{"id":6931,"date":"2024-04-02T09:16:40","date_gmt":"2024-04-02T01:16:40","guid":{"rendered":"https:\/\/www.1ai.net\/?p=6931"},"modified":"2024-04-02T09:16:40","modified_gmt":"2024-04-02T01:16:40","slug":"%e8%8b%b9%e6%9e%9c%e7%a0%94%e7%a9%b6%e4%ba%ba%e5%91%98%e7%a7%b0%e5%85%b6%e8%ae%be%e5%a4%87%e7%ab%af%e6%a8%a1%e5%9e%8b-realm-%e6%80%a7%e8%83%bd%e4%bc%98%e4%ba%8e-gpt-4%ef%bc%8c%e5%8f%af%e5%a4%a7","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/6931.html","title":{"rendered":"Apple researchers say their on-device model ReALM outperforms GPT-4 and can significantly improve Siri&#039;s intelligence"},"content":{"rendered":"<p data-vmark=\"952c\">Although currently <a href=\"https:\/\/www.1ai.net\/en\/tag\/siri\" title=\"_Other Organiser\" target=\"_blank\" >Siri<\/a> You can try to describe the images in the message, but the results are not stable. However,<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e8%8b%b9%e6%9e%9c\" title=\"[View articles tagged with [apple]]\" target=\"_blank\" >apple<\/a>The company hasn\u2019t given up on exploring artificial intelligence. In a recent research paper, Apple\u2019s AI team described a model that could significantly improve Siri\u2019s intelligence, and they believe the model, called AI, could help it improve its performance. <a href=\"https:\/\/www.1ai.net\/en\/tag\/realm\" title=\"_Other Organiser\" target=\"_blank\" >ReALM<\/a> The model outperformed OpenAI&#039;s well-known language model in the test <a href=\"https:\/\/www.1ai.net\/en\/tag\/gpt-4-0\" title=\"[SEE ARTICLES WITH [GPT-4.0] LABELS]\" target=\"_blank\" >GPT-4.0<\/a>.<\/p>\n<p data-vmark=\"5710\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-6932\" title=\"f0f74c31-8c82-4d96-8774-8d55ead6dd98\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/04\/f0f74c31-8c82-4d96-8774-8d55ead6dd98.jpg\" alt=\"f0f74c31-8c82-4d96-8774-8d55ead6dd98\" width=\"1500\" height=\"750\" \/><\/p>\n<p data-vmark=\"884f\">What\u2019s special about ReALM is that it can understand both what\u2019s on the user\u2019s screen and what actions they\u2019re taking at the same time. The paper divides this information into three types:<\/p>\n<ul class=\"list-paddingleft-2\">\n<li>\n<p data-vmark=\"55b7\">Screen entity: refers to the content currently displayed on the user&#039;s screen.<\/p>\n<\/li>\n<li>\n<p data-vmark=\"289d\">Dialogue entity: refers to the content related to the conversation. For example, if the user says &quot;call mom&quot;, then mom&#039;s contact information is the dialogue entity.<\/p>\n<\/li>\n<li>\n<p data-vmark=\"1493\">Background entities: refers to entities that may not be directly related to the user&#039;s current operation or the content displayed on the screen, such as the music being played or the alarm that is about to ring.<\/p>\n<\/li>\n<\/ul>\n<p data-vmark=\"619a\">If it works perfectly, ReALM will make Siri more intelligent and useful. They compared the performance of ReALM with OpenAI&#039;s GPT-3.5 and GPT-4.0:<\/p>\n<p data-vmark=\"10a9\">\u201cWe tested both the GPT-3.5 and GPT-4.0 models provided by OpenAI and provided them with contextual information to predict a range of possible entities. GPT-3.5 only accepts text input, so we only provided text prompts. GPT-4 can understand image information, so we provided it with screenshots, which significantly improved its screen entity recognition performance.\u201d<\/p>\n<p data-vmark=\"2782\">So how does Apple perform with ReALM?<\/p>\n<p data-vmark=\"38e7\">\u201cOur models have made significant progress in recognizing different types of entities. Even the smallest model has improved the accuracy of on-screen entity recognition by more than 5% compared to the original system.<span class=\"accentTextColor\">In a comparison with GPT-3.5 and GPT-4.0, our smallest model performs on par with GPT-4.0, while the larger models significantly outperform it.<\/span>. &quot;<\/p>\n<p data-vmark=\"3e24\">One of the conclusions of the paper is that<span class=\"accentTextColor\">Even with far fewer parameters than GPT-4, ReALM is able to match its performance and perform better when processing domain-specific user instructions.<\/span>, which makes ReALM a practical and efficient entity recognition system that can run on the device side.<\/p>\n<p data-vmark=\"501a\">For Apple, how to apply this technology to devices without affecting performance seems to be the key. With the WWDC 2024 Developer Conference to be held on June 10, the outside world generally expects Apple to demonstrate more artificial intelligence technology achievements in new systems such as iOS 18.<\/p>","protected":false},"excerpt":{"rendered":"<p>While Siri can currently attempt to describe images in messages, the results are inconsistent. However, Apple has not given up on exploring the field of artificial intelligence. In a recent research paper, Apple's AI team describes a model that could significantly improve Siri's intelligence, and they argue that the model, called ReALM, outperforms OpenAI's well-known language model GPT-4.0 in tests. What's special about ReALM is that it can understand both what's on the user's screen and what's being done. The paper categorizes information into the following three types: Screen entities: the content currently displayed on the user's screen. Conversation entities: This refers to the content associated with a conversation. For example, if the user says, \"Call mom\", then mom<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[2015,2014,1843,345],"collection":[],"class_list":["post-6931","post","type-post","status-publish","format-standard","hentry","category-news","tag-gpt-4-0","tag-realm","tag-siri","tag-345"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/6931","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=6931"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/6931\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=6931"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=6931"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=6931"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=6931"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}