{"id":3801,"date":"2024-12-10T22:40:08","date_gmt":"2024-12-10T14:40:08","guid":{"rendered":"https:\/\/infernews.com\/?page_id=3801"},"modified":"2026-10-03T00:16:50","modified_gmt":"2026-10-02T16:16:50","slug":"text-embedding","status":"publish","type":"page","link":"https:\/\/infernews.com\/blog\/text-embedding\/","title":{"rendered":"A.I. \u5165\u9580"},"content":{"rendered":"\n<figure class=\"wp-block-image size-full\"><img data-dominant-color=\"636c75\" data-has-transparency=\"false\" style=\"--dominant-color: #636c75;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"397\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/10\/image-2.png\" alt=\"\" class=\"wp-image-12864 not-transparent\" srcset=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/10\/image-2.png 1024w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/10\/image-2-300x116.png 300w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/10\/image-2-766x297.png 766w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n<div class=\"vlp-link-container vlp-layout-spotlight-clone block-editor-block-list__block wp-block wp-block-visual-link-preview-link\"><a href=\"https:\/\/infernews.com\/ai\/kids\/\" class=\"vlp-link\" title=\"Infer Kids AI\" rel=\"nofollow\" target=\"_blank\"><\/a><div class=\"vlp-layout-zone-main\"><span class=\"vlp-block-0 vlp-link-title\">Infer Kids AI<\/span><div class=\"vlp-block-1 vlp-link-summary\">\u8b93\u5b69\u5b50\u5feb\u6a02\u63a2\u7d22 AI \u4e16\u754c<\/div><div class=\"vlp-block-2 vlp-link-image\"><img decoding=\"async\" src=\"https:\/\/infernews.com\/ai\/kids\/transformer\/images\/banner01.jpg\" style=\"max-width: 1024px; max-height: 1024px\" \/><\/div><\/div><\/div>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"How Machine Learning Works, End to End (Full Basics)\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_du1B6633-hU\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fdu1B6633-hU%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/du1B6633-hU\" \/><meta itemprop=\"duration\" content=\"PT1H28S\" \/><meta itemprop=\"uploadDate\" content=\"2026-08-27T14:00:39Z\" \/><\/div><div id=\"lyte_du1B6633-hU\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fdu1B6633-hU%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">How Machine Learning Works, End to End (Full Basics)<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/du1B6633-hU\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fdu1B6633-hU%2F0.jpg\" alt=\"How Machine Learning Works, End to End (Full Basics)\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"Machine learning explained end to end, from a messy spreadsheet to a trained model that makes real predictions. Most explanations either stay at the buzzword level or drown you in math. We take the middle path. One housing-price example gets carried from raw data all the way through loss and gradient descent, then the exact same training loop scales up to how ChatGPT is actually built. By the end, the words weights, loss, epochs, tokens, and RLHF stop being jargon and start being one connected idea. \ud83e\uddea Try it yourself in the free lab: https:\/\/kode.wiki\/4gVrsfX \ud83d\udcda What you&#039;ll learn: 1\ufe0f\u20e3 Why if-else rules break on messy problems, and what &quot;learning from examples&quot; replaces them with 2\ufe0f\u20e3 What actually changes inside a model when it learns (weights, bias, and the loss that guides them) 3\ufe0f\u20e3 How gradient descent, batches, and epochs turn a wrong guess into a good prediction 4\ufe0f\u20e3 Why models overfit, and how train\/validation\/test splits catch it before you ship 5\ufe0f\u20e3 How the same loop becomes a neural network, and how GPT gets from next-token prediction to a helpful assistant \ud83d\udea8 Start Your AI Journey with KodeKloud: https:\/\/kode.wiki\/4qsrspX \ud83e\uddea Free hands-on lab: https:\/\/kode.wiki\/4gVrsfX \u23f0 Timestamps: 00:00 - Introduction to Machine Learning 00:48 - A Short History of AI and Machine learning 05:25 - What Machine Learning actually is 08:40 - Data, Features, and Labels 12:50 - Lab1: Clean a Messy Housing Dataset 13:37 - Training and Inference 16:00 - Models, Parameters, and Weights 20:13 - Loss functions: Scoring how Wrong the Model is 23:58 - Gradient Descent Explained 28:02 - Lab2: Tune the Model 28:26 - Train, validation, and Test Sets 29:14 - Overfitting and Data Leakage 32:16 - Lab 3: Catch Overfitting and a Data Leak 32:48 - Supervised vs unsupervised learning 36:36 - Classification and Regression 40:00 - Neural Networks Explained 47:20 - How ChatGPT and LLMs are actually Trained 55:58 - Lab4: Train a Neural Network on Handwritten Digits 56:28 - Modern AI: transformers, multimodal, and agents \ud83e\udd16 Learn AI from this Playlist: https:\/\/www.youtube.com\/playlist?list=PL2We04F3Y_43f3x3n9pawcEuAwru7bcMG \ud83d\udd14 Subscribe for more AI and machine learning fundamentals, explained for engineers #MachineLearning #HowAIWorks #DeepLearning #KodeKloud #NeuralNetworks #MLBasics #ArtificialIntelligence #GradientDescent #LLM #ChatGPT #AIforBeginners #SupervisedLearning #Transformers #Backpropagation #AIEngineering\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<p class=\"wp-block-paragraph\">\u4eba\u5de5\u667a\u6167\uff08AI\uff09\u8207\u6a5f\u5668\u5b78\u7fd2\u7684\u5e95\u5c64\u5b8c\u5168\u662f\u7531\u6578\u5b78\u69cb\u5efa\u800c\u6210\u7684\u3002\u5982\u679c\u8981\u7cfb\u7d71\u6027\u5730\u5217\u51fa<strong>\u6240\u6709<\/strong> AI \u9700\u8981\u7684\u6578\u5b78\u57fa\u790e\uff0c\u53ef\u4ee5\u5c07\u5b83\u5011\u5206\u70ba\u300c\u56db\u5927\u6838\u5fc3\u652f\u67f1\u300d\u4ee5\u53ca\u300c\u4e09\u5927\u5ef6\u4f38\u8207\u9ad8\u968e\u9818\u57df\u300d\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u4ee5\u4e0b\u662f\u70ba\u4f60\u6574\u7406\u7684\u5b8c\u6574 AI \u6578\u5b78\u5730\u5716\uff1a<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u6838\u5fc3\u652f\u67f1\u4e00\uff1a\u7dda\u6027\u4ee3\u6578 (Linear Algebra) \u2014\u2014 \u7a7a\u9593\u8207\u8cc7\u6599\u7684\u8f49\u63db<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u7dda\u6027\u4ee3\u6578\u662f AI \u8655\u7406\u8cc7\u6599\u7684\u8a9e\u8a00\u3002\u5728 AI \u4e2d\uff0c\u6240\u6709\u7684\u8cc7\u6599\uff08\u6587\u5b57\u3001\u5716\u7247\u3001\u8072\u97f3\uff09\u90fd\u6703\u88ab\u8f49\u63db\u6210\u5411\u91cf\u548c\u77e9\u9663\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u57fa\u790e\u6982\u5ff5\uff1a<\/strong> \u5411\u91cf (Vectors)\u3001\u77e9\u9663 (Matrices)\u3001\u5f35\u91cf (Tensors)\u3002<\/li>\n\n\n\n<li><strong>\u6838\u5fc3\u904b\u7b97\uff1a<\/strong> \u77e9\u9663\u4e58\u6cd5\u3001\u884c\u5217\u5f0f\u3001\u77e9\u9663\u7684\u8de1 (Trace)\u3002<\/li>\n\n\n\n<li><strong>\u77e9\u9663\u5206\u89e3\u8207\u7a7a\u9593\u8f49\u63db\uff1a<\/strong> * \u7279\u5fb5\u503c\u8207\u7279\u5fb5\u5411\u91cf (Eigenvalues &amp; Eigenvectors)\n<ul class=\"wp-block-list\">\n<li>\u5947\u7570\u503c\u5206\u89e3 (SVD, Singular Value Decomposition)<\/li>\n\n\n\n<li>\u4e3b\u6210\u5206\u5206\u6790 (PCA) \u2014\u2014 \u7528\u65bc\u8cc7\u6599\u964d\u7dad\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>AI \u61c9\u7528\uff1a<\/strong> \u795e\u7d93\u7db2\u8def\u6b0a\u91cd\u904b\u7b97\u3001\u5716\u50cf\u5377\u7a4d\u8655\u7406\u3001\u5927\u578b\u8a9e\u8a00\u6a21\u578b\uff08LLM\uff09\u7684 Embedding\uff08\u8a5e\u5d4c\u5165\u7a7a\u9593\uff09\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u6838\u5fc3\u652f\u67f1\u4e8c\uff1a\u5fae\u7a4d\u5206 (Calculus) \u2014\u2014 \u6a21\u578b\u7684\u5b78\u7fd2\u8207\u512a\u5316<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u5fae\u7a4d\u5206\u8ca0\u8cac\u89e3\u6c7a AI \u5982\u4f55\u300c\u5f9e\u932f\u8aa4\u4e2d\u5b78\u7fd2\u300d\u7684\u554f\u984c\u3002\u900f\u904e\u8a08\u7b97\u8b8a\u5316\u7387\uff0cAI \u624d\u80fd\u8abf\u6574\u81ea\u8eab\u53c3\u6578\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u55ae\u8b8a\u6578\u8207\u591a\u8b8a\u6578\u5fae\u7a4d\u5206\uff1a<\/strong> \u5c0e\u6578 (Derivatives)\u3001\u504f\u5c0e\u6578 (Partial Derivatives)\u3002<\/li>\n\n\n\n<li><strong>\u91cd\u8981\u51fd\u6578\u8207\u6cd5\u5247\uff1a<\/strong> * \u9023\u9396\u5f8b (Chain Rule) \u2014\u2014 \u6df1\u5ea6\u5b78\u7fd2<strong>\u53cd\u5411\u50b3\u64ad\u7b97\u6cd5 (Backpropagation)<\/strong> \u7684\u6838\u5fc3\u3002\n<ul class=\"wp-block-list\">\n<li>\u6fc0\u6d3b\u51fd\u6578\u7684\u5c0e\u6578\uff08\u5982 Sigmoid, ReLU \u7684\u5c0e\u51fd\u6578\uff09\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u5411\u91cf\u5fae\u7a4d\u5206\uff1a<\/strong> \u68af\u5ea6 ($\\nabla$, Gradient)\u3001\u96c5\u53ef\u6bd4\u77e9\u9663 (Jacobian Matrix)\u3001\u6d77\u68ee\u77e9\u9663 (Hessian Matrix)\u3002<\/li>\n\n\n\n<li><strong>AI \u61c9\u7528\uff1a<\/strong> \u8a08\u7b97\u640d\u5931\u51fd\u6578\u7684\u659c\u7387\uff0c\u6307\u5f15\u6a21\u578b\u53c3\u6578\u8abf\u6574\u7684\u65b9\u5411\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u6838\u5fc3\u652f\u67f1\u4e09\uff1a\u6a5f\u7387\u8207\u7d71\u8a08 (Probability &amp; Statistics) \u2014\u2014 \u8655\u7406\u4e0d\u78ba\u5b9a\u6027<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">AI \u7684\u672c\u8cea\u662f\u5728\u5145\u6eff\u96dc\u8a0a\u7684\u73fe\u5be6\u4e16\u754c\u4e2d\u505a\u51fa\u300c\u6700\u4f73\u731c\u6e2c\u300d\uff0c\u6a5f\u7387\u8ad6\u63d0\u4f9b\u4e86\u91cf\u5316\u4e0d\u78ba\u5b9a\u6027\u7684\u5de5\u5177\uff0c\u7d71\u8a08\u5b78\u5247\u63d0\u4f9b\u4e86\u5f9e\u8cc7\u6599\u4e2d\u62bd\u53d6\u51fa\u898f\u5f8b\u7684\u65b9\u6cd5\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u6a5f\u7387\u57fa\u790e\uff1a<\/strong> \u689d\u4ef6\u6a5f\u7387\u3001\u8c9d\u6c0f\u5b9a\u7406 (Bayes&#8217; Theorem)\u3002<\/li>\n\n\n\n<li><strong>\u96a8\u6a5f\u8b8a\u6578\u8207\u6a5f\u7387\u5206\u4f48\uff1a<\/strong> \u96e2\u6563\u8207\u9023\u7e8c\u5206\u4f48\uff08\u9ad8\u65af\/\u5e38\u614b\u5206\u4f48\u3001\u4e8c\u9805\u5206\u4f48\u3001\u5c0d\u6578\u5e38\u614b\u5206\u4f48\u3001Poisson \u5206\u4f48\uff09\u3002<\/li>\n\n\n\n<li><strong>\u7d71\u8a08\u63a8\u8ad6\u8207\u4f30\u8a08\uff1a<\/strong> * \u6700\u5927\u6982\u4f3c\u4f30\u8a08 (MLE) \u8207 \u6700\u5927\u5f8c\u9a57\u6a5f\u7387\u4f30\u8a08 (MAP)\u3002\n<ul class=\"wp-block-list\">\n<li>\u671f\u671b\u503c\u3001\u8b8a\u7570\u6578\u3001\u5354\u65b9\u5dee\u77e9\u9663 (Covariance Matrix)\u3002<\/li>\n\n\n\n<li>\u5047\u8a2d\u6aa2\u5b9a (Hypothesis Testing)\u3001\u986f\u8457\u6027\u5dee\u7570\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>AI \u61c9\u7528\uff1a<\/strong> \u9810\u6e2c\u5206\u985e\u6a5f\u7387\u3001\u64f4\u6563\u6a21\u578b\uff08Diffusion Models\uff09\u7684\u53bb\u566a\u3001\u5927\u578b\u8a9e\u8a00\u6a21\u578b\u7684\u4e0b\u4e00\u500b\u5b57\u9810\u6e2c\uff08Token Prediction\uff09\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u6838\u5fc3\u652f\u67f1\u56db\uff1a\u6700\u4f73\u5316\u7406\u8ad6 (Optimization Theory) \u2014\u2014 \u5c0b\u627e\u6700\u5b8c\u7f8e\u7684\u89e3\u7b54<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u5fae\u7a4d\u5206\u7d66\u4e86\u6211\u5011\u5de5\u5177\uff0c\u6700\u4f73\u5316\u7406\u8ad6\u5247\u7d66\u4e86\u6211\u5011\u300c\u5c0b\u627e\u6700\u4f73\u89e3\u7684\u7b56\u7565\u8207\u5730\u5716\u300d\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u7121\u7d04\u675f\u6700\u4f73\u5316\uff1a<\/strong> \u68af\u5ea6\u4e0b\u964d\u6cd5 (Gradient Descent)\u3001\u96a8\u6a5f\u68af\u5ea6\u4e0b\u964d (SGD)\u3001Adam \u512a\u5316\u5668\u3002<\/li>\n\n\n\n<li><strong>\u7d04\u675f\u6700\u4f73\u5316\uff1a<\/strong> \u62c9\u683c\u6717\u65e5\u4e58\u6578\u6cd5 (Lagrange Multipliers)\u3001KKT \u689d\u4ef6\u3002<\/li>\n\n\n\n<li><strong>\u51f8\u6700\u4f73\u5316 (Convex Optimization)\uff1a<\/strong> \u78ba\u4fdd\u80fd\u627e\u5230\u5168\u5c40\u6700\u4f73\u89e3\uff08Global Minima\uff09\u800c\u975e\u5c40\u90e8\u6700\u4f73\u89e3\uff08Local Minima\uff09\u3002<\/li>\n\n\n\n<li><strong>AI \u61c9\u7528\uff1a<\/strong> \u8b93 AI \u6a21\u578b\u5728\u8a13\u7df4\u6642\uff0c\u4ee5\u6700\u5feb\u3001\u6700\u6709\u6548\u7387\u7684\u65b9\u5f0f\u5c07\u640d\u5931\u51fd\u6578\uff08\u8aa4\u5dee\uff09\u964d\u5230\u6700\u4f4e\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u5ef6\u4f38\u9818\u57df\u4e94\uff1a\u8cc7\u8a0a\u8ad6 (Information Theory) \u2014\u2014 \u8861\u91cf\u5b78\u7fd2\u6548\u7387<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u8cc7\u8a0a\u8ad6\u539f\u672c\u7528\u65bc\u901a\u8a0a\uff0c\u4f46\u5728 AI \u4e2d\uff0c\u5b83\u88ab\u7528\u4f86\u91cf\u5316\u8cc7\u8a0a\u91cf\u4ee5\u53ca\u6a21\u578b\u9810\u6e2c\u7684\u6e96\u78ba\u5ea6\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u6838\u5fc3\u6982\u5ff5\uff1a<\/strong> \u8cc7\u8a0a\u71b5 (Entropy)\u3001\u4ea4\u53c9\u71b5 (Cross-Entropy)\u3001KL \u6563\u5ea6 (KL Divergence)\u3001\u4e92\u8cc7\u8a0a (Mutual Information)\u3002<\/li>\n\n\n\n<li><strong>AI \u61c9\u7528\uff1a<\/strong> * <strong>\u4ea4\u53c9\u71b5\u640d\u5931\u51fd\u6578\uff1a<\/strong> \u5e7e\u4e4e\u6240\u6709\u5206\u985e\u6a21\u578b\uff08\u5982\u8c93\u72d7\u8fa8\u8b58\u3001\u6587\u5b57\u5206\u985e\uff09\u7684\u6a19\u6e96\u8a55\u4f30\u6307\u6a19\u3002\n<ul class=\"wp-block-list\">\n<li><strong>KL \u6563\u5ea6\uff1a<\/strong> \u7528\u65bc VAE\uff08\u8b8a\u5206\u81ea\u7de8\u78bc\u5668\uff09\u8207 RLHF\uff08\u4eba\u985e\u56de\u994b\u5f37\u5316\u5b78\u7fd2\uff09\u4e2d\uff0c\u9632\u6b62\u6a21\u578b\u7b56\u7565\u504f\u96e2\u592a\u9060\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u5ef6\u4f38\u9818\u57df\u516d\uff1a\u96e2\u6563\u6578\u5b78\u8207\u5716\u8ad6 (Discrete Mathematics &amp; Graph Theory) \u2014\u2014 \u7d50\u69cb\u5316\u95dc\u806f<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u7576\u8cc7\u6599\u4e0d\u662f\u6574\u9f4a\u7684\u8868\u683c\uff0c\u800c\u662f\u8907\u96dc\u7684\u7db2\u8def\u6216\u908f\u8f2f\u95dc\u4fc2\u6642\uff0c\u5c31\u9700\u8981\u96e2\u6563\u6578\u5b78\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u5716\u8ad6\u57fa\u790e\uff1a<\/strong> \u7bc0\u9ede (Nodes)\u3001\u908a (Edges)\u3001\u9130\u63a5\u77e9\u9663 (Adjacency Matrix)\u3001\u6a39\u72c0\u7d50\u69cb (Trees)\u3002<\/li>\n\n\n\n<li><strong>\u6578\u7406\u908f\u8f2f\u8207\u7d44\u5408\u6578\u5b78\uff1a<\/strong> \u547d\u984c\u908f\u8f2f\u3001\u96c6\u5408\u8ad6\u3001\u6392\u5217\u7d44\u5408\u3002<\/li>\n\n\n\n<li><strong>AI \u61c9\u7528\uff1a<\/strong> * <strong>\u5716\u795e\u7d93\u7db2\u8def (GNN)\uff1a<\/strong> \u61c9\u7528\u65bc\u793e\u7fa4\u7db2\u8def\u5206\u6790\u3001\u86cb\u767d\u8cea\u5206\u5b50\u7d50\u69cb\u9810\u6e2c\u3001\u63a8\u85a6\u7cfb\u7d71\u3002\n<ul class=\"wp-block-list\">\n<li><strong>\u77e5\u8b58\u5716\u8b5c (Knowledge Graph)\uff1a<\/strong> \u8b93 AI \u5177\u5099\u908f\u8f2f\u63a8\u7406\u80fd\u529b\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u5ef6\u4f38\u9818\u57df\u4e03\uff1a\u6578\u503c\u5206\u6790\u8207\u9ad8\u7d1a\u5e7e\u4f55 (Advanced Domains) \u2014\u2014 \u7a69\u5b9a\u6027\u8207\u524d\u6cbf\u67b6\u69cb<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u9019\u5c6c\u65bc\u8f03\u70ba\u9ad8\u968e\u6216\u7279\u5b9a\u524d\u6cbf\u9818\u57df\u624d\u6703\u6df1\u5165\u6d89\u7375\u7684\u6578\u5b78\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u6578\u503c\u5206\u6790 (Numerical Analysis)\uff1a<\/strong> * \u7814\u7a76\u6d6e\u9ede\u6578\u904b\u7b97\u7684\u8aa4\u5dee\u3001\u77e9\u9663\u904b\u7b97\u7684\u6578\u503c\u7a69\u5b9a\u6027\uff08\u907f\u514d\u68af\u5ea6\u7206\u70b8\u6216\u6d88\u5931\uff09\u3002<\/li>\n\n\n\n<li><strong>\u5fae\u5206\u5e7e\u4f55\u8207\u62d3\u64b2\u5b78 (Differential Geometry &amp; Topology)\uff1a<\/strong>\n<ul class=\"wp-block-list\">\n<li><strong>\u5e7e\u4f55\u6df1\u5ea6\u5b78\u7fd2 (Geometric Deep Learning)\uff1a<\/strong> \u7814\u7a76\u975e\u6b50\u5e7e\u91cc\u5f97\u7a7a\u9593\uff08\u5982 3D \u9ede\u96f2\u3001\u6d41\u5f62\u7a7a\u9593 Manifold\uff09\u7684\u8cc7\u6599\u67b6\u69cb\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u5be6\u5206\u6790\u8207\u6e2c\u5ea6\u8ad6 (Real Analysis &amp; Measure Theory)\uff1a<\/strong>\n<ul class=\"wp-block-list\">\n<li>\u7528\u65bc\u56b4\u683c\u8b49\u660e\u6a5f\u7387\u8ad6\u8207\u6df1\u5ea6\u5b78\u7fd2\u7406\u8ad6\u7684\u5e95\u5c64\u6536\u6582\u6027\uff08\u504f\u5411\u5b78\u8853\u7814\u7a76\uff09\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">\ud83d\udca1 \u5b78\u7fd2\u5efa\u8b70\u7684\u512a\u5148\u9806\u5e8f<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">\u5982\u679c\u4f60\u662f AI \u7684\u521d\u5b78\u8005\u6216\u61c9\u7528\u7aef\u5de5\u7a0b\u5e2b\uff0c\u5efa\u8b70\u7684\u9ede\u6280\u80fd\u9806\u5e8f\u70ba\uff1a<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u7dda\u6027\u4ee3\u6578 &#8211; \u5fae\u7a4d\u5206 &#8211; \u6a5f\u7387\u8207\u7d71\u8a08 &#8211; \u6700\u4f73\u5316\u7406\u8ad6(\u68af\u5ea6\u4e0b\u964d) &#8211; \u8cc7\u8a0a\u8ad6(\u4ea4\u53c9\u71b5)<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Text Embedding<\/h2>\n\n\n\n<figure class=\"wp-block-image size-full\"><img data-dominant-color=\"d4dde4\" data-has-transparency=\"false\" style=\"--dominant-color: #d4dde4;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"254\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/05\/text-embedding-web.jpg\" alt=\"\" class=\"wp-image-8352 not-transparent\" srcset=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/05\/text-embedding-web.jpg 1024w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/05\/text-embedding-web-300x74.jpg 300w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/05\/text-embedding-web-768x191.jpg 768w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>\u6587\u672c\u5d4c\u5165\uff08Text Embedding\uff09\u6280\u8853\u8207\u61c9\u7528\u6307\u5357<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Text Embedding\uff08\u6587\u672c\u5d4c\u5165\uff09\u662f\u4e00\u7a2e\u81ea\u7136\u8a9e\u8a00\u8655\u7406\u6280\u8853\uff0c\u7528\u65bc\u5c07\u6587\u672c\u8f49\u63db\u6210\u6578\u503c\u5411\u91cf\uff0c\u4fdd\u7559\u5176\u539f\u59cb\u6587\u672c\u7684\u610f\u7fa9\u548c\u7d50\u69cb\u3002\u9019\u4e9b\u5411\u91cf\u88ab\u7a31\u70ba Embeddings\uff0c\u5b83\u5011\u53ef\u4ee5\u7528\u65bc\u8a31\u591a NLP \u4efb\u52d9\uff0c\u5982\u6587\u672c\u985e\u4f3c\u5ea6\u8a08\u7b97\u3001\u6587\u672c\u751f\u6210\u3001\u6587\u672c\u5206\u985e\u7b49\u3002 Text Embedding \u7684\u76ee\u6a19\u662f\u5c07\u6587\u672c\u8f49\u63db\u6210\u4e00\u7a2e\u53ef\u6578\u5b78\u5316\u7684\u5f62\u5f0f\uff0c\u4f7f\u5f97\u5b83\u5011\u80fd\u5920\u8207\u5176\u4ed6\u6578\u503c\u5411\u91cf\u9032\u884c\u6bd4\u8f03\u548c\u904b\u7b97\u3002\u9019\u6a23\u53ef\u4ee5\u8b93\u6a21\u578b\u5728\u8655\u7406\u6587\u672c\u6642\uff0c\u80fd\u5920\u5229\u7528\u50b3\u7d71\u7684\u795e\u7d93\u7db2\u8def\u6280\u8853\u4f86\u9032\u884c\u5206\u6790\u548c\u9810\u6e2c\u3002<\/p>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"\u7406\u89e3\u900f\u8fd9\u4e24\u4e2a\u57fa\u672c\u6982\u5ff5\uff0c\u4f60\u770b\u6240\u6709AI\u90fd\u5c06\u8c41\u7136\u5f00\u6717\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_kDmUHCKmA8E\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkDmUHCKmA8E%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/kDmUHCKmA8E\" \/><meta itemprop=\"duration\" content=\"PT8M39S\" \/><meta itemprop=\"uploadDate\" content=\"2025-09-16T08:01:48Z\" \/><\/div><div id=\"lyte_kDmUHCKmA8E\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkDmUHCKmA8E%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">\u7406\u89e3\u900f\u8fd9\u4e24\u4e2a\u57fa\u672c\u6982\u5ff5\uff0c\u4f60\u770b\u6240\u6709AI\u90fd\u5c06\u8c41\u7136\u5f00\u6717<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/kDmUHCKmA8E\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkDmUHCKmA8E%2F0.jpg\" alt=\"\u7406\u89e3\u900f\u8fd9\u4e24\u4e2a\u57fa\u672c\u6982\u5ff5\uff0c\u4f60\u770b\u6240\u6709AI\u90fd\u5c06\u8c41\u7136\u5f00\u6717\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"======= \u6578\u5b57\u8ca8\u5e63 \u64fa\u812b\u5974\u5f79 \u4eba\u5de5\u667a\u80fd \u64fa\u812b\u8089\u9ad4 \u5b87\u5b99\u6b96\u6c11 \u64fa\u812b\u820a\u4e16\u754c ======= X\uff1a https:\/\/x.com\/jidifeng ======= AI\u64ad\u5ba2\u751f\u6210\uff1a\u804a\u900fhttps:\/\/ww.Liao2.AI ======= ### \u60ac\u5ff5\/\u53cd\u5e38\u8bc6\u7c7b 1. **\u4e00\u4e2a\u7a7a\u683c\uff0c\u51ed\u4ec0\u4e48\u8017\u8d39AI\u6570\u5341\u4ebf\uff1f\u63ed\u79d8\u5927\u6a21\u578b\u773c\u4e2d\u7684\u201c\u5929\u5927\u7684\u4e8b\u201d** (\u5229\u7528\u5f00\u5934\u6700\u5f3a\u7684\u94a9\u5b50\uff0c\u5236\u9020\u60ac\u5ff5\u548c\u4ef7\u503c\u611f) 2. **\u4f60\u4ee5\u4e3aAI\u8bfb\u7684\u662f\u6587\u5b57\uff1f\u5927\u9519\u7279\u9519\uff01\u5b83\u773c\u4e2d\u7684\u4e16\u754c\u53ea\u662f\u4e00\u4e32\u6570\u5b57** (\u76f4\u63a5\u5f15\u7528\u6587\u4e2d\u6700\u5177\u98a0\u8986\u6027\u7684\u89c2\u70b9) 3. **AI\u5e76\u4e0d\u8ba4\u8bc6\u4f60\u5199\u7684\u5b57\uff0c\u90a3\u5b83\u5982\u4f55\u7406\u89e3\u4f60\u7684\uff1f** (\u63d0\u51fa\u4e00\u4e2a\u6839\u672c\u6027\u95ee\u9898\uff0c\u5f15\u53d1\u89c2\u4f17\u597d\u5947) 4. **\u4e3a\u4ec0\u4e48\u201c\u897f\u7ea2\u67ff\u201d\u662f4\u4e2a\u8bcd\uff1f\u62c6\u89e3AI\u773c\u4e2d\u88ab\u201c\u80a2\u89e3\u201d\u7684\u4e16\u754c** (\u7528\u4e00\u4e2a\u5177\u4f53\u7684\u3001\u53cd\u76f4\u89c9\u7684\u4f8b\u5b50\u4f5c\u4e3a\u5207\u5165\u70b9) 5. **AI\u5982\u4f55\u5b66\u4f1a\u201c\u770b\u4eba\u4e0b\u83dc\u789f\u201d\uff1f\u63ed\u79d8\u5b83\u8bfb\u61c2\u5bf9\u8bdd\u7684\u79d8\u5bc6\u6b66\u5668** (\u7528\u751f\u52a8\u7684\u6bd4\u55bb\u6765\u89e3\u91ca\u7279\u6b8aToken\u7684\u4f5c\u7528) ### \u79d1\u666e\/\u63ed\u79d8\u7c7b 6. **AI\u4e16\u754c\u7684\u57fa\u77f3\uff1aToken\u4e0eEmbedding\uff0c\u4e00\u6b21\u8bb2\u900f\uff01** (\u6e05\u6670\u5730\u544a\u8bc9\u89c2\u4f17\u89c6\u9891\u7684\u6838\u5fc3\u5185\u5bb9\uff0c\u627f\u8bfa\u201c\u8bb2\u900f\u201d) 7. **\u4ece\u6587\u5b57\u5230\u201c\u5b87\u5b99\u5750\u6807\u201d\uff1a\u8fd9\u624d\u662fAI\u7406\u89e3\u4e16\u754c\u7684\u771f\u6b63\u65b9\u5f0f** (\u4f7f\u7528\u6587\u4e2d\u201c\u5b87\u5b99\u5750\u6807\u201d\u8fd9\u4e2a\u7cbe\u5f69\u6bd4\u55bb\uff0c\u753b\u9762\u611f\u5341\u8db3) 8. **\u62c6\u89e3AI\u5927\u8111\uff08\u4e00\uff09\uff1a\u5b83\u5982\u4f55\u628a\u4f60\u7684\u8bdd\u5207\u6210\u788e\u7247\uff1f(Tokenizer)** (\u7cfb\u5217\u5316\u7684\u6807\u9898\uff0c\u6e05\u6670\u5b9a\u4f4d\uff0c\u9002\u5408\u505a\u6df1\u5ea6\u5185\u5bb9) 9. **AI\u7684\u201c\u7075\u9b42\u201d\u662f\u4ec0\u4e48\uff1f\u6df1\u5165\u5927\u6a21\u578b\u6700\u795e\u5947\u7684\u9ed1\u5323\u5b50\uff1aEmbedding** (\u5c06Embedding\u6bd4\u4f5c\u201c\u7075\u9b42\u201d\u548c\u201c\u9ed1\u5323\u5b50\u201d\uff0c\u5145\u6ee1\u795e\u79d8\u611f) 10. **GPT-4\u548c\u56fd\u4ea7\u6a21\u578b\u770b\u4e16\u754c\u6709\u4f55\u4e0d\u540c\uff1f\u5929\u58e4\u4e4b\u522b\u7adf\u5728\u4e00\u4e2a\u201c\u5206\u8bcd\u5668\u201d** (\u5f15\u5165\u6a21\u578b\u5bf9\u6bd4\uff0c\u589e\u52a0\u770b\u70b9\u548c\u8ba8\u8bba\u5ea6) ### \u4ef7\u503c\/\u6280\u80fd\u7c7b 11. **\u522b\u518d\u50bb\u50bb\u5730\u5582AI\u4e86\uff01\u5148\u641e\u61c2\u5b83\u600e\u4e48\u201c\u5403\u996d\u201d (Token &amp; Embedding)** (\u7528\u8f7b\u677e\u8c03\u4f83\u7684\u8bed\u6c14\uff0c\u5f3a\u8c03\u7406\u89e3\u8fd9\u4e9b\u6982\u5ff5\u7684\u5b9e\u7528\u4ef7\u503c) 12. **\u8fd9\u53ef\u80fd\u662f\u4f60\u770b\u8fc7\u6700\u901a\u4fd7\u6613\u61c2\u7684Token\u548cEmbedding\u8bb2\u89e3** (\u7ecf\u5178\u7684\u201c\u6700\u597d\u61c2\u201d\u6807\u9898\uff0c\u76f4\u63a5\u5438\u5f15\u60f3\u5f04\u61c2\u6982\u5ff5\u7684\u89c2\u4f17) 13. **\u7406\u89e3\u4e86\u8fd9\u4e24\u4e2a\u6982\u5ff5\uff0c\u4f60\u770b\u6240\u6709AI\u90fd\u5c06\u8c41\u7136\u5f00\u6717** (\u5f3a\u8c03\u77e5\u8bc6\u7684\u4ef7\u503c\uff0c\u7ed9\u89c2\u4f17\u4e00\u4e2a\u201c\u5347\u7ea7\u201d\u7684\u627f\u8bfa) 14. **AI\u80fd\u529b\u7684\u201c\u5b9a\u4e49\u8005\u201d\uff1a\u4f60\u7edd\u5bf9\u60f3\u4e0d\u5230\u201c\u5206\u8bcd\u5668\u201d\u6709\u591a\u91cd\u8981** (\u62d4\u9ad8Tokenizer\u7684\u5730\u4f4d\uff0c\u98a0\u8986\u89c2\u4f17\u7684\u666e\u904d\u8ba4\u77e5) 15. **\u6240\u6709AI\u795e\u5947\u80fd\u529b\u7684\u8d77\u70b9\uff1a\u4ece\u51b0\u51b7\u7684\u6570\u5b57\u5230\u5145\u6ee1\u201c\u8bed\u5883\u201d\u7684\u5750\u6807** (\u8ffd\u672c\u6eaf\u6e90\uff0c\u7ed9\u89c2\u4f17\u4e00\u4e2a\u5b8f\u5927\u7684\u89c6\u89d2) ### \u6545\u4e8b\/\u6bd4\u55bb\u7c7b 16. **\u4e3a\u4ec0\u4e48AI\u77e5\u9053\u201c\u738b\u5a46\u201d\u548c\u201c\u5356\u74dc\u201d\u5173\u7cfb\u8fd1\uff1f\u63ed\u79d8Embedding\u7684\u7a7a\u95f4\u9b54\u6cd5** (\u7528\u6587\u4e2d\u751f\u52a8\u7684\u4f8b\u5b50\uff0c\u8ba9\u6807\u9898\u66f4\u6709\u6545\u4e8b\u6027) 17. **\u201c\u82f9\u679c\u201d\u8fd8\u662f\u201c\u82f9\u679c\u516c\u53f8\u201d\uff1fAI\u5982\u4f55\u9760\u201c\u77ac\u95f4\u79fb\u52a8\u201d\u8bfb\u61c2\u4e0a\u4e0b\u6587** (\u5c06\u201c\u8bed\u5883\u5316\u5d4c\u5165\u201d\u6bd4\u4f5c\u201c\u77ac\u95f4\u79fb\u52a8\u201d\uff0c\u751f\u52a8\u6709\u8da3) 18. **AI\u5982\u4f55\u5b66\u4f1a\u201c\u4e3e\u4e00\u53cd\u4e09\u201d\uff1f\u79d8\u5bc6\u5c31\u5728\u8fd9\u5f20\u201c\u5b87\u5b99\u661f\u56fe\u201d\u91cc** (\u5c06Embedding\u77e9\u9635\u6bd4\u4f5c\u201c\u5b87\u5b99\u661f\u56fe\u201d\uff0c\u5145\u6ee1\u60f3\u8c61\u7a7a\u95f4) 19. **\u4e3aAI\u8d4b\u4e88\u201c\u7075\u9b42\u201d\u7684\u5168\u8fc7\u7a0b\uff0c\u6d53\u7f29\u5728\u8fd910\u5206\u949f** (\u603b\u7ed3\u89c6\u9891\u5185\u5bb9\uff0c\u5e76\u7528\u201c\u8d4b\u4e88\u7075\u9b42\u201d\u6765\u63d0\u5347\u683c\u5c40) 20. **\u4ece\u7a0b\u5e8f\u5458\u5230\u79d1\u5b66\u5bb6\uff0c\u9876\u7ea7AI\u90fd\u6234\u7740\u600e\u6837\u7684\u201c\u7279\u5236\u773c\u955c\u201d\u770b\u4e16\u754c\uff1f** (\u7528\u201c\u7279\u5236\u773c\u955c\u201d\u6bd4\u55bb\u4e0d\u540c\u5206\u8bcd\u5668\uff0c\u5f15\u51fa\u4e0d\u540c\u6a21\u578b\u7684\u7279\u70b9)\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n\n<p class=\"wp-block-paragraph\"><strong>\u5e38\u898b\u7684 Text Embedding \u65b9\u6cd5<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u96a8\u8457\u81ea\u7136\u8a9e\u8a00\u8655\u7406\u6280\u8853\u7684\u6f14\u9032\uff0cText Embedding \u7684\u751f\u6210\u65b9\u6cd5\u5df2\u5f9e\u65e9\u671f\u7684\u975c\u614b\u8a5e\u5411\u91cf\uff0c\u767c\u5c55\u81f3\u5982\u4eca\u57fa\u65bc\u5927\u8a9e\u8a00\u6a21\u578b\u7684\u52d5\u614b\u4e0a\u4e0b\u6587\u5411\u91cf\u3002\u4ee5\u4e0b\u662f\u76ee\u524d\u6700\u5e38\u898b\u4e14\u4e3b\u6d41\u7684\u65b9\u6cd5\uff1a<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>1. \u7d93\u5178\u975c\u614b\u5411\u91cf\u65b9\u6cd5\uff08Static Embeddings\uff09<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u9019\u985e\u65b9\u6cd5\u70ba\u6bcf\u500b\u55ae\u8a5e\u751f\u6210\u56fa\u5b9a\u7684\u5411\u91cf\uff0c\u8a08\u7b97\u901f\u5ea6\u5feb\uff0c\u4f46\u7121\u6cd5\u89e3\u6c7a\u4e00\u8a5e\u591a\u7fa9\u7684\u554f\u984c\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Word2Vec<\/strong>\uff1a\u662f\u4e00\u7a2e\u7528\u65bc\u5c07\u55ae\u8a5e\u8f49\u63db\u6210\u6578\u503c\u5411\u91cf\u7684\u65b9\u6cd5\uff0c\u5b83\u901a\u904e\u8a13\u7df4 Word Embeddings \u6a21\u578b\u4f86\u7372\u5f97\u6bcf\u500b\u55ae\u8a5e\u7684\u5411\u91cf\u8868\u9054\u3002<\/li>\n\n\n\n<li><strong>GloVe<\/strong>\uff1a\u662f\u4e00\u7a2e\u57fa\u65bc\u77e9\u9663\u5206\u89e3\u7684\u65b9\u6cd5\uff0c\u65e8\u5728\u7372\u53d6\u6bcf\u500b\u55ae\u8a5e\u7684\u5411\u91cf\u8868\u9054\uff0c\u4e26\u5229\u7528\u6587\u672c\u4e2d\u55ae\u8a5e\u4e4b\u9593\u7684\u95dc\u806f\u4fe1\u606f\u4f86\u9032\u884c\u5b78\u7fd2\u3002<\/li>\n\n\n\n<li><strong>FastText<\/strong>\uff1a\u7531 Facebook \u958b\u767c\u7684 Word2Vec \u5347\u7d1a\u7248\u3002\u5b83\u5f15\u5165\u4e86\u5b50\u8a5e\uff08Subword\uff09\u7684\u6982\u5ff5\uff0c\u5c07\u55ae\u8a5e\u62c6\u89e3\u70ba Character n-grams \u9032\u884c\u8a13\u7df4\uff0c\u56e0\u6b64\u80fd\u6709\u6548\u8655\u7406\u672a\u767b\u9304\u8a5e\uff08Out-of-Vocabulary, OOV\uff09\u8207\u932f\u5b57\u554f\u984c\u3002<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>2. \u52d5\u614b\u4e0a\u4e0b\u6587\u7de8\u78bc\u65b9\u6cd5\uff08Contextualized Embeddings\uff09<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u9019\u985e\u65b9\u6cd5\u5229\u7528\u9810\u8a13\u7df4\u6a21\u578b\uff0c\u80fd\u5920\u6839\u64da\u55ae\u8a5e\u5728\u53e5\u5b50\u4e2d\u7684\u4e0a\u4e0b\u6587\u4f4d\u7f6e\uff0c\u52d5\u614b\u751f\u6210\u4e0d\u540c\u7684\u5411\u91cf\u8868\u9054\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>BERT\uff08Bidirectional Encoder Representations from Transformers\uff09<\/strong>\uff1a\u662f\u4e00\u7a2e\u57fa\u65bc transformer \u67b6\u69cb\u7684\u9810\u8a13\u7df4\u6a21\u578b\uff0c\u5b83\u901a\u904e\u5c07\u6587\u672c\u8f49\u63db\u6210\u5411\u91cf\u8868\u9054\uff0c\u4e26\u4e14\u80fd\u5920\u6355\u6349\u5230\u6587\u672c\u4e2d\u7684\u9577\u8ddd\u96e2\u4f9d\u8cf4\u95dc\u4fc2\u3002<\/li>\n\n\n\n<li><strong>BGE\uff08Beijing Academy of Artificial Intelligence\uff09<\/strong>\uff1a\u5982 bge-large-zh-v1.5\uff0c\u5728 MTEB\uff08\u591a\u4efb\u52d9\u6587\u672c\u5d4c\u5165\u57fa\u6e96\uff09\u699c\u55ae\u4e0a\u540d\u5217\u524d\u8305\uff0c\u5c0d\u4e2d\u6587\u7684\u8a9e\u7fa9\u76f8\u4f3c\u5ea6\u8a08\u7b97\u8207\u6aa2\u7d22\u4efb\u52d9\u8868\u73fe\u6975\u70ba\u512a\u7570\u3002<\/li>\n\n\n\n<li><strong>mE5\uff08Multilingual E5\uff09<\/strong>\uff1a\u7531\u5fae\u8edf\u958b\u767c\u7684\u5d4c\u5165\u6a21\u578b\uff0c\u5176\u591a\u8a9e\u8a00\u7248\u672c\uff08multilingual-e5-large\uff09\u5728\u8655\u7406\u4e2d\u82f1\u6587\u6df7\u96dc\u7684\u6587\u672c\u6642\uff0c\u6548\u679c\u975e\u5e38\u7cbe\u6e96\u4e14\u7a69\u5b9a\u3002<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>3. \u524d\u6cbf\u5927\u6a21\u578b\u8207\u9577\u6587\u672c\u5d4c\u5165\u65b9\u6cd5\uff08LLM-based &amp; Long-Context Embeddings\uff09<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u56e0\u61c9\u5927\u8a9e\u8a00\u6a21\u578b\uff08LLM\uff09\u8207 RAG\uff08\u6aa2\u7d22\u589e\u5f37\u751f\u6210\uff09\u7684\u7206\u767c\uff0c\u65b0\u4e00\u4ee3\u7684 Embedding \u5177\u5099\u66f4\u5f37\u7684\u8de8\u8a9e\u8a00\u80fd\u529b\u8207\u9f90\u5927\u7684 Token \u5bb9\u7d0d\u91cf\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>OpenAI Embeddings\uff08\u5982 <\/strong><strong>text-embedding-3-large<\/strong><strong>\uff09<\/strong>\uff1a\u76ee\u524d\u5546\u696d\u61c9\u7528\u4e2d\u6700\u666e\u53ca\u7684\u65b9\u6848\u4e4b\u4e00\u3002\u652f\u63f4\u9ad8\u9054 3072 \u7dad\u5ea6\u7684\u5411\u91cf\uff0c\u5177\u5099\u6975\u5f37\u7684\u8de8\u8a9e\u8a00\u7406\u89e3\u80fd\u529b\uff0c\u4e26\u652f\u63f4\u7e2e\u6e1b\u7dad\u5ea6\u6280\u8853\u3002<\/li>\n\n\n\n<li><strong>Cohere Embeddings<\/strong>\uff1a\u5728\u4f01\u696d\u7d1a\u641c\u5c0b\u4e2d\u8868\u73fe\u51fa\u8272\uff0c\u7279\u5225\u91dd\u5c0d\u591a\u8a9e\u8a00\u4ee5\u53ca\u542b\u6709\u96dc\u8a0a\u7684\u7db2\u9801\u6587\u672c\u9032\u884c\u4e86\u6df1\u5ea6\u512a\u5316\u3002<\/li>\n\n\n\n<li><strong>Jina Embeddings<\/strong>\uff1a\u4e3b\u6253\u9577\u6587\u672c\u8655\u7406\u80fd\u529b\uff0c\u50b3\u7d71\u7684 BERT \u6a21\u578b\u901a\u5e38\u9650\u5236\u5728 512 \u500b Token\uff0c\u800c Jina \u652f\u63f4\u9ad8\u9054 8k \u751a\u81f3\u66f4\u9577\u7684\u8f38\u5165\uff0c\u9069\u5408\u5206\u6790\u6574\u7bc7\u9577\u5831\u544a\u6216\u8ad6\u6587\u3002<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Text Embedding \u7684\u91cd\u8981\u61c9\u7528<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Text Embedding \u5728\u81ea\u7136\u8a9e\u8a00\u8655\u7406\u9818\u57df\u6709\u8457\u5ee3\u6cdb\u7684\u61c9\u7528\uff0c\u4e3b\u8981\u5305\u62ec\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u6587\u672c\u985e\u4f3c\u5ea6\u8a08\u7b97\uff08Semantic Similarity\uff09<\/strong>\uff1a\u4f7f\u7528 Embeddings \u53ef\u4ee5\u6bd4\u8f03\u5169\u500b\u6587\u672c\u7684\u76f8\u4f3c\u7a0b\u5ea6\u3002\u9019\u5728\u554f\u7b54\u7cfb\u7d71\u3001\u641c\u5c0b\u5f15\u64ce\u4ee5\u53ca RAG\uff08\u6aa2\u7d22\u589e\u5f37\u751f\u6210\uff09\u7684\u77e5\u8b58\u5eab\u6aa2\u7d22\u4e2d\u662f\u6838\u5fc3\u6280\u8853\u3002<\/li>\n\n\n\n<li><strong>\u6587\u672c\u5206\u985e\uff08Text Classification\uff09<\/strong>\uff1a\u901a\u904e\u5c07\u6587\u672c\u8f49\u63db\u6210\u5411\u91cf\u8868\u9054\uff0c\u53ef\u4ee5\u9032\u884c\u60c5\u611f\u5206\u6790\u3001\u5783\u573e\u90f5\u4ef6\u904e\u6ffe\u3001\u65b0\u805e\u6a19\u7c64\u5206\u985e\u7b49 NLP \u4efb\u52d9\u3002<\/li>\n\n\n\n<li><strong>\u6587\u672c\u751f\u6210\uff08Text Generation\uff09<\/strong>\uff1a Embeddings \u53ef\u4ee5\u7528\u65bc\u751f\u6210\u65b0\u7684\u6587\u672c\uff0c\u8f14\u52a9\u5927\u6a21\u578b\u7406\u89e3\u8f38\u5165\u63d0\u793a\u8a5e\uff08Prompt\uff09\uff0c\u9032\u800c\u9054\u6210\u9ad8\u8cea\u91cf\u7684\u6587\u672c\u6458\u8981\u6216\u6587\u672c\u5b8c\u6210\u3002<\/li>\n<\/ul>\n\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"15\u5206\u949f\u5f04\u61c2Token\u548cEmbedding \u8be6\u89e3LLM\u4e0eRAG\u6570\u636e\u5904\u7406\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_BQveePDWavA\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBQveePDWavA%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/BQveePDWavA\" \/><meta itemprop=\"duration\" content=\"PT15M14S\" \/><meta itemprop=\"uploadDate\" content=\"2026-01-08T09:00:30Z\" \/><\/div><meta itemprop=\"accessibilityFeature\" content=\"captions\" \/><div id=\"lyte_BQveePDWavA\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBQveePDWavA%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">15\u5206\u949f\u5f04\u61c2Token\u548cEmbedding \u8be6\u89e3LLM\u4e0eRAG\u6570\u636e\u5904\u7406<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/BQveePDWavA\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBQveePDWavA%2F0.jpg\" alt=\"15\u5206\u949f\u5f04\u61c2Token\u548cEmbedding \u8be6\u89e3LLM\u4e0eRAG\u6570\u636e\u5904\u7406\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\u52a0\u5165\u77e5\u8bc6\u661f\u7403\uff08\u97a0\u8eac\u611f\u8c22\uff01\uff09: https:\/\/t.zsxq.com\/ubYr8 https:\/\/www.youtube.com\/watch?v=mpGiFuRYrDk https:\/\/www.youtube.com\/watch?v=PoQLjK2hlZk AI\u7cfb\u5217\u8bfe\u7a0b\u5217\u8868\uff1ahttps:\/\/www.youtube.com\/playlist?list=PLeYA2mrASg3E5ctJc80FgMRGb9m3NqFsu --- 00:00:00 Token 00:03:39 Token\u7684\u95ee\u9898 00:05:42 Embedding 00:08:55 Embedding\u5b9e\u73b0 00:10:51 RAG Embedding 00:14:14 \u54f2\u5b66\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"Learn Text Embeddings in 20 Minutes (full guide for beginners)\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_Q6TBHDgWCDQ\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FQ6TBHDgWCDQ%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/Q6TBHDgWCDQ\" \/><meta itemprop=\"duration\" content=\"PT19M57S\" \/><meta itemprop=\"uploadDate\" content=\"2026-05-24T06:54:38Z\" \/><\/div><div id=\"lyte_Q6TBHDgWCDQ\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FQ6TBHDgWCDQ%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">Learn Text Embeddings in 20 Minutes (full guide for beginners)<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/Q6TBHDgWCDQ\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2FQ6TBHDgWCDQ%2F0.jpg\" alt=\"Learn Text Embeddings in 20 Minutes (full guide for beginners)\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\ud83d\udcdd Download the full notes from this video and code to run yourself \ud83d\udc49 https:\/\/thu-vu.kit.com\/8d439091c8 \ud83d\udce9 Get my FREE weekly AI &amp; data insights \ud83d\udc49 https:\/\/thu-vu.ck.page\/49c5ee08f6 \ud83c\udf1f Learn to build AI Projects \ud83d\udc49 https:\/\/python-course-earlybird.framer.website\/ In this video I explain the concept of text embeddings, a essential tool for large language models (LLM) and other modern generative AI models. We&#039;ll learn how text can be represented as numerical vectors using different methods. Understanding word embeddings, you gain valuable insights into natural language processing and how AI models interpret text. \ud83d\udd11 TIMESTAMPS \u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580\u2580 0:00 - What are word embeddings? 0:55 - Frequency-based methods 4:26 - Embeddings 8:56 - What is embedding space? 10:39 - How are embedding created? 16:28 - How to use pre-trained embedding models #deeplearning #ai #datascience #ThuVu\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>\n\n<figure class=\"wp-block-embed-youtube wp-block-embed is-type-video is-provider-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"lyte-wrapper\" title=\"\u9762\u8bd5\u5b98\uff1aRAG\u600e\u4e48\u5904\u7406Word\u548cPDF?\" style=\"width:853px;max-width:100%;margin:5px auto;\"><div class=\"lyMe\" id=\"WYL_90YSUJ2q-5I\" itemprop=\"video\" itemscope itemtype=\"https:\/\/schema.org\/VideoObject\"><div><meta itemprop=\"thumbnailUrl\" content=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F90YSUJ2q-5I%2Fhqdefault.jpg\" \/><meta itemprop=\"embedURL\" content=\"https:\/\/www.youtube.com\/embed\/90YSUJ2q-5I\" \/><meta itemprop=\"duration\" content=\"PT7M30S\" \/><meta itemprop=\"uploadDate\" content=\"2026-07-11T01:22:49Z\" \/><\/div><div id=\"lyte_90YSUJ2q-5I\" data-src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F90YSUJ2q-5I%2Fhqdefault.jpg\" class=\"pL\"><div class=\"tC\"><div class=\"tT\" itemprop=\"name\">\u9762\u8bd5\u5b98\uff1aRAG\u600e\u4e48\u5904\u7406Word\u548cPDF?<\/div><\/div><button tabindex=\"0\" class=\"play\"><\/button><div class=\"ctrl\"><div class=\"Lctrl\"><\/div><div class=\"Rctrl\"><\/div><\/div><\/div><noscript><a href=\"https:\/\/youtu.be\/90YSUJ2q-5I\" rel=\"nofollow\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/plugins\/wp-youtube-lyte\/lyteCache.php?origThumbUrl=https%3A%2F%2Fi.ytimg.com%2Fvi%2F90YSUJ2q-5I%2F0.jpg\" alt=\"\u9762\u8bd5\u5b98\uff1aRAG\u600e\u4e48\u5904\u7406Word\u548cPDF?\" width=\"853\" height=\"460\" \/><br \/>Watch this video on YouTube<\/a><\/noscript><meta itemprop=\"description\" content=\"\u7cfb\u7edf\u5b66\u4e60+AI\u5927\u6a21\u578b\u5b66\u4e60\u8def\u7ebf+\u5b9e\u6218\u9879\u76ee+\u89c6\u9891\u7535\u5b50\u4e66+\u9762\u8bd5\u771f\u9898 \u9700\u8981\u7684\u670b\u53cb\u4eec\uff0c\u53ef\u4ee5\u5728\u9891\u9053\u7684\u7f6e\u9876\u94fe\u63a5\u514d\u8d39\u83b7\u53d6~\"><\/div><\/div><div class=\"lL\" style=\"max-width:100%;width:853px;margin:5px auto;\"><\/div><figcaption><\/figcaption><\/figure>","protected":false},"excerpt":{"rendered":"<p>\u4eba\u5de5\u667a\u6167\uff08AI\uff09\u8207\u6a5f\u5668\u5b78\u7fd2\u7684\u5e95\u5c64\u5b8c\u5168\u662f\u7531\u6578\u5b78\u69cb\u5efa\u800c\u6210\u7684\u3002\u5982\u679c\u8981\u7cfb\u7d71\u6027\u5730\u5217\u51fa\u6240\u6709 AI \u9700\u8981\u7684\u6578\u5b78\u57fa\u790e\uff0c\u53ef\u4ee5\u5c07\u5b83\u5011\u5206\u70ba\u300c\u56db\u5927\u6838\u5fc3\u652f\u67f1\u300d\u4ee5\u53ca\u300c\u4e09\u5927\u5ef6\u4f38\u8207\u9ad8\u968e\u9818\u57df\u300d\u3002 \u4ee5\u4e0b\u662f\u70ba\u4f60\u6574\u7406\u7684\u5b8c\u6574 AI \u6578\u5b78\u5730\u5716\uff1a \u6838\u5fc3\u652f\u67f1\u4e00\uff1a\u7dda\u6027\u4ee3\u6578 (Linear Algebra) \u2014\u2014 \u7a7a\u9593\u8207\u8cc7\u6599\u7684\u8f49\u63db \u7dda\u6027\u4ee3\u6578\u662f AI \u8655\u7406\u8cc7\u6599\u7684\u8a9e\u8a00\u3002\u5728 AI \u4e2d\uff0c\u6240\u6709\u7684\u8cc7\u6599\uff08\u6587\u5b57\u3001\u5716\u7247\u3001\u8072\u97f3\uff09\u90fd\u6703\u88ab\u8f49\u63db\u6210\u5411\u91cf\u548c\u77e9\u9663\u3002 \u6838\u5fc3\u652f\u67f1\u4e8c\uff1a\u5fae\u7a4d\u5206 (Calculus) \u2014\u2014 \u6a21\u578b\u7684\u5b78\u7fd2\u8207\u512a\u5316 \u5fae\u7a4d\u5206\u8ca0\u8cac\u89e3\u6c7a AI \u5982\u4f55\u300c\u5f9e\u932f\u8aa4\u4e2d\u5b78\u7fd2\u300d\u7684\u554f\u984c\u3002\u900f\u904e\u8a08\u7b97\u8b8a\u5316\u7387\uff0cAI \u624d\u80fd\u8abf\u6574\u81ea\u8eab\u53c3\u6578\u3002 \u6838\u5fc3\u652f\u67f1\u4e09\uff1a\u6a5f\u7387\u8207\u7d71\u8a08 (Probability &amp; Statistics) \u2014\u2014 \u8655\u7406\u4e0d\u78ba\u5b9a\u6027 AI \u7684\u672c\u8cea\u662f\u5728\u5145\u6eff\u96dc\u8a0a\u7684\u73fe\u5be6\u4e16\u754c\u4e2d\u505a\u51fa\u300c\u6700\u4f73\u731c\u6e2c\u300d\uff0c\u6a5f\u7387\u8ad6\u63d0\u4f9b\u4e86\u91cf\u5316\u4e0d\u78ba\u5b9a\u6027\u7684\u5de5\u5177\uff0c\u7d71\u8a08\u5b78\u5247\u63d0\u4f9b\u4e86\u5f9e\u8cc7\u6599\u4e2d\u62bd\u53d6\u51fa\u898f\u5f8b\u7684\u65b9\u6cd5\u3002 \u6838\u5fc3\u652f\u67f1\u56db\uff1a\u6700\u4f73\u5316\u7406\u8ad6 (Optimization Theory) \u2014\u2014 \u5c0b\u627e\u6700\u5b8c\u7f8e\u7684\u89e3\u7b54 \u5fae\u7a4d\u5206\u7d66\u4e86\u6211\u5011\u5de5\u5177\uff0c\u6700\u4f73\u5316\u7406\u8ad6\u5247\u7d66\u4e86\u6211\u5011\u300c\u5c0b\u627e\u6700\u4f73\u89e3\u7684\u7b56\u7565\u8207\u5730\u5716\u300d\u3002 \u5ef6\u4f38\u9818\u57df\u4e94\uff1a\u8cc7\u8a0a\u8ad6 (Information Theory) \u2014\u2014 \u8861\u91cf\u5b78\u7fd2\u6548\u7387 \u8cc7\u8a0a\u8ad6\u539f\u672c\u7528\u65bc\u901a\u8a0a\uff0c\u4f46\u5728 AI \u4e2d\uff0c\u5b83\u88ab\u7528\u4f86\u91cf\u5316\u8cc7\u8a0a\u91cf\u4ee5\u53ca\u6a21\u578b\u9810\u6e2c\u7684\u6e96\u78ba\u5ea6\u3002 \u5ef6\u4f38\u9818\u57df\u516d\uff1a\u96e2\u6563\u6578\u5b78\u8207\u5716\u8ad6 (Discrete Mathematics &amp; Graph Theory) \u2014\u2014 \u7d50\u69cb\u5316\u95dc\u806f \u7576\u8cc7\u6599\u4e0d\u662f\u6574\u9f4a\u7684\u8868\u683c\uff0c\u800c\u662f\u8907\u96dc\u7684\u7db2\u8def\u6216\u908f\u8f2f\u95dc\u4fc2\u6642\uff0c\u5c31\u9700\u8981\u96e2\u6563\u6578\u5b78\u3002 [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"wpai_generated_summary":"","footnotes":""},"class_list":["post-3801","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages\/3801","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=3801"}],"version-history":[{"count":4,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages\/3801\/revisions"}],"predecessor-version":[{"id":12865,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/pages\/3801\/revisions\/12865"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=3801"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}