{"id":13401,"date":"2024-07-14T14:53:31","date_gmt":"2024-07-14T14:53:31","guid":{"rendered":"https:\/\/dailyai.com\/?p=13401"},"modified":"2024-07-14T14:53:31","modified_gmt":"2024-07-14T14:53:31","slug":"ai-model-performance-is-it-reasoning-or-simply-reciting","status":"publish","type":"post","link":"https:\/\/dailyai.com\/it\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/","title":{"rendered":"Prestazioni dei modelli di intelligenza artificiale: Si tratta di ragionare o semplicemente di recitare?"},"content":{"rendered":"

Quando ChatGPT fornisce la risposta giusta alla richiesta, ragiona sulla richiesta o ricorda semplicemente la risposta dai suoi dati di addestramento?<\/strong><\/p>\n

I ricercatori del Computer Science and Artificial Intelligence Laboratory (CSAIL) del MIT hanno progettato una serie di test per verificare se i modelli di intelligenza artificiale \"pensano\" o hanno solo una buona memoria.<\/p>\n

Quando si chiede a un modello di intelligenza artificiale di risolvere un problema matematico come \"Quanto fa 27+62?\", il modello risponde rapidamente con la risposta corretta: 89. Come possiamo dire se il modello comprende l'aritmetica sottostante o se ha semplicemente visto il problema nei suoi dati di addestramento?<\/p>\n

In la loro carta<\/a>I ricercatori hanno testato GPT-4, GPT-3.5 Turbo, Claude 1.3 e PaLM2 per vedere se erano in grado di \"generalizzare non solo a istanze non viste di compiti noti, ma anche a nuovi compiti\".<\/p>\n

Hanno progettato una serie di 11 compiti che differivano leggermente dai compiti standard in cui i LLM generalmente ottengono buoni risultati.<\/p>\n

I LLM dovrebbero ottenere le stesse prestazioni con i \"compiti controfattuali\" se utilizzano procedure di risoluzione dei compiti generali e trasferibili.<\/p>\n

Se un LLM \"capisce\" la matematica, allora dovrebbe fornire la risposta corretta a un problema matematico in base 10 e nella raramente utilizzata base 9, per esempio.<\/p>\n

Ecco alcuni esempi di attivit\u00e0 e di prestazioni del GPT-4.<\/p>\n

\"\"
Prestazioni del GPT-4 con compiti standard predefiniti (blu) e compiti controfattuali leggermente modificati (arancione). Qui sono mostrati esempi di compiti e risposte corrette. Fonte: arXiv<\/figcaption><\/figure>\n

Le prestazioni del GPT-4 nei test standard (linea blu) sono buone, ma le sue abilit\u00e0 matematiche, di ragionamento logico, spaziale e di altro tipo (linea arancione) peggiorano significativamente quando il compito viene leggermente modificato.<\/p>\n

Gli altri modelli hanno mostrato un degrado simile, con il GPT-4 in testa.<\/p>\n

Nonostante il degrado, le prestazioni nei compiti controfattuali erano ancora migliori del caso. I modelli di intelligenza artificiale cercano di ragionare su questi compiti, ma non sono molto bravi.<\/p>\n

I risultati mostrano che le impressionanti prestazioni dei modelli di intelligenza artificiale in compiti come gli esami universitari si basano su un eccellente richiamo dei dati di formazione, non sul ragionamento. Ci\u00f2 evidenzia ulteriormente che i modelli di intelligenza artificiale non possono generalizzarsi a compiti inediti,<\/p>\n

Zhaofeng Wu, dottorando in ingegneria elettrica e informatica del MIT, affiliato al CSAIL e autore principale del lavoro, ha dichiarato: \"Abbiamo scoperto un aspetto affascinante dei modelli linguistici di grandi dimensioni: eccellono in scenari familiari, quasi come un sentiero ben battuto, ma faticano quando il terreno diventa sconosciuto\". Questa intuizione \u00e8 fondamentale per migliorare l'adattabilit\u00e0 di questi modelli e ampliare i loro orizzonti applicativi\".<\/p>\n

Abbiamo visto una dimostrazione simile di questa incapacit\u00e0 di generalizzare quando abbiamo esplorato quanto siano scadenti i modelli di intelligenza artificiale a risolvere un puzzle semplificato di attraversamento del fiume.<\/a><\/p>\n

I ricercatori hanno concluso che, quando gli sviluppatori analizzano i loro modelli, dovrebbero \"considerare l'abilit\u00e0 astratta del compito come distaccata dalla prestazione osservata del compito\".<\/p>\n

L'approccio \"train-to-test\" pu\u00f2 far avanzare un modello nei benchmark, ma non offre una vera misura di come il modello si comporter\u00e0 quando gli verr\u00e0 presentato un nuovo compito su cui ragionare.<\/p>\n

I ricercatori suggeriscono che parte del problema \u00e8 che questi modelli sono addestrati solo su testi di forma superficiale.<\/p>\n

Se i LLM sono esposti a dati contestualizzati e a rappresentazioni semantiche pi\u00f9 reali, potrebbero essere in grado di generalizzare quando vengono presentate variazioni del compito.<\/p>","protected":false},"excerpt":{"rendered":"

Quando ChatGPT vi d\u00e0 la risposta giusta al vostro quesito, ragiona sulla richiesta o ricorda semplicemente la risposta dai suoi dati di addestramento? I ricercatori del Computer Science and Artificial Intelligence Laboratory (CSAIL) del MIT hanno progettato una serie di test per verificare se i modelli di intelligenza artificiale \"pensano\" o hanno semplicemente una buona memoria. Quando si chiede a un modello di intelligenza artificiale di risolvere un problema matematico come \"Quanto fa 27+62?\", il modello risponde rapidamente con la risposta corretta: 89. Come possiamo dire se il modello comprende l'aritmetica sottostante o se ha semplicemente visto il problema nei suoi dati di addestramento? Nel loro articolo, i ricercatori hanno testato il GPT-4,<\/p>","protected":false},"author":6,"featured_media":13404,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[118],"class_list":["post-13401","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-llms"],"yoast_head":"\nAI model performance: Is it reasoning or simply reciting? | DailyAI<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dailyai.com\/it\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/\" \/>\n<meta property=\"og:locale\" content=\"it_IT\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI model performance: Is it reasoning or simply reciting? | DailyAI\" \/>\n<meta property=\"og:description\" content=\"When ChatGPT gives you the right answer to your prompt, does it reason through the request or simply remember the answer from its training data? MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) researchers designed a series of tests to see if AI models \u201cthink\u201d or just have good memories. When you prompt an AI model to solve a math problem like \u201cWhat is 27+62?\u201d it comes back quickly with the correct answer: 89. How could we tell if it understands the underlying arithmetic or simply saw the problem in its training data? In their paper, the researchers tested GPT-4,\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/it\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-07-14T14:53:31+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/07\/AI-reasoning.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1792\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Eugene van der Watt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"Scritto da\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tempo di lettura stimato\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minuti\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"AI model performance: Is it reasoning or simply reciting?\",\"datePublished\":\"2024-07-14T14:53:31+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/\"},\"wordCount\":532,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/AI-reasoning.webp\",\"keywords\":[\"LLMS\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"it-IT\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/\",\"name\":\"AI model performance: Is it reasoning or simply reciting? | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/AI-reasoning.webp\",\"datePublished\":\"2024-07-14T14:53:31+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#breadcrumb\"},\"inLanguage\":\"it-IT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/AI-reasoning.webp\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/07\\\/AI-reasoning.webp\",\"width\":1792,\"height\":1024},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/07\\\/ai-model-performance-is-it-reasoning-or-simply-reciting\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI model performance: Is it reasoning or simply reciting?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"name\":\"DailyAI\",\"description\":\"Your Daily Dose of AI News\",\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/dailyai.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"it-IT\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"width\":4501,\"height\":934,\"caption\":\"DailyAI\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/DailyAIOfficial\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/dailyaiofficial\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@DailyAIOfficial\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\",\"name\":\"Eugene van der Watt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"caption\":\"Eugene van der Watt\"},\"description\":\"Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.\",\"sameAs\":[\"www.linkedin.com\\\/in\\\/eugene-van-der-watt-16828119\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/it\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Prestazioni dei modelli di intelligenza artificiale: Ragionare o semplicemente recitare? | QuotidianoAI","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":"https:\/\/dailyai.com\/it\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/","og_locale":"it_IT","og_type":"article","og_title":"AI model performance: Is it reasoning or simply reciting? | DailyAI","og_description":"When ChatGPT gives you the right answer to your prompt, does it reason through the request or simply remember the answer from its training data? MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) researchers designed a series of tests to see if AI models \u201cthink\u201d or just have good memories. When you prompt an AI model to solve a math problem like \u201cWhat is 27+62?\u201d it comes back quickly with the correct answer: 89. How could we tell if it understands the underlying arithmetic or simply saw the problem in its training data? In their paper, the researchers tested GPT-4,","og_url":"https:\/\/dailyai.com\/it\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/","og_site_name":"DailyAI","article_published_time":"2024-07-14T14:53:31+00:00","og_image":[{"width":1792,"height":1024,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/07\/AI-reasoning.webp","type":"image\/webp"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Scritto da":"Eugene van der Watt","Tempo di lettura stimato":"3 minuti"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"AI model performance: Is it reasoning or simply reciting?","datePublished":"2024-07-14T14:53:31+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/"},"wordCount":532,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/07\/AI-reasoning.webp","keywords":["LLMS"],"articleSection":["Industry"],"inLanguage":"it-IT"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/","url":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/","name":"Prestazioni dei modelli di intelligenza artificiale: Ragionare o semplicemente recitare? | QuotidianoAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/07\/AI-reasoning.webp","datePublished":"2024-07-14T14:53:31+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#breadcrumb"},"inLanguage":"it-IT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/"]}]},{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/07\/AI-reasoning.webp","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/07\/AI-reasoning.webp","width":1792,"height":1024},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/07\/ai-model-performance-is-it-reasoning-or-simply-reciting\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"AI model performance: Is it reasoning or simply reciting?"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"La vostra dose quotidiana di notizie sull'intelligenza artificiale","publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dailyai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"it-IT"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","width":4501,"height":934,"caption":"DailyAI"},"image":{"@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DailyAIOfficial","https:\/\/www.linkedin.com\/company\/dailyaiofficial\/","https:\/\/www.youtube.com\/@DailyAIOfficial"]},{"@type":"Person","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa","name":"Eugene van der Watt","image":{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","caption":"Eugene van der Watt"},"description":"Eugene proviene da un background di ingegneria elettronica e ama tutto ci\u00f2 che \u00e8 tecnologico. Quando si prende una pausa dal consumo di notizie sull'intelligenza artificiale, lo si pu\u00f2 trovare al tavolo da biliardo.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/it\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts\/13401","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/comments?post=13401"}],"version-history":[{"count":3,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts\/13401\/revisions"}],"predecessor-version":[{"id":13406,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts\/13401\/revisions\/13406"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/media\/13404"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/media?parent=13401"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/categories?post=13401"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/tags?post=13401"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}