{"id":476002,"date":"2023-08-09T07:25:33","date_gmt":"2023-08-09T07:25:33","guid":{"rendered":""},"modified":"2023-09-05T11:11:49","modified_gmt":"2023-09-05T11:11:49","slug":"bert","status":"publish","type":"wiki","link":"https:\/\/oneproxy.pro\/vn\/wiki\/bert\/","title":{"rendered":"BERT"},"content":{"rendered":"

BERT, hay \u0110\u1ea1i di\u1ec7n b\u1ed9 m\u00e3 h\u00f3a hai chi\u1ec1u t\u1eeb Transformers, l\u00e0 m\u1ed9t ph\u01b0\u01a1ng ph\u00e1p mang t\u00ednh c\u00e1ch m\u1ea1ng trong l\u0129nh v\u1ef1c x\u1eed l\u00fd ng\u00f4n ng\u1eef t\u1ef1 nhi\u00ean (NLP) s\u1eed d\u1ee5ng c\u00e1c m\u00f4 h\u00ecnh Transformer \u0111\u1ec3 hi\u1ec3u ng\u00f4n ng\u1eef theo c\u00e1ch m\u00e0 c\u00e1c c\u00f4ng ngh\u1ec7 tr\u01b0\u1edbc \u0111\u00f3 kh\u00f4ng th\u1ec3 th\u1ef1c hi\u1ec7n \u0111\u01b0\u1ee3c.<\/p>\n

Ngu\u1ed3n g\u1ed1c v\u00e0 l\u1ecbch s\u1eed c\u1ee7a BERT<\/h2>\n

BERT \u0111\u01b0\u1ee3c c\u00e1c nh\u00e0 nghi\u00ean c\u1ee9u t\u1ea1i Google AI Language gi\u1edbi thi\u1ec7u v\u00e0o n\u0103m 2018. M\u1ee5c ti\u00eau \u0111\u1eb1ng sau vi\u1ec7c t\u1ea1o ra BERT l\u00e0 cung c\u1ea5p m\u1ed9t gi\u1ea3i ph\u00e1p c\u00f3 th\u1ec3 kh\u1eafc ph\u1ee5c nh\u1eefng h\u1ea1n ch\u1ebf c\u1ee7a c\u00e1c m\u00f4 h\u00ecnh bi\u1ec3u di\u1ec5n ng\u00f4n ng\u1eef tr\u01b0\u1edbc \u0111\u00e2y. L\u1ea7n \u0111\u1ea7u ti\u00ean \u0111\u1ec1 c\u1eadp \u0111\u1ebfn BERT l\u00e0 trong b\u00e0i b\u00e1o \u201cBERT: \u0110\u00e0o t\u1ea1o tr\u01b0\u1edbc v\u1ec1 M\u00e1y bi\u1ebfn \u00e1p hai chi\u1ec1u s\u00e2u \u0111\u1ec3 hi\u1ec3u ng\u00f4n ng\u1eef,\u201d \u0111\u01b0\u1ee3c xu\u1ea5t b\u1ea3n tr\u00ean arXiv.<\/p>\n

Hi\u1ec3u BERT<\/h2>\n

BERT l\u00e0 m\u1ed9t ph\u01b0\u01a1ng ph\u00e1p \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc c\u00e1c bi\u1ec3u di\u1ec5n ng\u00f4n ng\u1eef, c\u00f3 ngh\u0129a l\u00e0 \u0111\u00e0o t\u1ea1o m\u00f4 h\u00ecnh \u201chi\u1ec3u ng\u00f4n ng\u1eef\u201d c\u00f3 m\u1ee5c \u0111\u00edch chung tr\u00ean m\u1ed9t l\u01b0\u1ee3ng l\u1edbn d\u1eef li\u1ec7u v\u0103n b\u1ea3n, sau \u0111\u00f3 tinh ch\u1ec9nh m\u00f4 h\u00ecnh \u0111\u00f3 cho c\u00e1c t\u00e1c v\u1ee5 c\u1ee5 th\u1ec3. BERT \u0111\u00e3 c\u00e1ch m\u1ea1ng h\u00f3a l\u0129nh v\u1ef1c NLP v\u00ec n\u00f3 \u0111\u01b0\u1ee3c thi\u1ebft k\u1ebf \u0111\u1ec3 m\u00f4 h\u00ecnh h\u00f3a v\u00e0 hi\u1ec3u s\u1ef1 ph\u1ee9c t\u1ea1p c\u1ee7a ng\u00f4n ng\u1eef m\u1ed9t c\u00e1ch ch\u00ednh x\u00e1c h\u01a1n.<\/p>\n

S\u1ef1 \u0111\u1ed5i m\u1edbi quan tr\u1ecdng c\u1ee7a BERT l\u00e0 \u0111\u00e0o t\u1ea1o hai chi\u1ec1u cho Transformers. Kh\u00f4ng gi\u1ed1ng nh\u01b0 c\u00e1c m\u00f4 h\u00ecnh tr\u01b0\u1edbc \u0111\u00e2y x\u1eed l\u00fd d\u1eef li\u1ec7u v\u0103n b\u1ea3n theo m\u1ed9t h\u01b0\u1edbng (t\u1eeb tr\u00e1i sang ph\u1ea3i ho\u1eb7c t\u1eeb ph\u1ea3i sang tr\u00e1i), BERT \u0111\u1ecdc to\u00e0n b\u1ed9 chu\u1ed7i t\u1eeb c\u00f9ng m\u1ed9t l\u00fac. \u0110i\u1ec1u n\u00e0y cho ph\u00e9p m\u00f4 h\u00ecnh t\u00ecm hi\u1ec3u ng\u1eef c\u1ea3nh c\u1ee7a m\u1ed9t t\u1eeb d\u1ef1a tr\u00ean t\u1ea5t c\u1ea3 m\u00f4i tr\u01b0\u1eddng xung quanh n\u00f3 (tr\u00e1i v\u00e0 ph\u1ea3i c\u1ee7a t\u1eeb \u0111\u00f3).<\/p>\n

C\u1ea5u tr\u00fac v\u00e0 ch\u1ee9c n\u0103ng b\u00ean trong c\u1ee7a BERT<\/h2>\n

BERT t\u1eadn d\u1ee5ng ki\u1ebfn tr\u00fac c\u00f3 t\u00ean Transformer. M\u00e1y bi\u1ebfn \u00e1p bao g\u1ed3m b\u1ed9 m\u00e3 h\u00f3a v\u00e0 b\u1ed9 gi\u1ea3i m\u00e3, nh\u01b0ng BERT ch\u1ec9 s\u1eed d\u1ee5ng ph\u1ea7n m\u00e3 h\u00f3a. M\u1ed7i b\u1ed9 m\u00e3 h\u00f3a Transformer c\u00f3 hai ph\u1ea7n:<\/p>\n

    \n
  1. C\u01a1 ch\u1ebf t\u1ef1 ch\u00fa \u00fd: N\u00f3 x\u00e1c \u0111\u1ecbnh nh\u1eefng t\u1eeb n\u00e0o trong c\u00e2u c\u00f3 li\u00ean quan v\u1edbi nhau. N\u00f3 l\u00e0m \u0111\u01b0\u1ee3c \u0111i\u1ec1u \u0111\u00f3 b\u1eb1ng c\u00e1ch cho \u0111i\u1ec3m m\u1ee9c \u0111\u1ed9 li\u00ean quan c\u1ee7a t\u1eebng t\u1eeb v\u00e0 s\u1eed d\u1ee5ng nh\u1eefng \u0111i\u1ec3m s\u1ed1 n\u00e0y \u0111\u1ec3 c\u00e2n nh\u1eafc t\u00e1c \u0111\u1ed9ng c\u1ee7a c\u00e1c t\u1eeb \u0111\u1ed1i v\u1edbi nhau.<\/li>\n
  2. M\u1ea1ng n\u01a1-ron chuy\u1ec3n ti\u1ebfp ngu\u1ed3n c\u1ea5p d\u1eef li\u1ec7u: Sau c\u01a1 ch\u1ebf ch\u00fa \u00fd, c\u00e1c t\u1eeb s\u1ebd \u0111\u01b0\u1ee3c chuy\u1ec3n \u0111\u1ebfn m\u1ea1ng n\u01a1-ron chuy\u1ec3n ti\u1ebfp ngu\u1ed3n c\u1ea5p d\u1eef li\u1ec7u.<\/li>\n<\/ol>\n

    Lu\u1ed3ng th\u00f4ng tin trong BERT l\u00e0 hai chi\u1ec1u, cho ph\u00e9p n\u00f3 nh\u00ecn th\u1ea5y c\u00e1c t\u1eeb tr\u01b0\u1edbc v\u00e0 sau t\u1eeb hi\u1ec7n t\u1ea1i, mang l\u1ea1i s\u1ef1 hi\u1ec3u bi\u1ebft theo ng\u1eef c\u1ea3nh ch\u00ednh x\u00e1c h\u01a1n.<\/p>\n

    C\u00e1c t\u00ednh n\u0103ng ch\u00ednh c\u1ee7a BERT<\/h2>\n
      \n
    1. \n

      T\u00ednh hai chi\u1ec1u<\/strong>: Kh\u00f4ng gi\u1ed1ng nh\u01b0 c\u00e1c m\u00f4 h\u00ecnh tr\u01b0\u1edbc \u0111\u00f3, BERT xem x\u00e9t ng\u1eef c\u1ea3nh \u0111\u1ea7y \u0111\u1ee7 c\u1ee7a m\u1ed9t t\u1eeb b\u1eb1ng c\u00e1ch xem x\u00e9t c\u00e1c t\u1eeb xu\u1ea5t hi\u1ec7n tr\u01b0\u1edbc v\u00e0 sau t\u1eeb \u0111\u00f3.<\/p>\n<\/li>\n

    2. \n

      M\u00e1y bi\u1ebfn \u00e1p<\/strong>: BERT s\u1eed d\u1ee5ng ki\u1ebfn tr\u00fac Transformer, cho ph\u00e9p n\u00f3 x\u1eed l\u00fd c\u00e1c chu\u1ed7i t\u1eeb d\u00e0i m\u1ed9t c\u00e1ch hi\u1ec7u qu\u1ea3 v\u00e0 n\u0103ng su\u1ea5t h\u01a1n.<\/p>\n<\/li>\n

    3. \n

      \u0110\u00e0o t\u1ea1o tr\u01b0\u1edbc v\u00e0 tinh ch\u1ec9nh<\/strong>: BERT \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc tr\u00ean m\u1ed9t kho d\u1eef li\u1ec7u v\u0103n b\u1ea3n l\u1edbn ch\u01b0a \u0111\u01b0\u1ee3c g\u1eafn nh\u00e3n v\u00e0 sau \u0111\u00f3 \u0111\u01b0\u1ee3c tinh ch\u1ec9nh cho m\u1ed9t t\u00e1c v\u1ee5 c\u1ee5 th\u1ec3.<\/p>\n<\/li>\n<\/ol>\n

      C\u00e1c lo\u1ea1i BERT<\/h2>\n

      BERT c\u00f3 hai k\u00edch c\u1ee1:<\/p>\n

        \n
      1. C\u01a1 s\u1edf BERT<\/strong>: 12 l\u1edbp (kh\u1ed1i bi\u1ebfn \u00e1p), 12 \u0111\u1ea7u ch\u00fa \u00fd v\u00e0 110 tri\u1ec7u tham s\u1ed1.<\/li>\n
      2. BERT-L\u1edbn<\/strong>: 24 l\u1edbp (kh\u1ed1i bi\u1ebfn \u00e1p), 16 \u0111\u1ea7u ch\u00fa \u00fd v\u00e0 340 tri\u1ec7u tham s\u1ed1.<\/li>\n<\/ol>\n\n\n\n\n\n\n\n
        <\/th>\nC\u01a1 s\u1edf BERT<\/th>\nBERT-L\u1edbn<\/th>\n<\/tr>\n<\/thead>\n
        L\u1edbp (Kh\u1ed1i bi\u1ebfn \u00e1p)<\/td>\n12<\/td>\n24<\/td>\n<\/tr>\n
        ng\u01b0\u1eddi \u0111\u1ee9ng \u0111\u1ea7u ch\u00fa \u00fd<\/td>\n12<\/td>\n16<\/td>\n<\/tr>\n
        Th\u00f4ng s\u1ed1<\/td>\n110 tri\u1ec7u<\/td>\n340 tri\u1ec7u<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

        C\u00e1ch s\u1eed d\u1ee5ng, th\u00e1ch th\u1ee9c v\u00e0 gi\u1ea3i ph\u00e1p v\u1edbi BERT<\/h2>\n

        BERT \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng r\u1ed9ng r\u00e3i trong nhi\u1ec1u nhi\u1ec7m v\u1ee5 NLP nh\u01b0 h\u1ec7 th\u1ed1ng tr\u1ea3 l\u1eddi c\u00e2u h\u1ecfi, ph\u00e2n lo\u1ea1i c\u00e2u v\u00e0 nh\u1eadn d\u1ea1ng th\u1ef1c th\u1ec3.<\/p>\n

        Nh\u1eefng th\u00e1ch th\u1ee9c v\u1edbi BERT bao g\u1ed3m:<\/p>\n

          \n
        1. \n

          T\u00e0i nguy\u00ean t\u00ednh to\u00e1n<\/strong>: BERT y\u00eau c\u1ea7u t\u00e0i nguy\u00ean t\u00ednh to\u00e1n \u0111\u00e1ng k\u1ec3 \u0111\u1ec3 \u0111\u00e0o t\u1ea1o do s\u1ed1 l\u01b0\u1ee3ng tham s\u1ed1 l\u1edbn v\u00e0 ki\u1ebfn tr\u00fac s\u00e2u.<\/p>\n<\/li>\n

        2. \n

          Thi\u1ebfu minh b\u1ea1ch<\/strong>: Gi\u1ed1ng nh\u01b0 nhi\u1ec1u m\u00f4 h\u00ecnh h\u1ecdc s\u00e2u, BERT c\u00f3 th\u1ec3 ho\u1ea1t \u0111\u1ed9ng nh\u01b0 m\u1ed9t \u201ch\u1ed9p \u0111en\u201d, khi\u1ebfn vi\u1ec7c hi\u1ec3u c\u00e1ch n\u00f3 \u0111\u01b0a ra m\u1ed9t quy\u1ebft \u0111\u1ecbnh c\u1ee5 th\u1ec3 tr\u1edf n\u00ean kh\u00f3 kh\u0103n.<\/p>\n<\/li>\n<\/ol>\n

          Gi\u1ea3i ph\u00e1p cho nh\u1eefng v\u1ea5n \u0111\u1ec1 n\u00e0y bao g\u1ed3m:<\/p>\n

            \n
          1. \n

            S\u1eed d\u1ee5ng c\u00e1c m\u00f4 h\u00ecnh \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc<\/strong>: Thay v\u00ec \u0111\u00e0o t\u1ea1o t\u1eeb \u0111\u1ea7u, ng\u01b0\u1eddi ta c\u00f3 th\u1ec3 s\u1eed d\u1ee5ng c\u00e1c m\u00f4 h\u00ecnh BERT \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc v\u00e0 tinh ch\u1ec9nh ch\u00fang cho c\u00e1c nhi\u1ec7m v\u1ee5 c\u1ee5 th\u1ec3, \u0111\u00f2i h\u1ecfi \u00edt t\u00e0i nguy\u00ean t\u00ednh to\u00e1n h\u01a1n.<\/p>\n<\/li>\n

          2. \n

            C\u00f4ng c\u1ee5 gi\u1ea3i th\u00edch<\/strong>: C\u00e1c c\u00f4ng c\u1ee5 nh\u01b0 LIME v\u00e0 SHAP c\u00f3 th\u1ec3 gi\u00fap \u0111\u01b0a ra c\u00e1c quy\u1ebft \u0111\u1ecbnh c\u1ee7a m\u00f4 h\u00ecnh BERT d\u1ec5 hi\u1ec3u h\u01a1n.<\/p>\n<\/li>\n<\/ol>\n

            BERT v\u00e0 c\u00e1c c\u00f4ng ngh\u1ec7 t\u01b0\u01a1ng t\u1ef1<\/h2>\n\n\n\n\n\n\n\n
            <\/th>\nBERT<\/th>\nLSTM<\/th>\n<\/tr>\n<\/thead>\n
            Ph\u01b0\u01a1ng h\u01b0\u1edbng<\/td>\nhai chi\u1ec1u<\/td>\nM\u1ed9t chi\u1ec1u<\/td>\n<\/tr>\n
            Ng\u00e0nh ki\u1ebfn tr\u00fac<\/td>\nM\u00e1y bi\u1ebfn \u00e1p<\/td>\n\u0110\u1ecbnh k\u1ef3<\/td>\n<\/tr>\n
            Hi\u1ec3u bi\u1ebft theo ng\u1eef c\u1ea3nh<\/td>\nT\u1ed1t h\u01a1n<\/td>\nGi\u1edbi h\u1ea1n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

            Vi\u1ec5n c\u1ea3nh t\u01b0\u01a1ng lai v\u00e0 c\u00f4ng ngh\u1ec7 li\u00ean quan \u0111\u1ebfn BERT<\/h2>\n

            BERT ti\u1ebfp t\u1ee5c truy\u1ec1n c\u1ea3m h\u1ee9ng cho c\u00e1c m\u00f4 h\u00ecnh m\u1edbi trong NLP. DistilBERT, phi\u00ean b\u1ea3n BERT nh\u1ecf h\u01a1n, nhanh h\u01a1n v\u00e0 nh\u1eb9 h\u01a1n v\u00e0 RoBERTa, phi\u00ean b\u1ea3n BERT lo\u1ea1i b\u1ecf m\u1ee5c ti\u00eau hu\u1ea5n luy\u1ec7n tr\u01b0\u1edbc c\u00e2u ti\u1ebfp theo, l\u00e0 nh\u1eefng v\u00ed d\u1ee5 v\u1ec1 nh\u1eefng ti\u1ebfn b\u1ed9 g\u1ea7n \u0111\u00e2y.<\/p>\n

            Nghi\u00ean c\u1ee9u trong t\u01b0\u01a1ng lai v\u1ec1 BERT c\u00f3 th\u1ec3 t\u1eadp trung v\u00e0o vi\u1ec7c l\u00e0m cho m\u00f4 h\u00ecnh hi\u1ec7u qu\u1ea3 h\u01a1n, d\u1ec5 hi\u1ec3u h\u01a1n v\u00e0 x\u1eed l\u00fd c\u00e1c chu\u1ed7i d\u00e0i h\u01a1n t\u1ed1t h\u01a1n.<\/p>\n

            M\u00e1y ch\u1ee7 BERT v\u00e0 Proxy<\/h2>\n

            BERT ph\u1ea7n l\u1edbn kh\u00f4ng li\u00ean quan \u0111\u1ebfn m\u00e1y ch\u1ee7 proxy, v\u00ec BERT l\u00e0 m\u00f4 h\u00ecnh NLP v\u00e0 m\u00e1y ch\u1ee7 proxy l\u00e0 c\u00f4ng c\u1ee5 m\u1ea1ng. Tuy nhi\u00ean, khi t\u1ea3i xu\u1ed1ng c\u00e1c m\u00f4 h\u00ecnh BERT \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc ho\u1eb7c s\u1eed d\u1ee5ng ch\u00fang th\u00f4ng qua API, m\u00e1y ch\u1ee7 proxy \u0111\u00e1ng tin c\u1eady, nhanh ch\u00f3ng v\u00e0 an to\u00e0n nh\u01b0 OneProxy c\u00f3 th\u1ec3 \u0111\u1ea3m b\u1ea3o truy\u1ec1n d\u1eef li\u1ec7u \u1ed5n \u0111\u1ecbnh v\u00e0 an to\u00e0n.<\/p>\n

            Li\u00ean k\u1ebft li\u00ean quan<\/h2>\n
              \n
            1. \n

              BERT: \u0110\u00e0o t\u1ea1o tr\u01b0\u1edbc M\u00e1y bi\u1ebfn \u00e1p hai chi\u1ec1u s\u00e2u \u0111\u1ec3 hi\u1ec3u ng\u00f4n ng\u1eef<\/a><\/p>\n<\/li>\n

            2. \n

              Blog AI c\u1ee7a Google: Ngu\u1ed3n m\u1edf BERT<\/a><\/p>\n<\/li>\n

            3. \n

              Gi\u1ea3i th\u00edch v\u1ec1 BERT: H\u01b0\u1edbng d\u1eabn \u0111\u1ea7y \u0111\u1ee7 v\u1ec1 L\u00fd thuy\u1ebft v\u00e0 H\u01b0\u1edbng d\u1eabn<\/a><\/p>\n<\/li>\n<\/ol>","protected":false},"featured_media":467710,"menu_order":0,"template":"","meta":{"_acf_changed":false,"content-type":"","inline_featured_image":false,"footnotes":""},"class_list":["post-476002","wiki","type-wiki","status-publish","has-post-thumbnail","hentry"],"acf":{"faq_title":"Frequently Asked Questions about Bidirectional Encoder Representations from Transformers (BERT)<\/mark>","faq_items":[{"question":"What is BERT?","answer":"

              BERT, or Bidirectional Encoder Representations from Transformers, is a cutting-edge method in the field of natural language processing (NLP) that leverages Transformer models to understand language in a way that surpasses earlier technologies.<\/p>"},{"question":"Who introduced BERT and when?","answer":"

              BERT was introduced by researchers at Google AI Language in 2018. The paper titled \"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,\" published on arXiv, was the first to mention BERT.<\/p>"},{"question":"What is the key innovation of BERT?","answer":"

              The key innovation of BERT is its bidirectional training of Transformers. This is a departure from previous models that processed text data in one direction only. BERT reads the entire sequence of words at once, learning the context of a word based on all its surroundings.<\/p>"},{"question":"How does BERT work internally?","answer":"

              BERT uses an architecture known as Transformer, specifically its encoder part. Each Transformer encoder comprises a self-attention mechanism, which determines the relevance of words to each other, and a feed-forward neural network, which the words pass through after the attention mechanism. BERT's bidirectional information flow gives it a richer contextual understanding of language.<\/p>"},{"question":"What are the main types of BERT?","answer":"

              BERT primarily comes in two sizes: BERT-Base and BERT-Large. BERT-Base has 12 layers, 12 attention heads, and 110 million parameters. BERT-Large, on the other hand, has 24 layers, 16 attention heads, and 340 million parameters.<\/p>"},{"question":"What challenges might one face when using BERT?","answer":"

              BERT requires substantial computational resources for training due to its large number of parameters and deep architecture. Furthermore, like many deep learning models, BERT can be a \"black box,\" making it challenging to understand how it makes a particular decision.<\/p>"},{"question":"How do BERT and proxy servers relate?","answer":"

              While BERT and proxy servers operate in different spheres (NLP and networking, respectively), a proxy server can be crucial when downloading pre-trained BERT models or using them via APIs. A reliable proxy server like OneProxy ensures secure and stable data transmission.<\/p>"},{"question":"What are the future prospects related to BERT?","answer":"

              BERT continues to inspire new models in NLP like DistilBERT and RoBERTa. Future research in BERT may focus on making the model more efficient, more interpretable, and better at handling longer sequences.<\/p>"}]},"_links":{"self":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/wiki\/476002","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/wiki"}],"about":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/types\/wiki"}],"version-history":[{"count":0,"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/wiki\/476002\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/media\/467710"}],"wp:attachment":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/media?parent=476002"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}