{"id":91456,"date":"2024-10-10T08:30:00","date_gmt":"2024-10-10T06:30:00","guid":{"rendered":"https:\/\/www.cocus.com\/?p=91456"},"modified":"2024-10-10T09:33:39","modified_gmt":"2024-10-10T07:33:39","slug":"ai-in-companies-rag-in-complex-data-environments","status":"publish","type":"post","link":"https:\/\/www.cocus.com\/en\/ai-in-companies-rag-in-complex-data-environments\/","title":{"rendered":"Rethinking RAG: A fresh approach to leveraging context for AI-powered solutions"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"91456\" class=\"elementor elementor-91456 elementor-91167\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-1e0e333 elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"1e0e333\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-1cb3cf2 lottie-bg-no\" data-id=\"1cb3cf2\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-9024e63 elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"9024e63\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-7893620 lottie-bg-no\" data-id=\"7893620\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-aa1ad2e elementor-widget elementor-widget-heading\" data-id=\"aa1ad2e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h6 class=\"elementor-heading-title elementor-size-default\">Guest contribution by<\/h6>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-c0a2d0f elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"c0a2d0f\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-c26a486 lottie-bg-no\" data-id=\"c26a486\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-1ff2643 elementor-widget elementor-widget-image\" data-id=\"1ff2643\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" src=\"https:\/\/www.cocus.com\/wp-content\/uploads\/elementor\/thumbs\/Image_Capa@2x-qivtxnkioovvu4zrtdsw0jolwtf85b611gvf6f2ibs.png\" title=\"Francisco Capa\" alt=\"Francisco Capa\" loading=\"lazy\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-f0549a5 lottie-bg-no\" data-id=\"f0549a5\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c3715c5 elementor-widget elementor-widget-heading\" data-id=\"c3715c5\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h6 class=\"elementor-heading-title elementor-size-default\">Francisco Capa, Data Enginer COCUS<\/h6>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ae9c513 elementor-icon-list--layout-inline elementor-list-item-link-full_width elementor-widget elementor-widget-icon-list\" data-id=\"ae9c513\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"icon-list.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<ul class=\"elementor-icon-list-items elementor-inline-items\">\n\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item elementor-inline-item\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/www.linkedin.com\/in\/francisco-capa-424087154\/\" target=\"_blank\" rel=\"noopener\">\n\n\t\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fab-linkedin-in\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M100.28 448H7.4V148.9h92.88zM53.79 108.1C24.09 108.1 0 83.5 0 53.8a53.79 53.79 0 0 1 107.58 0c0 29.7-24.1 54.3-53.79 54.3zM447.9 448h-92.68V302.4c0-34.7-.7-79.2-48.29-79.2-48.29 0-55.69 37.7-55.69 76.7V448h-92.78V148.9h89.08v40.8h1.3c12.4-23.5 42.69-48.3 87.88-48.3 94 0 111.28 61.9 111.28 142.3V448z\"><\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\"><\/span>\n\t\t\t\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-9b003f1 lottie-bg-no\" data-id=\"9b003f1\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap\">\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<div class=\"elementor-element elementor-element-fd4942e elementor-widget elementor-widget-text-editor\" data-id=\"fd4942e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>By rethinking how we apply Retrieval-Augmented Generation (RAG), businesses can unlock new levels of AI integration that are both practical and powerful. The concept of RAG holds significant potential for simplicity, accuracy and cost efficiency in AI-powered solutions and maximizes the value of existing data. Here you can find out about the attractive options RAG offers for a wide range of applications.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-44f35b2 elementor-widget elementor-widget-heading\" data-id=\"44f35b2\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">What is Retrieval-Augmented Generation (RAG)?<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-026524e elementor-widget elementor-widget-text-editor\" data-id=\"026524e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Retrieval-Augmented Generation is a concept that many of us have encountered without even realizing we are using it. If you ever used a Large Language Model (LLM) like ChatGPT or Gemini and provided it with specific context before asking a question, you\u2019ve used a form of RAG. At its core, RAG involves supplying context or additional information to an LLM to guide and refine its responses.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-3384b2a elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"3384b2a\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-83b75bb lottie-bg-no\" data-id=\"83b75bb\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-294d3e7 elementor-widget elementor-widget-text-editor\" data-id=\"294d3e7\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>As an example, consider the following simple prompt using ChatGPT 3.5 as the LLM for testing this concept: \u201cWrite a small paragraph about quantum physics.\u201d. The answers will be limited to 50 words for simplicity. The model gives us a standard, informative response as expected.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-6cdb249 lottie-bg-no\" data-id=\"6cdb249\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;background_background&quot;:&quot;classic&quot;,&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-508a5da elementor-widget elementor-widget-text-editor\" data-id=\"508a5da\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<h5>\u201eQuantum physics is the branch of science that studies the smallest particles in the universe, like atoms and photons. It reveals how these particles behave in surprising ways, differing from the laws of classical physics. Quantum mechanics underlies the fundamental principles of how particles move, interact, and give rise to the properties of matter, highlighting phenomena such as superposition and entanglement.\u201c<\/h5>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-2120c9c elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"2120c9c\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-a60a1c9 lottie-bg-no\" data-id=\"a60a1c9\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3359756 elementor-widget elementor-widget-text-editor\" data-id=\"3359756\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>But what if the context changes? Let\u2019s now ask the same question with the following context: \u201cThe audience are children\u201d. The LLM\u2019s response is now slightly different, adapting the answer to the new context while trying to keep the information accurate and relevant.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-e0280d6 lottie-bg-no\" data-id=\"e0280d6\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;background_background&quot;:&quot;classic&quot;,&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-dfab69c elementor-widget elementor-widget-text-editor\" data-id=\"dfab69c\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<h5>\u201eQuantum physics is like a magical rulebook for the tiniest things in the universe, such as atoms and light particles. It tells us that these tiny bits can do surprising tricks, like being in two places at once or talking to each other instantly over long distances, making the world a mysterious place!\u201c<\/h5>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<div class=\"elementor-element elementor-element-925aa3b elementor-widget elementor-widget-text-editor\" data-id=\"925aa3b\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>This adaptability highlights the power of RAG, and it\u2019s not limited to just text. This works in pretty much every type of LLM from text-to-text to text-to-code and even text-to-image. <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-b9ec376 elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"b9ec376\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-33 elementor-inner-column elementor-element elementor-element-b2840d5 lottie-bg-no\" data-id=\"b2840d5\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-f0989b2 elementor-widget elementor-widget-text-editor\" data-id=\"f0989b2\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>As an example, imagine using an AI image generator like DALL-E, where you upload a picture of your dog and ask it to place it in a different scenario. The resulting image is not perfect, but it will still recognize the breed and colour, giving context to the image you want to generate. So, context can be provided in a multitude of ways.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-33 elementor-inner-column elementor-element elementor-element-c2dffb9 lottie-bg-no\" data-id=\"c2dffb9\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-dadb554 elementor-widget elementor-widget-image\" data-id=\"dadb554\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img fetchpriority=\"high\" decoding=\"async\" width=\"484\" height=\"365\" src=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/RAG_Dog1.jpg\" class=\"attachment-full size-full wp-image-91176\" alt=\"Original image of a dog on a sofa.\" srcset=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/RAG_Dog1.jpg 484w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/RAG_Dog1-300x226.jpg 300w\" sizes=\"(max-width: 484px) 100vw, 484px\" title=\"\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-33 elementor-inner-column elementor-element elementor-element-066b2b8 lottie-bg-no\" data-id=\"066b2b8\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-e0a264a elementor-widget elementor-widget-image\" data-id=\"e0a264a\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"484\" height=\"365\" src=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/RAG_Dog2.jpg\" class=\"attachment-full size-full wp-image-91178\" alt=\"Generated image of the dog in an autumnal landscape.\" srcset=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/RAG_Dog2.jpg 484w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/RAG_Dog2-300x226.jpg 300w\" sizes=\"(max-width: 484px) 100vw, 484px\" title=\"\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<div class=\"elementor-element elementor-element-3276b3f elementor-widget elementor-widget-heading\" data-id=\"3276b3f\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">The Real-World Benefits of RAG in Business<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d820b20 elementor-widget elementor-widget-text-editor\" data-id=\"d820b20\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>The concept of RAG extends beyond playful interactions with AI. When applied for AI in companies, it can be used to unlock significant advantages: <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1fe6e3d elementor-align-start elementor-icon-list--layout-traditional elementor-list-item-link-full_width elementor-widget elementor-widget-icon-list\" data-id=\"1fe6e3d\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"icon-list.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<ul class=\"elementor-icon-list-items\">\n\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-circle\" viewBox=\"0 0 512 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8z\"><\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\"><b>Up-to-Date and Accurate Responses: <\/b>Language models, like GPT-3.5, are trained on data up until a certain point\u2014in this case, 2023. By providing real-time data or additional context from external sources, businesses can ensure that the model\u2019s responses are not only accurate but also current. <\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-circle\" viewBox=\"0 0 512 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8z\"><\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\"><b>Reducing Hallucinations: <\/b>Hallucinations, or instances where the model generates incorrect or nonsensical information, can be a concern. RAG helps to \"ground\" the model by providing focused context, thereby reducing the likelihood of these errors. <\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-circle\" viewBox=\"0 0 512 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8z\"><\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\"><b>Domain-Specific, Relevant Responses: <\/b>By supplying proprietary or domain-specific data as context, businesses can tailor the model\u2019s output to be more relevant and aligned with their specific needs.<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t<li class=\"elementor-icon-list-item\">\n\t\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-icon\">\n\t\t\t\t\t\t\t<svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-circle\" viewBox=\"0 0 512 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8z\"><\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t<span class=\"elementor-icon-list-text\"><b>Efficiency and Cost-Effectiveness: <\/b>Compared to more resource-intensive methods like retraining or fine-tuning the entire model, RAG offers a simpler and more cost-effective solution for customizing language models.<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-28f2da5 elementor-widget elementor-widget-text-editor\" data-id=\"28f2da5\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>While retraining or fine-tuning the base model are valid approaches depending on the use case, they are not mutually exclusive with RAG. In fact, combining these strategies or choosing the most appropriate one based on the use case can lead to optimal results. <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2e190f1 elementor-widget elementor-widget-heading\" data-id=\"2e190f1\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">A Practical Example for AI in Companies: Customer Service<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-cecf459 elementor-widget elementor-widget-text-editor\" data-id=\"cecf459\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>To illustrate the practical application of RAG, let\u2019s consider a customer service scenario, keeping in mind that the example is very simple, but the approach could be used in a much more complex data model logic. Imagine a customer asking a language model when their next payment is due (this can be done via a simple text chat or a more complex voice-to-text system). How would the model know this?  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2bc428d elementor-widget elementor-widget-text-editor\" data-id=\"2bc428d\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>One option is to pre-train the model on every customer\u2019s contract information. However, this would require frequent retraining and could become expensive very fast. A more efficient approach is to use RAG \u2013 provide the model with the relevant contract details stored in an external source (e.g., a SQL database) on-the-fly, enabling it to answer the question accurately. However, this approach presents a challenge: How do we determine which specific data the model needs?<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4672033 elementor-widget elementor-widget-heading\" data-id=\"4672033\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">A New Approach: Layered RAG<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-8981e30 elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"8981e30\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-6fa6aa1 lottie-bg-no\" data-id=\"6fa6aa1\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-dff7e0a elementor-widget elementor-widget-text-editor\" data-id=\"dff7e0a\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>The typical solution to this problem involves using semantic search to filter the data before feeding it to the model. This process involves converting both the data points and the query into model embeddings, followed by a semantic search to identify the most relevant information. While effective, this method can become a bottleneck, especially when dealing with structured data that is easily filtered because of the format nature. Here\u2019s where a different approach comes into play. A layered RAG strategy.   <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-e292752 lottie-bg-no\" data-id=\"e292752\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-4e33d49 elementor-widget elementor-widget-image\" data-id=\"4e33d49\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell1.jpg\" data-elementor-open-lightbox=\"yes\" data-elementor-lightbox-title=\"RAG\" data-e-action-hash=\"#elementor-action%3Aaction%3Dlightbox%26settings%3DeyJpZCI6OTEyMDcsInVybCI6Imh0dHBzOlwvXC93d3cuY29jdXMuY29tXC93cC1jb250ZW50XC91cGxvYWRzXC8yMDI0XC8wOVwvQmxvZ19SQUdfTW9kZWxsMS5qcGcifQ%3D%3D\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"1454\" height=\"439\" src=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell1.jpg\" class=\"attachment-full size-full wp-image-91207\" alt=\"RAG Model 1: The process begins with a user entering a question. This question is then embedded and compared with external data stored in a vector database. The relevant data is retrieved through semantic search. The question and relevant data are then fed into a model to generate an answer, which is ultimately presented to the user.\" srcset=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell1.jpg 1454w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell1-300x91.jpg 300w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell1-1024x309.jpg 1024w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell1-768x232.jpg 768w\" sizes=\"(max-width: 1454px) 100vw, 1454px\" title=\"\">\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-inner-section elementor-element elementor-element-987edcc elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"987edcc\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-854ca74 lottie-bg-no\" data-id=\"854ca74\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-793c9b2 elementor-widget elementor-widget-image\" data-id=\"793c9b2\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell2.jpg\" data-elementor-open-lightbox=\"yes\" data-elementor-lightbox-title=\"RAG\" data-e-action-hash=\"#elementor-action%3Aaction%3Dlightbox%26settings%3DeyJpZCI6OTEyMDUsInVybCI6Imh0dHBzOlwvXC93d3cuY29jdXMuY29tXC93cC1jb250ZW50XC91cGxvYWRzXC8yMDI0XC8wOVwvQmxvZ19SQUdfTW9kZWxsMi5qcGcifQ%3D%3D\">\n\t\t\t\t\t\t\t<img loading=\"lazy\" decoding=\"async\" width=\"1503\" height=\"505\" src=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell2.jpg\" class=\"attachment-full size-full wp-image-91205\" alt=\"RAG Model 2: A person asks the model a question. The model is based on a data schema, filters the relevant data from external data, and combines it with the question. The model then gives the person an answer.\" srcset=\"https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell2.jpg 1503w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell2-300x101.jpg 300w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell2-1024x344.jpg 1024w, https:\/\/www.cocus.com\/wp-content\/uploads\/2024\/09\/Blog_RAG_Modell2-768x258.jpg 768w\" sizes=\"(max-width: 1503px) 100vw, 1503px\" title=\"\">\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-50 elementor-inner-column elementor-element elementor-element-2a1ecbd lottie-bg-no\" data-id=\"2a1ecbd\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-108cdb0 elementor-widget elementor-widget-text-editor\" data-id=\"108cdb0\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Instead of relying solely on semantic search, we can eliminate this step by directly leveraging the structured schema of our data. By providing the schema as context to the LLM, we can ask it to identify the necessary data points needed to answer the query. Once identified, we simply query our external data sources, supply the relevant context, and allow the model to generate the final response. So, in a way we are using RAG to feed the next level of RAG.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<div class=\"elementor-element elementor-element-7e5a80f elementor-widget elementor-widget-text-editor\" data-id=\"7e5a80f\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>This should only be an option where the data structure is known prior to the question, this would not work if the context is images or documents for example because the data structure is unknown at that point. <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-6f37421 elementor-widget elementor-widget-heading\" data-id=\"6f37421\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">Revisiting the Customer Service Example: Benefits of RAG for AI in Companies<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-b96ef12 elementor-widget elementor-widget-text-editor\" data-id=\"b96ef12\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Let\u2019s revisit the customer service example. Suppose a customer asks about their next payment due date. Instead of performing a semantic search on a vast amount of unstructured data, we could explain to the model that our database contains several structured tables.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-18958fa elementor-widget elementor-widget-text-editor\" data-id=\"18958fa\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>The model could then determine that it needs information from the &#8220;contract&#8221; and &#8220;transaction&#8221; tables. We query these tables, provide the relevant data points, and the model responds accordingly. This approach is straightforward, cost-effective, and well-suited for business cases involving structured data.  <\/p>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-adcaba8 elementor-widget elementor-widget-heading\" data-id=\"adcaba8\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h4 class=\"elementor-heading-title elementor-size-default\">Successful RAG Application<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-f8e1306 elementor-widget elementor-widget-text-editor\" data-id=\"f8e1306\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>RAG offers significant potential for integrating LLMs into business processes to improve efficiency and accuracy in AI-powered solutions. We support your organization to make the implementation efficient, secure and cost-effective and ensure that the full potential of your data is realized.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-449b987 elementor-section-boxed elementor-section-height-default elementor-section-height-default lottie-bg-no\" data-id=\"449b987\" data-element_type=\"section\" data-e-type=\"section\" data-settings=\"{&quot;background_background&quot;:&quot;gradient&quot;,&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t\t\t\t\t<div class=\"elementor-background-overlay\"><\/div>\n\t\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-9ee9fe3 lottie-bg-no\" data-id=\"9ee9fe3\" data-element_type=\"column\" data-e-type=\"column\" data-settings=\"{&quot;enable_lottie_background&quot;:&quot;no&quot;}\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-22b0bff elementor-widget elementor-widget-heading\" data-id=\"22b0bff\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Connecting Data \u2013 <br>Empowering Innovation<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-171ec15 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"171ec15\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"https:\/\/www.cocus.com\/en\/it-services\/data\/\" element=\"blog-contact\" type=\"services\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">COCUS Data  AI Services<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Guest contribution by Francisco Capa, Data Enginer COCUS By rethinking how we apply Retrieval-Augmented Generation (RAG), businesses can unlock new levels of AI integration that are both practical and powerful. The concept of RAG holds significant potential for simplicity, accuracy and cost efficiency in AI-powered solutions and maximizes the value of existing data. Here you [&hellip;]<\/p>\n","protected":false},"author":24,"featured_media":91218,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","footnotes":""},"categories":[436,570],"tags":[584,586],"class_list":["post-91456","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-data-en","category-news-en","tag-ai","tag-automation-en"],"_links":{"self":[{"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/posts\/91456","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/users\/24"}],"replies":[{"embeddable":true,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/comments?post=91456"}],"version-history":[{"count":3,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/posts\/91456\/revisions"}],"predecessor-version":[{"id":91466,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/posts\/91456\/revisions\/91466"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/media\/91218"}],"wp:attachment":[{"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/media?parent=91456"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/categories?post=91456"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.cocus.com\/en\/wp-json\/wp\/v2\/tags?post=91456"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}