{"id":6851,"date":"2023-10-27T19:21:39","date_gmt":"2023-10-27T19:21:39","guid":{"rendered":"https:\/\/dailyai.com\/?p=6851"},"modified":"2023-10-27T22:55:24","modified_gmt":"2023-10-27T22:55:24","slug":"ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization","status":"publish","type":"post","link":"https:\/\/dailyai.com\/de\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/","title":{"rendered":"AI in Tests zur Theory of Mind und zur systematischen Verallgemeinerung"},"content":{"rendered":"<p><b>Forscher haben FANToM eingef\u00fchrt, einen neuartigen Benchmark, mit dem das Verst\u00e4ndnis und die Anwendung der Theory of Mind (ToM) von gro\u00dfen Sprachmodellen (LLMs) rigoros getestet und bewertet werden kann.<\/b><\/p>\n<p><span style=\"font-weight: 400;\">Unter Theory of Mind versteht man die F\u00e4higkeit, sich selbst und anderen \u00dcberzeugungen, W\u00fcnsche und Wissen zuzuordnen und zu verstehen, dass andere andere \u00dcberzeugungen und Sichtweisen haben als man selbst.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">ToM wird als Grundlage f\u00fcr das Bewusstsein intelligenter Tiere angesehen. Neben dem Menschen verf\u00fcgen auch Primaten wie Orang-Utans, Gorillas und Schimpansen sowie einige Nichtprimaten wie Papageien und Rabenv\u00f6gel (Kr\u00e4hen) \u00fcber ToM.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Da KI-Modelle immer komplexer werden, suchen KI-Forscher nach neuen Methoden zur Bewertung von F\u00e4higkeiten wie ToM.<\/span><\/p>\n<p><a href=\"https:\/\/hyunw.kim\/fantom\/\"><span style=\"font-weight: 400;\">Ein neuer Benchmark namens FANToM<\/span><\/a><span style=\"font-weight: 400;\">das von Forschern des Allen Institute for AI, der University of Washington, der Carnegie Mellon University und der Seoul National University entwickelt wurde, unterzieht Modelle f\u00fcr maschinelles Lernen dynamischen Szenarien, die reale Interaktionen widerspiegeln.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In FANToM treten die Charaktere in Gespr\u00e4che ein und aus, was die KI-Modelle vor die Herausforderung stellt, genau zu wissen, wer was zu einem bestimmten Zeitpunkt wei\u00df.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Bei der Pr\u00fcfung gro\u00dfer Sprachmodelle (LLMs) mit FANToM zeigte sich, dass selbst die fortschrittlichsten Modelle Schwierigkeiten haben, eine konsistente ToM aufrechtzuerhalten.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die Leistung der Modelle war deutlich geringer als die der menschlichen Teilnehmer, was die Grenzen der KI beim Verstehen und Steuern komplexer sozialer Interaktionen verdeutlicht.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Tats\u00e4chlich dominierten die Menschen in jeder Kategorie, wie unten dargestellt.\u00a0<\/span><\/p>\n<figure id=\"attachment_6852\" aria-describedby=\"caption-attachment-6852\" style=\"width: 1024px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-6852 size-large\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-1024x213.png\" alt=\"AI ToM\" width=\"1024\" height=\"213\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-1024x213.png 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-300x63.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-768x160.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-370x77.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-800x167.png 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-740x154.png 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-20x4.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-1320x275.png 1320w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart-230x48.png 230w, https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/scores_barchart.png 1367w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption id=\"caption-attachment-6852\" class=\"wp-caption-text\">Menschen waren bei der Beantwortung von ToM-bezogenen Fragen im Vergleich zu beliebten LLMs weit \u00fcberlegen. Quelle: <a href=\"https:\/\/hyunw.kim\/fantom\/\">FANToM<\/a>.<\/figcaption><\/figure>\n<p><span style=\"font-weight: 400;\">Ein interessanter Nebenaspekt ist, dass die Oktober-Version des GPT-4-Modells von einer fr\u00fcheren Juni-Version \u00fcbertroffen wurde, was die j\u00fcngsten Anekdoten von Nutzern best\u00e4tigen k\u00f6nnte, dass <\/span><a href=\"https:\/\/dailyai.com\/de\/2023\/07\/is-chatgpt-getting-worse-heres-everything-we-know-so-far\/\"><span style=\"font-weight: 400;\">ChatGPT wird immer schlechter<\/span><\/a><span style=\"font-weight: 400;\">.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">FANToM hat auch Techniken zur Verbesserung des LLM ToM aufgezeigt, wie z.B. das Denken in Gedankenketten und andere Feinabstimmungsmethoden. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die Kluft zwischen KI und menschlichen ToM-F\u00e4higkeiten ist jedoch nach wie vor gro\u00df.\u00a0<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">KI n\u00e4hert sich menschen\u00e4hnlichen Sprachkenntnissen<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">In einem etwas verwandten, aber separaten <\/span><a href=\"https:\/\/www.nature.com\/articles\/d41586-023-03272-3\"><span style=\"font-weight: 400;\">Studie ver\u00f6ffentlicht in Nature<\/span><\/a><span style=\"font-weight: 400;\">haben Wissenschaftler ein neuronales Netz entwickelt, das in der Lage ist, Sprache \u00e4hnlich wie der Mensch zu verallgemeinern.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dieses neue neuronale Netz zeigte eine beeindruckende F\u00e4higkeit, neu gelernte W\u00f6rter in seinen bestehenden Wortschatz zu integrieren. Es konnte diese W\u00f6rter dann in verschiedenen Kontexten verwenden - eine kognitive F\u00e4higkeit, die als systematische Generalisierung bekannt ist.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Menschen zeigen von Natur aus eine systematische Verallgemeinerung und nehmen neue Vokabeln nahtlos in ihr Repertoire auf.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Wenn jemand zum Beispiel den Begriff \"Fotobombe\" gelernt hat, kann er ihn fast sofort in verschiedenen Situationen anwenden. St\u00e4ndig tauchen neue Slangs auf, und die Menschen nehmen sie ganz nat\u00fcrlich in ihren Wortschatz auf.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die Forscher unterzogen sowohl ihr eigenes benutzerdefiniertes neuronales Netzwerk als auch ChatGPT einer Reihe von Tests und stellten fest, dass ChatGPT in der Leistung hinter dem benutzerdefinierten Modell zur\u00fcckblieb.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">W\u00e4hrend LLMs wie ChatGPT in vielen Konversationsszenarien hervorragende Leistungen erbringen, weisen sie in anderen Szenarien auff\u00e4llige Inkonsistenzen und L\u00fccken auf, ein Problem, das dieses neue neuronale Netzwerk angeht.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Um diesen Aspekt der sprachlichen Kommunikation zu untersuchen, f\u00fchrten die Forscher ein Experiment mit 25 menschlichen Teilnehmern durch, bei dem ihre F\u00e4higkeit, neu gelernte W\u00f6rter in verschiedenen Kontexten anzuwenden, bewertet wurde. <\/span><span style=\"font-weight: 400;\">Die Probanden wurden in eine Pseudosprache eingef\u00fchrt, die aus Nonsens-W\u00f6rtern besteht, die verschiedene Handlungen und Regeln darstellen.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Nach einer Trainingsphase wendeten die Teilnehmer diese abstrakten Regeln mit Bravour auf neue Situationen an und zeigten so eine systematische Generalisierung.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Als das neu entwickelte neuronale Netz mit dieser Aufgabe konfrontiert wurde, entsprach es der menschlichen Leistung. <\/span><span style=\"font-weight: 400;\">Als ChatGPT jedoch der gleichen Herausforderung ausgesetzt wurde, hatte es erhebliche Schwierigkeiten und versagte je nach Aufgabe in 42 bis 86% der Zeit.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dies ist aus zwei Gr\u00fcnden von Bedeutung. Erstens k\u00f6nnte man argumentieren, dass dieses neue neuronale Netz das GPT-4 bei dieser speziellen Aufgabe tats\u00e4chlich \u00fcbertrifft - was schon beeindruckend genug ist. Zweitens zeigt diese Studie neue Methoden auf, mit denen KI-Modellen beigebracht werden kann, neue Sprache wie Menschen zu verallgemeinern.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Wie Elia Bruni, Spezialist f\u00fcr die Verarbeitung nat\u00fcrlicher Sprache an der Universit\u00e4t Osnabr\u00fcck in Deutschland, beschreibt, \"ist es eine gro\u00dfe Sache, Systematik in neuronale Netze zu bringen\".<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Zusammen bieten diese beiden Studien neue Ans\u00e4tze f\u00fcr das Training intelligenterer KI-Modelle, die es in wichtigen Bereichen wie Linguistik und Theory of Mind mit dem Menschen aufnehmen k\u00f6nnen. <\/span><\/p>","protected":false},"excerpt":{"rendered":"<p>Forscher haben FANToM eingef\u00fchrt, ein neuartiges Benchmarking, mit dem das Verst\u00e4ndnis und die Anwendung der Theory of Mind (ToM) von gro\u00dfen Sprachmodellen (LLMs) rigoros getestet und bewertet werden soll. Unter Theory of Mind versteht man die F\u00e4higkeit, sich selbst und anderen \u00dcberzeugungen, W\u00fcnsche und Wissen zuzuordnen und zu verstehen, dass andere Menschen andere \u00dcberzeugungen und Perspektiven haben als man selbst.  Die Theory of Mind wird als Grundlage f\u00fcr das Bewusstsein intelligenter Tiere angesehen. Neben dem Menschen verf\u00fcgen auch Primaten wie Orang-Utans, Gorillas und Schimpansen \u00fcber ToM, ebenso wie einige Nicht-Primaten wie Papageien und Rabenv\u00f6gel (Kr\u00e4hen).  Als KI-Modelle<\/p>","protected":false},"author":2,"featured_media":6853,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[404,115,105,93],"class_list":["post-6851","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-agi","tag-chatgpt","tag-machine-learning","tag-openai"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>AI subjected to tests on Theory of Mind and systematic generalization | DailyAI<\/title>\n<meta name=\"description\" content=\"Researchers have introduced FANToM, a novel benchmark designed to rigorously test and evaluate large language models\u2019 (LLMs) understanding and application of Theory of Mind (ToM).\" \/>\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\/de\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI subjected to tests on Theory of Mind and systematic generalization | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Researchers have introduced FANToM, a novel benchmark designed to rigorously test and evaluate large language models\u2019 (LLMs) understanding and application of Theory of Mind (ToM).\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/de\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-27T19:21:39+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-27T22:55:24+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_435449356.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"667\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Sam Jeans\" \/>\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=\"Verfasst von\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sam Jeans\" \/>\n\t<meta name=\"twitter:label2\" content=\"Gesch\u00e4tzte Lesezeit\" \/>\n\t<meta name=\"twitter:data2\" content=\"4\u00a0Minuten\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"AI subjected to tests on Theory of Mind and systematic generalization\",\"datePublished\":\"2023-10-27T19:21:39+00:00\",\"dateModified\":\"2023-10-27T22:55:24+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/\"},\"wordCount\":665,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_435449356.jpg\",\"keywords\":[\"AGI\",\"ChatGPT\",\"machine learning\",\"OpenAI\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"de\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/\",\"name\":\"AI subjected to tests on Theory of Mind and systematic generalization | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_435449356.jpg\",\"datePublished\":\"2023-10-27T19:21:39+00:00\",\"dateModified\":\"2023-10-27T22:55:24+00:00\",\"description\":\"Researchers have introduced FANToM, a novel benchmark designed to rigorously test and evaluate large language models\u2019 (LLMs) understanding and application of Theory of Mind (ToM).\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_435449356.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_435449356.jpg\",\"width\":1000,\"height\":667,\"caption\":\"Theory of Mind AI\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI subjected to tests on Theory of Mind and systematic generalization\"}]},{\"@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\":\"de\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@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\\\/711e81f945549438e8bbc579efdeb3c9\",\"name\":\"Sam Jeans\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"caption\":\"Sam Jeans\"},\"description\":\"Sam is a science and technology writer who has worked in various AI startups. When he\u2019s not writing, he can be found reading medical journals or digging through boxes of vinyl records.\",\"sameAs\":[\"https:\\\/\\\/www.linkedin.com\\\/in\\\/sam-jeans-6746b9142\\\/\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/de\\\/author\\\/samjeans\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"KI unterzieht sich Tests zur Theory of Mind und zur systematischen Verallgemeinerung | DailyAI","description":"Forscher haben FANToM eingef\u00fchrt, einen neuartigen Benchmark, mit dem das Verst\u00e4ndnis und die Anwendung der Theory of Mind (ToM) von gro\u00dfen Sprachmodellen (LLMs) rigoros getestet und bewertet werden kann.","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\/de\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/","og_locale":"de_DE","og_type":"article","og_title":"AI subjected to tests on Theory of Mind and systematic generalization | DailyAI","og_description":"Researchers have introduced FANToM, a novel benchmark designed to rigorously test and evaluate large language models\u2019 (LLMs) understanding and application of Theory of Mind (ToM).","og_url":"https:\/\/dailyai.com\/de\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/","og_site_name":"DailyAI","article_published_time":"2023-10-27T19:21:39+00:00","article_modified_time":"2023-10-27T22:55:24+00:00","og_image":[{"width":1000,"height":667,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_435449356.jpg","type":"image\/jpeg"}],"author":"Sam Jeans","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Verfasst von":"Sam Jeans","Gesch\u00e4tzte Lesezeit":"4\u00a0Minuten"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"AI subjected to tests on Theory of Mind and systematic generalization","datePublished":"2023-10-27T19:21:39+00:00","dateModified":"2023-10-27T22:55:24+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/"},"wordCount":665,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_435449356.jpg","keywords":["AGI","ChatGPT","machine learning","OpenAI"],"articleSection":["Industry"],"inLanguage":"de"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/","url":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/","name":"KI unterzieht sich Tests zur Theory of Mind und zur systematischen Verallgemeinerung | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_435449356.jpg","datePublished":"2023-10-27T19:21:39+00:00","dateModified":"2023-10-27T22:55:24+00:00","description":"Forscher haben FANToM eingef\u00fchrt, einen neuartigen Benchmark, mit dem das Verst\u00e4ndnis und die Anwendung der Theory of Mind (ToM) von gro\u00dfen Sprachmodellen (LLMs) rigoros getestet und bewertet werden kann.","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/"]}]},{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_435449356.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_435449356.jpg","width":1000,"height":667,"caption":"Theory of Mind AI"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/10\/ai-subjected-to-tests-on-theory-of-mind-and-systematic-generalization\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"AI subjected to tests on Theory of Mind and systematic generalization"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"Ihre t\u00e4gliche Dosis an AI-Nachrichten","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":"de"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"de","@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\/711e81f945549438e8bbc579efdeb3c9","name":"Sam Jeans","image":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","caption":"Sam Jeans"},"description":"Sam ist ein Wissenschafts- und Technologiewissenschaftler, der in verschiedenen KI-Startups gearbeitet hat. Wenn er nicht gerade schreibt, liest er medizinische Fachzeitschriften oder kramt in Kisten mit Schallplatten.","sameAs":["https:\/\/www.linkedin.com\/in\/sam-jeans-6746b9142\/"],"url":"https:\/\/dailyai.com\/de\/author\/samjeans\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/6851","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/comments?post=6851"}],"version-history":[{"count":5,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/6851\/revisions"}],"predecessor-version":[{"id":6866,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/6851\/revisions\/6866"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media\/6853"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media?parent=6851"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/categories?post=6851"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/tags?post=6851"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}