer with non-parallel data. We explore two models to achieve this goal,\nand the key idea behind the proposed models is to learn separate content\nrepresentations and style representations using adversarial networks. We also\npropose novel evaluation metrics which measure two aspects of style transfer:\ntransfer strength and content preservation. We access our models and the\nevaluation metrics on two tasks: paper-news title transfer, and\npositive-negative review transfer. Results show that the proposed content\npreservation metric is highly correlate to human judgments, and the proposed\nmodels are able to generate sentences with higher style transfer strength and\nsimilar content preservation score comparing to auto-encoder.","url_abs":"http://arxiv.org/abs/1711.06861v2","url_pdf":"http://arxiv.org/pdf/1711.06861v2.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":"style-transfer-in-text-exploration-and","repo_url":"https://github.com/fuzhenxin/text_style_transfer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"style-transfer-in-text-exploration-and","repo_url":"https://github.com/rj-IIITH18/style_transfer_in_text","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"style-transfer","task_name":"Style Transfer"},{"task_slug":"text-style-transfoer","task_name":"Text Style Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/text-style-transfer-on-yelp-review-dataset","task":"Text Style Transfer","dataset":"Yelp Review Dataset (Small)","model":"MultiDecoder","rank_in_archive_order":6,"of":8,"metrics":{"G-Score (BLEU, Accuracy)":"45.02"},"uses_additional_data":false},{"leaderboard":"/sota/text-style-transfer-on-yelp-review-dataset","task":"Text Style Transfer","dataset":"Yelp Review Dataset (Small)","model":"StyleEmbedding","rank_in_archive_order":8,"of":8,"metrics":{"G-Score (BLEU, Accuracy)":"31.31"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.06861","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}