{"id":3895,"date":"2023-11-04T23:13:55","date_gmt":"2023-11-04T23:13:55","guid":{"rendered":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/"},"modified":"2023-11-05T05:48:28","modified_gmt":"2023-11-05T05:48:28","slug":"how-to-evaluate-the-accuracy-and-bias-of-llms","status":"publish","type":"post","link":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/","title":{"rendered":"How to evaluate the accuracy and bias of LLMs"},"content":{"rendered":"<h1>How to Evaluate the Accuracy and Bias of Language Models<\/h1>\n<p>Language models have become increasingly sophisticated in recent years, thanks to advancements in deep learning and natural language processing algorithms. However, with this improvement in complexity comes a need to carefully evaluate the accuracy and potential biases of these models.<\/p>\n<p>In this tutorial, we will explore various methods and techniques for evaluating the accuracy and bias of language models, particularly focusing on Large Language Models (LLMs). LLMs are often used for tasks like text generation, translation, summarization, and sentiment analysis.<\/p>\n<h2>Table of Contents<\/h2>\n<ol>\n<li>Introduction to Language Model Evaluation<\/li>\n<li>Accuracy Evaluation Techniques\n<ul>\n<li>Perplexity Calculation<\/li>\n<li>Language Modeling Evaluation Datasets<\/li>\n<\/ul>\n<\/li>\n<li>Bias Evaluation Techniques\n<ul>\n<li>Word Embedding Analysis<\/li>\n<li>Dataset Analysis<\/li>\n<\/ul>\n<\/li>\n<li>Conclusion<\/li>\n<\/ol>\n<h2>1. Introduction to Language Model Evaluation<\/h2>\n<p>Language model evaluation is the process of assessing the performance, accuracy, and potential biases of language models. Evaluating language models is crucial to ensure that they produce reliable and high-quality results.<\/p>\n<p>There are two primary aspects to consider when evaluating language models:<\/p>\n<ol>\n<li>Accuracy: This refers to how well a language model performs on specific language tasks, such as text generation or sentiment analysis. Accuracy evaluation helps determine whether a language model is producing plausible and coherent outputs.<\/p>\n<\/li>\n<li>\n<p>Bias: Language models, like any AI system, can reflect biases present in the training data. Bias evaluation aims to identify and mitigate any biases present in the language model to ensure that it produces fair and unbiased results.<\/p>\n<\/li>\n<\/ol>\n<p>In the following sections, we will discuss specific techniques and approaches for evaluating both the accuracy and bias of language models.<\/p>\n<h2>2. Accuracy Evaluation Techniques<\/h2>\n<p>Evaluating the accuracy of a language model is crucial to ensure that its generated outputs are reliable and coherent. Here are two popular techniques for accuracy evaluation:<\/p>\n<h3>Perplexity Calculation<\/h3>\n<p>Perplexity is a widely used metric for evaluating the accuracy of language models. It measures how well a language model predicts a given text. The lower the perplexity value, the better the language model&#8217;s performance.<\/p>\n<p>Perplexity can be calculated using the following formula:<\/p>\n<pre><code class=\"language-math\">perplexity = 2^{-frac{1}{N} sum_{i=1}^{N} log_{2}(p(w_i | w_1, w_2, ..., w_{i-1}))}\n<\/code><\/pre>\n<p>Where:<br \/>\n&#8211; <code>N<\/code> is the total number of words in the evaluation dataset.<br \/>\n&#8211; <code>w_i<\/code> represents the <code>i-th<\/code> word in the evaluation dataset.<\/p>\n<p>To calculate perplexity, you need an evaluation dataset and a trained language model. First, you feed each word in the evaluation dataset to the language model and calculate the model&#8217;s predicted probability for each word. Then, you compute the average of the logarithm of these probabilities, and finally, transform it using 2 as the base.<\/p>\n<h3>Language Modeling Evaluation Datasets<\/h3>\n<p>Another approach to evaluate the accuracy of a language model is to use language modeling evaluation datasets. These datasets are designed to test how well a language model can generate coherent and grammatically correct text.<\/p>\n<p>Popular language modeling evaluation datasets include:<\/p>\n<ul>\n<li><strong>Penn Treebank:<\/strong> This dataset consists of annotated data from articles published in the Wall Street Journal. It is widely used for evaluating language models.<\/p>\n<\/li>\n<li>\n<p><strong>WikiText:<\/strong> WikiText is another popular dataset for language modeling evaluation. It includes a large amount of text from Wikipedia articles.<\/p>\n<\/li>\n<\/ul>\n<p>Using these datasets, you can assess the language model&#8217;s performance by measuring various metrics such as perplexity, BLEU scores, or even qualitative analysis of the generated text.<\/p>\n<h2>3. Bias Evaluation Techniques<\/h2>\n<p>Evaluating and mitigating biases in language models is crucial to ensure fair and unbiased results. Here are two techniques for evaluating bias in language models:<\/p>\n<h3>Word Embedding Analysis<\/h3>\n<p>Word embeddings are dense vector representations of words that capture semantic meaning. Analyzing word embeddings can help identify potential biases present in the language model. For instance, biased word embeddings may exhibit gender, racial, or cultural biases.<\/p>\n<p>You can evaluate bias in word embeddings using techniques such as:<\/p>\n<ul>\n<li><strong>Analogy-based evaluation:<\/strong> Test a language model&#8217;s embeddings for gender biases by analyzing analogical relationships (e.g., &#8220;man&#8221; is to &#8220;woman&#8221; as &#8220;king&#8221; is to &#8220;queen&#8221;).<\/p>\n<\/li>\n<li>\n<p><strong>Word similarity evaluation:<\/strong> Measure the cosine similarity between pairs of words to determine if the embeddings cluster words according to biases (e.g., gender, profession, or race).<\/p>\n<\/li>\n<\/ul>\n<h3>Dataset Analysis<\/h3>\n<p>Biased datasets can contribute to biased language models. Evaluating the training data used to train a language model is essential to identify potential biases. This can involve analyzing the demographic distribution, representation, and fairness of the training data.<\/p>\n<p>You can evaluate dataset biases using the following techniques:<\/p>\n<ul>\n<li><strong>Demographic parity:<\/strong> Assess whether the distribution of sensitive attributes (e.g., gender or race) in the training data is proportional to their distribution in the real world.<\/p>\n<\/li>\n<li>\n<p><strong>Word usage analysis:<\/strong> Examine the frequency of specific words and their potential biases. Biased datasets may contain over- or under-representation of certain groups, or contain explicit biases within the text.<\/p>\n<\/li>\n<\/ul>\n<h2>4. Conclusion<\/h2>\n<p>Evaluating the accuracy and bias of language models is essential to ensure their reliability, fairness, and usefulness. In this tutorial, we discussed various techniques for evaluating both accuracy and bias in language models.<\/p>\n<p>For accuracy evaluation, perplexity calculation and language modeling evaluation datasets are commonly used. Perplexity provides a quantitative measure of a language model&#8217;s performance, while evaluation datasets allow for qualitative analysis and comparison.<\/p>\n<p>To evaluate bias, word embedding analysis and dataset analysis are crucial. Analyzing word embeddings helps identify potential biases in semantic representation, while dataset analysis allows for the detection of biases in the training data.<\/p>\n<p>By employing these techniques, developers and researchers can assess and improve the accuracy and fairness of language models, making them more reliable and unbiased for a wide range of applications.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>How to Evaluate the Accuracy and Bias of Language Models Language models have become increasingly sophisticated in recent years, thanks to advancements in deep learning and natural language processing algorithms. However, with this improvement in complexity comes a need to carefully evaluate the accuracy and potential biases of these models. <a href=\"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\" class=\"btn btn-link continue-link\">Continue Reading<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_import_markdown_pro_load_document_selector":0,"_import_markdown_pro_submit_text_textarea":"","footnotes":""},"categories":[1],"tags":[182,179,178,181,185,177,183,180,184,176],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.5 (Yoast SEO v21.5) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>How to evaluate the accuracy and bias of LLMs - Pantherax Blogs<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"How to evaluate the accuracy and bias of LLMs\" \/>\n<meta property=\"og:description\" content=\"How to Evaluate the Accuracy and Bias of Language Models Language models have become increasingly sophisticated in recent years, thanks to advancements in deep learning and natural language processing algorithms. However, with this improvement in complexity comes a need to carefully evaluate the accuracy and potential biases of these models. Continue Reading\" \/>\n<meta property=\"og:url\" content=\"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\" \/>\n<meta property=\"og:site_name\" content=\"Pantherax Blogs\" \/>\n<meta property=\"article:published_time\" content=\"2023-11-04T23:13:55+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-11-05T05:48:28+00:00\" \/>\n<meta name=\"author\" content=\"Panther\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Panther\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\n\t    \"@context\": \"https:\/\/schema.org\",\n\t    \"@graph\": [\n\t        {\n\t            \"@type\": \"Article\",\n\t            \"@id\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/#article\",\n\t            \"isPartOf\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\"\n\t            },\n\t            \"author\": {\n\t                \"name\": \"Panther\",\n\t                \"@id\": \"http:\/\/localhost:10003\/#\/schema\/person\/b63d816f4964b163e53cbbcffaa0f3d7\"\n\t            },\n\t            \"headline\": \"How to evaluate the accuracy and bias of LLMs\",\n\t            \"datePublished\": \"2023-11-04T23:13:55+00:00\",\n\t            \"dateModified\": \"2023-11-05T05:48:28+00:00\",\n\t            \"mainEntityOfPage\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\"\n\t            },\n\t            \"wordCount\": 903,\n\t            \"publisher\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/#organization\"\n\t            },\n\t            \"keywords\": [\n\t                \"\\\"Accuracy and bias in LLMs evaluation\\\"\",\n\t                \"\\\"Analyze LLM bias\\\"\",\n\t                \"\\\"Assess LLM accuracy\\\"\",\n\t                \"\\\"Assessing bias in LLMs\\\"\",\n\t                \"\\\"Bias analysis in LLMs\\\"]\",\n\t                \"\\\"Bias in LLMs\\\"\",\n\t                \"\\\"Evaluating LLMs: accuracy and bias\\\"\",\n\t                \"\\\"Evaluating the accuracy of LLMs\\\"\",\n\t                \"\\\"LLM accuracy assessment\\\"\",\n\t                \"[\\\"Evaluate accuracy of LLMs\\\"\"\n\t            ],\n\t            \"inLanguage\": \"en-US\"\n\t        },\n\t        {\n\t            \"@type\": \"WebPage\",\n\t            \"@id\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\",\n\t            \"url\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\",\n\t            \"name\": \"How to evaluate the accuracy and bias of LLMs - Pantherax Blogs\",\n\t            \"isPartOf\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/#website\"\n\t            },\n\t            \"datePublished\": \"2023-11-04T23:13:55+00:00\",\n\t            \"dateModified\": \"2023-11-05T05:48:28+00:00\",\n\t            \"breadcrumb\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/#breadcrumb\"\n\t            },\n\t            \"inLanguage\": \"en-US\",\n\t            \"potentialAction\": [\n\t                {\n\t                    \"@type\": \"ReadAction\",\n\t                    \"target\": [\n\t                        \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/\"\n\t                    ]\n\t                }\n\t            ]\n\t        },\n\t        {\n\t            \"@type\": \"BreadcrumbList\",\n\t            \"@id\": \"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/#breadcrumb\",\n\t            \"itemListElement\": [\n\t                {\n\t                    \"@type\": \"ListItem\",\n\t                    \"position\": 1,\n\t                    \"name\": \"Home\",\n\t                    \"item\": \"http:\/\/localhost:10003\/\"\n\t                },\n\t                {\n\t                    \"@type\": \"ListItem\",\n\t                    \"position\": 2,\n\t                    \"name\": \"How to evaluate the accuracy and bias of LLMs\"\n\t                }\n\t            ]\n\t        },\n\t        {\n\t            \"@type\": \"WebSite\",\n\t            \"@id\": \"http:\/\/localhost:10003\/#website\",\n\t            \"url\": \"http:\/\/localhost:10003\/\",\n\t            \"name\": \"Pantherax Blogs\",\n\t            \"description\": \"\",\n\t            \"publisher\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/#organization\"\n\t            },\n\t            \"potentialAction\": [\n\t                {\n\t                    \"@type\": \"SearchAction\",\n\t                    \"target\": {\n\t                        \"@type\": \"EntryPoint\",\n\t                        \"urlTemplate\": \"http:\/\/localhost:10003\/?s={search_term_string}\"\n\t                    },\n\t                    \"query-input\": \"required name=search_term_string\"\n\t                }\n\t            ],\n\t            \"inLanguage\": \"en-US\"\n\t        },\n\t        {\n\t            \"@type\": \"Organization\",\n\t            \"@id\": \"http:\/\/localhost:10003\/#organization\",\n\t            \"name\": \"Pantherax Blogs\",\n\t            \"url\": \"http:\/\/localhost:10003\/\",\n\t            \"logo\": {\n\t                \"@type\": \"ImageObject\",\n\t                \"inLanguage\": \"en-US\",\n\t                \"@id\": \"http:\/\/localhost:10003\/#\/schema\/logo\/image\/\",\n\t                \"url\": \"http:\/\/localhost:10003\/wp-content\/uploads\/2023\/11\/cropped-9e7721cb-2d62-4f72-ab7f-7d1d8db89226.jpeg\",\n\t                \"contentUrl\": \"http:\/\/localhost:10003\/wp-content\/uploads\/2023\/11\/cropped-9e7721cb-2d62-4f72-ab7f-7d1d8db89226.jpeg\",\n\t                \"width\": 1024,\n\t                \"height\": 1024,\n\t                \"caption\": \"Pantherax Blogs\"\n\t            },\n\t            \"image\": {\n\t                \"@id\": \"http:\/\/localhost:10003\/#\/schema\/logo\/image\/\"\n\t            }\n\t        },\n\t        {\n\t            \"@type\": \"Person\",\n\t            \"@id\": \"http:\/\/localhost:10003\/#\/schema\/person\/b63d816f4964b163e53cbbcffaa0f3d7\",\n\t            \"name\": \"Panther\",\n\t            \"image\": {\n\t                \"@type\": \"ImageObject\",\n\t                \"inLanguage\": \"en-US\",\n\t                \"@id\": \"http:\/\/localhost:10003\/#\/schema\/person\/image\/\",\n\t                \"url\": \"http:\/\/2.gravatar.com\/avatar\/b8c0eda5a49f8f31ec32d0a0f9d6f838?s=96&d=mm&r=g\",\n\t                \"contentUrl\": \"http:\/\/2.gravatar.com\/avatar\/b8c0eda5a49f8f31ec32d0a0f9d6f838?s=96&d=mm&r=g\",\n\t                \"caption\": \"Panther\"\n\t            },\n\t            \"sameAs\": [\n\t                \"http:\/\/localhost:10003\"\n\t            ],\n\t            \"url\": \"http:\/\/localhost:10003\/author\/pepethefrog\/\"\n\t        }\n\t    ]\n\t}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"How to evaluate the accuracy and bias of LLMs - Pantherax Blogs","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/","og_locale":"en_US","og_type":"article","og_title":"How to evaluate the accuracy and bias of LLMs","og_description":"How to Evaluate the Accuracy and Bias of Language Models Language models have become increasingly sophisticated in recent years, thanks to advancements in deep learning and natural language processing algorithms. However, with this improvement in complexity comes a need to carefully evaluate the accuracy and potential biases of these models. Continue Reading","og_url":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/","og_site_name":"Pantherax Blogs","article_published_time":"2023-11-04T23:13:55+00:00","article_modified_time":"2023-11-05T05:48:28+00:00","author":"Panther","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Panther","Est. reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/#article","isPartOf":{"@id":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/"},"author":{"name":"Panther","@id":"http:\/\/localhost:10003\/#\/schema\/person\/b63d816f4964b163e53cbbcffaa0f3d7"},"headline":"How to evaluate the accuracy and bias of LLMs","datePublished":"2023-11-04T23:13:55+00:00","dateModified":"2023-11-05T05:48:28+00:00","mainEntityOfPage":{"@id":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/"},"wordCount":903,"publisher":{"@id":"http:\/\/localhost:10003\/#organization"},"keywords":["\"Accuracy and bias in LLMs evaluation\"","\"Analyze LLM bias\"","\"Assess LLM accuracy\"","\"Assessing bias in LLMs\"","\"Bias analysis in LLMs\"]","\"Bias in LLMs\"","\"Evaluating LLMs: accuracy and bias\"","\"Evaluating the accuracy of LLMs\"","\"LLM accuracy assessment\"","[\"Evaluate accuracy of LLMs\""],"inLanguage":"en-US"},{"@type":"WebPage","@id":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/","url":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/","name":"How to evaluate the accuracy and bias of LLMs - Pantherax Blogs","isPartOf":{"@id":"http:\/\/localhost:10003\/#website"},"datePublished":"2023-11-04T23:13:55+00:00","dateModified":"2023-11-05T05:48:28+00:00","breadcrumb":{"@id":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/"]}]},{"@type":"BreadcrumbList","@id":"http:\/\/localhost:10003\/how-to-evaluate-the-accuracy-and-bias-of-llms\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"http:\/\/localhost:10003\/"},{"@type":"ListItem","position":2,"name":"How to evaluate the accuracy and bias of LLMs"}]},{"@type":"WebSite","@id":"http:\/\/localhost:10003\/#website","url":"http:\/\/localhost:10003\/","name":"Pantherax Blogs","description":"","publisher":{"@id":"http:\/\/localhost:10003\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"http:\/\/localhost:10003\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"en-US"},{"@type":"Organization","@id":"http:\/\/localhost:10003\/#organization","name":"Pantherax Blogs","url":"http:\/\/localhost:10003\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"http:\/\/localhost:10003\/#\/schema\/logo\/image\/","url":"http:\/\/localhost:10003\/wp-content\/uploads\/2023\/11\/cropped-9e7721cb-2d62-4f72-ab7f-7d1d8db89226.jpeg","contentUrl":"http:\/\/localhost:10003\/wp-content\/uploads\/2023\/11\/cropped-9e7721cb-2d62-4f72-ab7f-7d1d8db89226.jpeg","width":1024,"height":1024,"caption":"Pantherax Blogs"},"image":{"@id":"http:\/\/localhost:10003\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"http:\/\/localhost:10003\/#\/schema\/person\/b63d816f4964b163e53cbbcffaa0f3d7","name":"Panther","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"http:\/\/localhost:10003\/#\/schema\/person\/image\/","url":"http:\/\/2.gravatar.com\/avatar\/b8c0eda5a49f8f31ec32d0a0f9d6f838?s=96&d=mm&r=g","contentUrl":"http:\/\/2.gravatar.com\/avatar\/b8c0eda5a49f8f31ec32d0a0f9d6f838?s=96&d=mm&r=g","caption":"Panther"},"sameAs":["http:\/\/localhost:10003"],"url":"http:\/\/localhost:10003\/author\/pepethefrog\/"}]}},"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"","_links":{"self":[{"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/posts\/3895"}],"collection":[{"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/comments?post=3895"}],"version-history":[{"count":1,"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/posts\/3895\/revisions"}],"predecessor-version":[{"id":4631,"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/posts\/3895\/revisions\/4631"}],"wp:attachment":[{"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/media?parent=3895"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/categories?post=3895"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/localhost:10003\/wp-json\/wp\/v2\/tags?post=3895"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}