[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-1-en-105":3,"doc-seo-195220-105":53,"doc-detail-195220-en":126},{"code":4,"msg":5,"data":6},0,"success",[7,14,19,24,29,34,39,44,49],{"id":8,"doc_module":9,"doc_module_name":10,"category_name":11,"show_sort_weight":12,"slug":13},11,1,"Template","Presentations",90,"presentations",{"id":15,"doc_module":9,"doc_module_name":10,"category_name":16,"show_sort_weight":17,"slug":18},12,"Resumes",80,"resumes",{"id":20,"doc_module":9,"doc_module_name":10,"category_name":21,"show_sort_weight":22,"slug":23},14,"Invoices",70,"invoices",{"id":25,"doc_module":9,"doc_module_name":10,"category_name":26,"show_sort_weight":27,"slug":28},15,"Posters",60,"posters",{"id":30,"doc_module":9,"doc_module_name":10,"category_name":31,"show_sort_weight":32,"slug":33},16,"Social Media",50,"social-media",{"id":35,"doc_module":9,"doc_module_name":10,"category_name":36,"show_sort_weight":37,"slug":38},17,"Forms",40,"forms",{"id":40,"doc_module":9,"doc_module_name":10,"category_name":41,"show_sort_weight":42,"slug":43},18,"Letters",30,"letters",{"id":45,"doc_module":9,"doc_module_name":10,"category_name":46,"show_sort_weight":47,"slug":48},21,"Paper Templates",5,"papers-templates",{"id":50,"doc_module":9,"doc_module_name":10,"category_name":51,"show_sort_weight":4,"slug":52},158,"General","general-158",{"code":4,"msg":54,"data":55},"ok",{"site_id":56,"language":57,"slug":58,"title":59,"keywords":60,"description":61,"schema_data":62,"social_meta":119,"head_meta":121,"extra_data":123,"updated_unix":125},105,"en","text-style-transfer-a-review-and-experimental-evaluation","Text Style Transfer - A Review and Experimental Evaluation","","This document explores the field of Text Style Transfer (TST), providing a comprehensive review and experimental evaluation of existing techniques. It defines key terms and notations used in TST, such as source attribute value (s), target attribute value (t), and the set of attribute values (A). The paper discusses the encoder (E) and generator (G) components of TST models, along with the style classifier (D), and their respective parameters. Various data settings and strategies for content-style disentanglement are presented, including Parallel Supervised, Non-Parallel Supervised, and Purely Unsupervised approaches. Specific techniques like Sequence-to-Sequence, Explicit Style Keyword Replacement, Back-Translation, Adversarial Learning, Attribute Control Generation, Entangled Latent Representation Edition, and Reinforcement Learning are detailed within their respective data settings. The document also outlines an experimental setup involving an encoder and decoder for translating sentences from English to French and vice versa, highlighting the inputs and outputs of such a system. The overall aim is to provide a structured overview of TST methods and their experimental foundations.",{"@graph":63,"@context":118},[64,80,101],{"@type":65,"itemListElement":66},"BreadcrumbList",[67,71,74,77],{"item":68,"name":69,"@type":70,"position":9},"https://docshare.wps.com","Home","ListItem",{"item":72,"name":10,"@type":70,"position":73},"https://docshare.wps.com/template/",2,{"item":75,"name":51,"@type":70,"position":76},"https://docshare.wps.com/template/general/",3,{"item":78,"name":59,"@type":70,"position":79},"https://docshare.wps.com/template/text-style-transfer-a-review-and-experimental-evaluation/195220/",4,{"url":78,"name":59,"@type":81,"image":82,"author":87,"headline":59,"publisher":90,"fileFormat":93,"inLanguage":57,"description":61,"dateModified":94,"datePublished":95,"encodingFormat":93,"isAccessibleForFree":96,"interactionStatistic":97},"DigitalDocument",{"url":83,"@type":84,"width":85,"height":86},"https://docshare.wps.com/thumbnails/text-style-transfer-a-review-and-experimental-evaluation/195220.png","ImageObject",442,249,{"name":88,"@type":89},"Valentina","Person",{"url":68,"name":91,"@type":92},"DocShare","Organization","application/pdf","2026-09-27","2026-09-03",true,{"@type":98,"interactionType":99,"userInteractionCount":76},"InteractionCounter",{"@type":100},"ViewAction",{"@type":102,"mainEntity":103},"FAQPage",[104,110,114],{"name":105,"@type":106,"acceptedAnswer":107},"What are the core components of a Text Style Transfer (TST) model discussed in the document?","Question",{"text":108,"@type":109},"The core components of a TST model are the Encoder (E), the Generator (G), and the style classifier or Discriminator (D). Each component has associated parameters listed in the document.","Answer",{"name":111,"@type":106,"acceptedAnswer":112},"What are the main data settings for content-style disentanglement in TST?",{"text":113,"@type":109},"The main data settings are Parallel Supervised, Non-Parallel Supervised, and Purely Unsupervised. Each setting employs different strategies and techniques for separating content from style.",{"name":115,"@type":106,"acceptedAnswer":116},"What is an example of an experimental setup for TST mentioned in the document?",{"text":117,"@type":109},"An experimental setup involving an encoder and decoder for translating sentences between English and French is described, illustrating the input and output processes for TST.","https://schema.org",{"og:url":78,"og:type":120,"og:title":59,"og:site_name":91,"og:description":61},"article",{"robots":122,"canonical":78},"index,follow",{"doc_id":124,"site_id":56},195220,1788446460,{"code":4,"msg":5,"data":127},{"doc_id":124,"user_id":128,"nickname":88,"user_avatar":129,"doc_module":9,"category_id":50,"category_name":51,"doc_title":59,"doc_description":61,"doc_content":130,"file_id":131,"file_url":132,"file_type":133,"file_size":134,"view_count":76,"is_deleted":4,"is_public":9,"is_downloadable":9,"audit_status":9,"page_count":135,"language":136,"language_code":57,"site_id":56,"html_lang":57,"table_of_contents":137,"faqs":138,"seo_title":139,"seo_description":61,"update_tm":125,"read_time":8},13056703020460,"https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923","| Not. | Meaning |\n| --- | --- |\n| s | The source attribute value, e.g., the formal style |\n| t | The target attribute value different from s, e.g., the informal style |\n| A | A predefined set of attribute values, s, t ∈ A |\n| x | A sentence with the source attribute value s |\n| x′ | The transferred sentence of x with the target attribute value t |\n| X | The corpus of sentences with different attribute values |\n| E | Encoder of a TST model |\n| G | Generator of a TST model |\n| D | style classifier or discriminator |\n| 􀀂 E | Parameters of the encoder |\n| 􀀂 G | Parameters of the generator |\n| 􀀂 D | Parameters of the style classifier |\n| z | Latent representation of text, i.e. , z , E(x) |\n| a | Latent representation of the attribute value in text |\n\n| Data Setting | Strategy (Content-Style Disentanglement) | Technique | Literature |\n| --- | --- | --- | --- |\n| Parallel Supervised | - | Sequence-to-Sequence | [48, 10, 111, 130, 50, 89, 73, 140, 149] |\n| Non-Parallel Supervised | Explicit | Explicit Style Keyword Replacement | [72, 137, 148, 117, 132] |\n|  | Implicit | Back-Translation | [98, 150] |\n|  |  | Adversarial Learning | [112, 151, 27, 13, 76, 145, 152, 144, 65, 52, 96] |\n|  |  | Attribute Control Generation | [40, 122] |\n|  |  | Other Peculiar Methods | [15] |\n|  | Without | Attribute Control Generation | [67, 17, 147, 46, 153] |\n|  |  | Entangled Latent Representation Edition | [88, 138, 127, 74] |\n|  |  | Reinforcement Learning | [78, 31] |\n|  |  | Other Peculiar Methods | [35, 15, 120] |\n| Purely Unsupervised | - | Purely Unsupervised | [100, 138, 113, 19] |\n\n|  | | 107 􀀋(QJOLVK􀀐WR􀀐)UHQFK􀀌 |  | |  | | 107 􀀋)UHQFK􀀐WR􀀐(QJOLVK􀀌 | |\n| --- | --- | --- | --- | --- | --- | --- | --- | --- |\n| ,QSXWVHQWHQFH [ LQ\u003Cbr>(QJOLVK |  |  |  |  | 2XWSXWVHQWHQFH [ LQ\u003Cbr>)UHQFK |  |  |  |\n|  |  | (QFRGHU | 'HFRGHU |  |  |  | (QFRGHU |  |\n|  |  |  |  |  |  |  |  |  |","cbCainMbGVnyd2eS","https://ap.wps.com/l/cbCainMbGVnyd2eS","pdf",1244648,32,"English","# Text Style Transfer (TST)\n## Notation\n## Data Setting, Strategy, Technique, and Literature\n## Experimental Setup","[{\"question\":\"What are the core components of a Text Style Transfer (TST) model discussed in the document?\",\"answer\":\"The core components of a TST model are the Encoder (E), the Generator (G), and the style classifier or Discriminator (D). Each component has associated parameters listed in the document.\"},{\"question\":\"What are the main data settings for content-style disentanglement in TST?\",\"answer\":\"The main data settings are Parallel Supervised, Non-Parallel Supervised, and Purely Unsupervised. Each setting employs different strategies and techniques for separating content from style.\"},{\"question\":\"What is an example of an experimental setup for TST mentioned in the document?\",\"answer\":\"An experimental setup involving an encoder and decoder for translating sentences between English and French is described, illustrating the input and output processes for TST.\"}]","Text Style Transfer - A Review and Experimental Evaluation | PDF"]