{"id":7671,"date":"2023-11-23T15:10:15","date_gmt":"2023-11-23T15:10:15","guid":{"rendered":"https:\/\/dailyai.com\/?p=7671"},"modified":"2023-11-23T15:13:38","modified_gmt":"2023-11-23T15:13:38","slug":"system-2-attention-improves-accuracy-of-llm-responses","status":"publish","type":"post","link":"https:\/\/dailyai.com\/es\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/","title":{"rendered":"La atenci\u00f3n del Sistema 2 mejora la precisi\u00f3n de las respuestas del LLM"},"content":{"rendered":"<p><strong>Los grandes modelos ling\u00fc\u00edsticos (LLM) a menudo se dejan enga\u00f1ar por los prejuicios o el contexto irrelevante de una pregunta. Los investigadores de Meta han encontrado una forma aparentemente sencilla de solucionarlo.<\/strong><\/p>\n<p>A medida que aumentan las ventanas de contexto, las instrucciones que introducimos en un LLM pueden hacerse m\u00e1s largas y detalladas. Los LLM son cada vez m\u00e1s capaces de captar los matices o peque\u00f1os detalles de nuestras indicaciones, pero a veces esto puede confundirlos.<\/p>\n<p>Los primeros sistemas de aprendizaje autom\u00e1tico utilizaban un enfoque de \"atenci\u00f3n dura\" que seleccionaba la parte m\u00e1s relevante de una entrada y respond\u00eda s\u00f3lo a ella. Esto funciona bien cuando se trata de subtitular una imagen, pero mal cuando se traduce una frase o se responde a una pregunta con varias capas.<\/p>\n<p>La mayor\u00eda de los LLM utilizan ahora un enfoque de \"atenci\u00f3n suave\" que tokeniza toda la pregunta y asigna pesos a cada una de ellas.<\/p>\n<p>Meta propone un enfoque denominado <a href=\"https:\/\/arxiv.org\/pdf\/2311.11829.pdf\" target=\"_blank\" rel=\"noopener\">Sistema 2 Atenci\u00f3n<\/a> (S2A) para obtener lo mejor de ambos mundos. S2A utiliza la capacidad de procesamiento del lenguaje natural de un LLM para tomar su solicitud y eliminar los sesgos y la informaci\u00f3n irrelevante antes de empezar a trabajar en una respuesta.<\/p>\n<p>He aqu\u00ed un ejemplo.<\/p>\n<figure id=\"attachment_7673\" aria-describedby=\"caption-attachment-7673\" style=\"width: 1200px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-7673 size-full\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example.png\" alt=\"\" width=\"1200\" height=\"750\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example.png 1200w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-300x188.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-1024x640.png 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-768x480.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-370x231.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-800x500.png 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-20x13.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-740x463.png 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Math-example-77x48.png 77w\" sizes=\"auto, (max-width: 1200px) 100vw, 1200px\" \/><figcaption id=\"caption-attachment-7673\" class=\"wp-caption-text\">Ejemplo de S2A Math. Fuente: arXiv<\/figcaption><\/figure>\n<p>S2A se deshace de la informaci\u00f3n relativa a Max, ya que es irrelevante para la pregunta. S2A regenera una pregunta optimizada antes de empezar a trabajar en ella. Los LLM son notoriamente malos en <a href=\"https:\/\/dailyai.com\/es\/2023\/10\/chatgpts-accounting-skills-are-put-to-the-test\/\">matem\u00e1ticas<\/a> por lo que hacer el aviso menos confuso es de gran ayuda.<\/p>\n<p>A los LLM les gusta complacer a la gente y est\u00e1n encantados de estar de acuerdo contigo, incluso cuando est\u00e1s equivocado. S2A elimina cualquier sesgo en una pregunta y s\u00f3lo procesa las partes relevantes de la misma. Esto reduce lo que los investigadores de IA llaman \"adulancia\", o la propensi\u00f3n de un modelo de IA a besar culos.<\/p>\n<figure id=\"attachment_7674\" aria-describedby=\"caption-attachment-7674\" style=\"width: 1190px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-7674 size-full\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction.png\" alt=\"\" width=\"1190\" height=\"584\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction.png 1190w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-300x147.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-1024x503.png 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-768x377.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-370x182.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-800x393.png 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-740x363.png 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-20x10.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-sycophancy-reduction-98x48.png 98w\" sizes=\"auto, (max-width: 1190px) 100vw, 1190px\" \/><figcaption id=\"caption-attachment-7674\" class=\"wp-caption-text\">Reducci\u00f3n de la sicofan\u00eda S2A. Fuente: arXiv<\/figcaption><\/figure>\n<p>En realidad, S2A no es m\u00e1s que un sistema que indica al LLM que perfeccione un poco la pregunta original antes de ponerse a trabajar en ella. Los resultados que obtuvieron los investigadores con preguntas matem\u00e1ticas, objetivas y largas fueron impresionantes.<\/p>\n<p>A modo de ejemplo, he aqu\u00ed las mejoras conseguidas por S2A en las preguntas sobre hechos. La l\u00ednea de base eran las respuestas a preguntas que conten\u00edan sesgos, mientras que la indicaci\u00f3n de Oracle era una indicaci\u00f3n ideal refinada por humanos.<\/p>\n<p>S2A se acerca mucho a los resultados de Oracle y mejora la precisi\u00f3n en casi 50% con respecto a la l\u00ednea de base.<\/p>\n<figure id=\"attachment_7675\" aria-describedby=\"caption-attachment-7675\" style=\"width: 586px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-7675 size-full\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Comparison-of-results.png\" alt=\"\" width=\"586\" height=\"342\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Comparison-of-results.png 586w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Comparison-of-results-300x175.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Comparison-of-results-370x216.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Comparison-of-results-20x12.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/S2A-Comparison-of-results-82x48.png 82w\" sizes=\"auto, (max-width: 586px) 100vw, 586px\" \/><figcaption id=\"caption-attachment-7675\" class=\"wp-caption-text\">S2A Comparaci\u00f3n de resultados. Fuente: arXiv<\/figcaption><\/figure>\n<p>\u00bfCu\u00e1l es el problema? El preprocesamiento de la pregunta original antes de responderla a\u00f1ade requisitos de c\u00e1lculo adicionales al proceso. Si la pregunta es larga y contiene mucha informaci\u00f3n relevante, regenerarla puede suponer un coste considerable.<\/p>\n<p>Es poco probable que los usuarios mejoren escribiendo prompts bien elaborados, por lo que S2A puede ser una buena forma de evitarlo.<\/p>\n<p>\u00bfIncorporar\u00e1 Meta S2A a su <a href=\"https:\/\/dailyai.com\/es\/2023\/07\/meta-and-microsoft-release-advanced-ai-llama-2-for-free\/\">Llama<\/a> modelo? No lo sabemos, pero usted mismo puede aprovechar el enfoque S2A.<\/p>\n<p>Si tienes cuidado de omitir las opiniones o las sugerencias de tus preguntas, es m\u00e1s probable que obtengas respuestas precisas de estos modelos.<\/p>","protected":false},"excerpt":{"rendered":"<p>Los grandes modelos ling\u00fc\u00edsticos (LLM) a menudo se dejan enga\u00f1ar por los prejuicios o el contexto irrelevante de una pregunta. Los investigadores de Meta han encontrado una forma aparentemente sencilla de solucionar este problema. A medida que aumentan las ventanas de contexto, las instrucciones que introducimos en un LLM pueden hacerse m\u00e1s largas y detalladas. Los LLM han mejorado a la hora de captar los matices o los detalles m\u00e1s peque\u00f1os de nuestras indicaciones, pero a veces esto puede confundirlos. Las primeras m\u00e1quinas de aprendizaje utilizaban un enfoque de \"atenci\u00f3n dura\" que seleccionaba la parte m\u00e1s relevante de una entrada y s\u00f3lo respond\u00eda a ella. Esto funciona bien cuando se trata de subtitular una imagen,<\/p>","protected":false},"author":6,"featured_media":7676,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[118,131],"class_list":["post-7671","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-llms","tag-meta"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>System 2 Attention improves accuracy of LLM responses | 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\/es\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/\" \/>\n<meta property=\"og:locale\" content=\"es_ES\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"System 2 Attention improves accuracy of LLM responses | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Large Language Models (LLM) are often mislead by bias or irrelevant context in a prompt. Researchers at Meta have found a seemingly simple way to fix that. As context windows increase the prompts that we enter into an LLM can become longer and increasingly detailed. LLMs have become better at picking up on the nuances or smaller details in our prompts, but sometimes this can confuse them. Early machine learning used a \u201chard attention\u201d approach that singled out the most relevant part of an input and responded only to that. This works fine when you\u2019re trying to caption an image,\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/es\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-11-23T15:10:15+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-11-23T15:13:38+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/Simplify.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"666\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\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=\"Escrito por\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tiempo de lectura\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutos\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"System 2 Attention improves accuracy of LLM responses\",\"datePublished\":\"2023-11-23T15:10:15+00:00\",\"dateModified\":\"2023-11-23T15:13:38+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/\"},\"wordCount\":520,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/11\\\/Simplify.jpg\",\"keywords\":[\"LLMS\",\"Meta\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"es\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/\",\"name\":\"System 2 Attention improves accuracy of LLM responses | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/11\\\/Simplify.jpg\",\"datePublished\":\"2023-11-23T15:10:15+00:00\",\"dateModified\":\"2023-11-23T15:13:38+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#breadcrumb\"},\"inLanguage\":\"es\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/11\\\/Simplify.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/11\\\/Simplify.jpg\",\"width\":1000,\"height\":666},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/11\\\/system-2-attention-improves-accuracy-of-llm-responses\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"System 2 Attention improves accuracy of LLM responses\"}]},{\"@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\":\"es\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@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\":\"es\",\"@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\\\/es\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"La atenci\u00f3n del Sistema 2 mejora la precisi\u00f3n de las respuestas del LLM | DailyAI","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\/es\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/","og_locale":"es_ES","og_type":"article","og_title":"System 2 Attention improves accuracy of LLM responses | DailyAI","og_description":"Large Language Models (LLM) are often mislead by bias or irrelevant context in a prompt. Researchers at Meta have found a seemingly simple way to fix that. As context windows increase the prompts that we enter into an LLM can become longer and increasingly detailed. LLMs have become better at picking up on the nuances or smaller details in our prompts, but sometimes this can confuse them. Early machine learning used a \u201chard attention\u201d approach that singled out the most relevant part of an input and responded only to that. This works fine when you\u2019re trying to caption an image,","og_url":"https:\/\/dailyai.com\/es\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/","og_site_name":"DailyAI","article_published_time":"2023-11-23T15:10:15+00:00","article_modified_time":"2023-11-23T15:13:38+00:00","og_image":[{"width":1000,"height":666,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/Simplify.jpg","type":"image\/jpeg"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Escrito por":"Eugene van der Watt","Tiempo de lectura":"3 minutos"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"System 2 Attention improves accuracy of LLM responses","datePublished":"2023-11-23T15:10:15+00:00","dateModified":"2023-11-23T15:13:38+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/"},"wordCount":520,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/Simplify.jpg","keywords":["LLMS","Meta"],"articleSection":["Industry"],"inLanguage":"es"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/","url":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/","name":"La atenci\u00f3n del Sistema 2 mejora la precisi\u00f3n de las respuestas del LLM | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/Simplify.jpg","datePublished":"2023-11-23T15:10:15+00:00","dateModified":"2023-11-23T15:13:38+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#breadcrumb"},"inLanguage":"es","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/"]}]},{"@type":"ImageObject","inLanguage":"es","@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/Simplify.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/11\/Simplify.jpg","width":1000,"height":666},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/11\/system-2-attention-improves-accuracy-of-llm-responses\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"System 2 Attention improves accuracy of LLM responses"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"Su dosis diaria de noticias sobre IA","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":"es"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"es","@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":"es","@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 es ingeniero electr\u00f3nico y le encanta todo lo relacionado con la tecnolog\u00eda. Cuando descansa de consumir noticias sobre IA, lo encontrar\u00e1 jugando al billar.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/es\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/posts\/7671","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/comments?post=7671"}],"version-history":[{"count":7,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/posts\/7671\/revisions"}],"predecessor-version":[{"id":7682,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/posts\/7671\/revisions\/7682"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/media\/7676"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/media?parent=7671"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/categories?post=7671"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/es\/wp-json\/wp\/v2\/tags?post=7671"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}