[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-169391-105":59,"doc-detail-169391-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","document-metadata-extraction-multimodal","Document Metadata Extraction (Multimodal)","","This document outlines a critical process for extracting metadata from various sources, including text and images, using a multimodal approach. The primary goal is to generate structured metadata in a raw JSON format, adhering to strict formatting and content rules. The system is designed to analyze both textual content and visual information from images to create comprehensive metadata. Key instructions include extracting readable text from images, combining it with document text, and generating specific metadata fields such as title, language, abstract, keywords, category, table of contents, and frequently asked questions (FAQs). The system prioritizes language detection based on complete sentences and mandates language consistency across all generated text fields. Specific rules are provided for title generation, including handling orphan serials, file names, and document titles, with a strict character limit and formatting requirements. The abstract generation emphasizes directness, a third-person objective tone, and adherence to length constraints, focusing on descriptive precision rather than filler. Keywords are expected to be the most relevant terms, limited to five. The table of contents should be a markdown-formatted outline with at most two levels, inferring structure if explicit headings are absent. Finally, FAQs are to be generated in an array of objects, with each question addressing a distinct aspect of the document and answers being concise and textually grounded. The category classification relies on a provided JSON definition, with a fallback to 'General' if no specific category fits with high confidence.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":57,"@type":76,"position":81},"https://docshare.wps.com/document/general/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/document-metadata-extraction-multimodal/169391/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/document-metadata-extraction-multimodal/169391.png","ImageObject",300,407,{"name":92,"@type":93},"Skyler","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-21","2026-09-01",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",11,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What is the primary output format for the extracted metadata?","Question",{"text":113,"@type":114},"The extracted metadata must be in a single raw JSON object, starting with '{' and ending with '}' without any additional characters or formatting.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How is the document's language determined?",{"text":118,"@type":114},"Language is detected based on meaningful natural language sentences, prioritizing body text and image text. The dominant language is determined by the majority of full sentences, with a default to 'en' if no confident result is obtained.",{"name":120,"@type":111,"acceptedAnswer":121},"What are the constraints on the generated title?",{"text":122,"@type":114},"The final title must be 100 characters or fewer, use only the document's detected original language, and follow specific formatting rules including hyphenation for multi-component titles and preservation of feature identifiers like volume or chapter numbers.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},169391,1790018909,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":56,"category_name":57,"doc_title":65,"doc_description":67,"doc_content":66,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145},2336464648746,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","cbCaimd4V3v7qyTb","https://ap.wps.com/l/cbCaimd4V3v7qyTb","pdf",80495,14,"English","# Document Metadata Extraction (Multimodal)\n## Instructions\n## Language Rules\n## Field Specifications\n### Title Rules\n### `title_en`\n### `language`\n### abstract\n### `keywords`\n### `toc`\n### `faqs`\n### `category`\n## Output Requirements (STRICT)","[{\"question\":\"What is the primary output format for the extracted metadata?\",\"answer\":\"The extracted metadata must be in a single raw JSON object, starting with '{' and ending with '}' without any additional characters or formatting.\"},{\"question\":\"How is the document's language determined?\",\"answer\":\"Language is detected based on meaningful natural language sentences, prioritizing body text and image text. The dominant language is determined by the majority of full sentences, with a default to 'en' if no confident result is obtained.\"},{\"question\":\"What are the constraints on the generated title?\",\"answer\":\"The final title must be 100 characters or fewer, use only the document's detected original language, and follow specific formatting rules including hyphenation for multi-component titles and preservation of feature identifiers like volume or chapter numbers.\"}]","Document Metadata Extraction (Multimodal) | PDF",1788253334,22]