{"id":5526,"date":"2024-09-30T03:29:00","date_gmt":"2024-09-29T18:29:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5526"},"modified":"2024-09-30T03:29:00","modified_gmt":"2024-09-29T18:29:00","slug":"openai-o1%e3%81%ae%e8%a9%95%e4%be%a1%e3%80%81a-preliminary-study-of-o1-in-medicine-are-we-closer-to-an-ai-doctor","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5526","title":{"rendered":"OpenAI o1\u306e\u8a55\u4fa1\u3001A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor?"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">OpenAI o1\u306e\u8a55\u4fa1\u7d50\u679c\u304c\u69d8\u3005\u51fa\u3066\u3044\u308b\u3002\u533b\u7642\u30b7\u30ca\u30ea\u30aa\u3067\u306e\u8a55\u4fa1\u306f\u7279\u306b\u8208\u5473\u6df1\u3044\u3002Gemini \u306e\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3082\u3042\u308a\u3001Claude 3.5 Opus\u306e\u5642\u3082\u3042\u308a\u3001\u5546\u7528\u30e2\u30c7\u30eb\u306e\u7af6\u4e89\u3082\u6fc0\u3057\u3044\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/developers.googleblog.com\/en\/updated-gemini-models-reduced-15-pro-pricing-increased-rate-limits-and-more\/\">Updated production-ready Gemini models, reduced 1.5 Pro pricing, increased rate limits, and more &#8211; Google Developers Blog (googleblog.com)<\/a><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor?&nbsp;<\/strong>[33.7]<br>OpenAI\u306eo1\u306f\u3001\u5f37\u5316\u5b66\u7fd2\u6226\u7565\u3092\u4f7f\u3063\u305f\u30c1\u30a7\u30fc\u30f3\u30fb\u30aa\u30d6\u30fb\u30bd\u30fc\u30c8\u6280\u8853\u3092\u4f7f\u3063\u305f\u6700\u521d\u306e\u30e2\u30c7\u30eb\u3068\u3057\u3066\u969b\u7acb\u3063\u3066\u3044\u308b\u3002 \u672c\u5831\u544a\u3067\u306f\u3001\u69d8\u3005\u306a\u533b\u7642\u30b7\u30ca\u30ea\u30aa\u306b\u304a\u3051\u308bo1\u306e\u7dcf\u5408\u7684\u306a\u63a2\u7d22\u3092\u884c\u3044\u3001\u7406\u89e3\u3001\u63a8\u8ad6\u3001\u591a\u8a00\u8a9e\u6027\u3068\u3044\u30463\u3064\u306e\u91cd\u8981\u306a\u5074\u9762\u306b\u3064\u3044\u3066\u691c\u8a0e\u3059\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2409.15277v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2409.15277v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Mon, 23 Sep 2024 17:59:43 GMT)<\/li>\n\n\n\n<li>\u300cOur analysis of o1 suggests that the enhanced reasoning ability of LLMs may (significantly) benefit their capability to understand various medical instructions and reason through complex clinical scenarios.\u300d\u3068\u306e\u8a55\u4fa1\u3067\u3001GPT-4o\u30843.5\u3092\u4e0a\u56de\u308b\u7d50\u679c\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/ucsc-vlaa.github.io\/o1_medicine\/\">A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor? (ucsc-vlaa.github.io)<\/a><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>A Case Study of Web App Coding with OpenAI Reasoning Models&nbsp;<\/strong>[1.7]<br>\u6211\u3005\u306f,OpenAI\u306e\u6700\u65b0\u63a8\u8ad6\u30e2\u30c7\u30eb\u3067\u3042\u308bo1-preview\u3068o1-mini\u306b\u3088\u308b\u30b3\u30fc\u30c7\u30a3\u30f3\u30b0\u30bf\u30b9\u30af\u306e\u30b1\u30fc\u30b9\u30b9\u30bf\u30c7\u30a3\u3092,\u4ed6\u306e\u30d5\u30ed\u30f3\u30c6\u30a3\u30a2\u30e2\u30c7\u30eb\u3068\u6bd4\u8f03\u3057\u305f\u3002 o1\u30e2\u30c7\u30eb\u306f\u3001\u30b7\u30f3\u30b0\u30eb\u30bf\u30b9\u30af\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308bWebApp1K\u306b\u5bfe\u3057\u3066\u3001SOTA\u7d50\u679c\u3092\u63d0\u4f9b\u3059\u308b\u3002\u3053\u306e\u7d50\u679c\u3001WebApp1K-Duo\u306f\u3001\u591a\u304f\u306e\u30bf\u30b9\u30af\u3068\u30c6\u30b9\u30c8\u30b1\u30fc\u30b9\u3092\u500d\u306b\u3059\u308b\u3001\u3088\u308a\u96e3\u3057\u3044\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2409.13773v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2409.13773v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Thu, 19 Sep 2024 06:58:02 GMT)<\/li>\n\n\n\n<li>WebApp1K\uff08<a href=\"https:\/\/github.com\/onekq\/WebApp1k\">GitHub &#8211; onekq\/WebApp1k: WebApp1k benchmark<\/a>\uff09\u306b\u5bfe\u3057\u3066\u306fo1\u304cSoTA\u3067\u3042\u308b\u4e00\u65b9\u3067\u3001\u3088\u308a\u9577\u3044\u51fa\u529b\u304c\u8981\u6c42\u3055\u308c\u308bWebApp1K-Duo\uff08<a href=\"https:\/\/huggingface.co\/datasets\/onekq-ai\/WebApp1K-Duo-React\">onekq-ai\/WebApp1K-Duo-React \u00b7 Datasets at Hugging Face<\/a>\uff09\u3067\u306fClaude 3.5 sonnet\u306b\u8ca0\u3051\u308b\u7d50\u679c\u3002<\/li>\n\n\n\n<li>\u300cSpeci\ufb01cally, the reasoning mechanism boosts performance when all expectations are captured, meanwhile exacerbates errors when key expectations are missed, potentially impacted by input lengths.\u300d\u3068\u3044\u3046\u6307\u6458\u304c\u8208\u5473\u6df1\u3044\u3002<\/li>\n\n\n\n<li><a href=\"https:\/\/huggingface.co\/spaces\/onekq-ai\/WebApp1K-models-leaderboard\">WebApp1K Models Leaderboard &#8211; a Hugging Face Space by onekq-ai<\/a>\u3000\u306bLeader board\u304c\u3042\u308b<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Towards a Realistic Long-Term Benchmark for Open-Web Research Agents&nbsp;<\/strong>[0.2]<br>\u30d5\u30a1\u30a4\u30ca\u30f3\u30b9\u3084\u30b3\u30f3\u30b5\u30eb\u30c6\u30a3\u30f3\u30b0\u306b\u304a\u3044\u3066\u65e5\u5e38\u7684\u306b\u884c\u308f\u308c\u3066\u3044\u308b,\u5b9f\u4e16\u754c\u306e\u30aa\u30fc\u30d7\u30f3\u30a6\u30a7\u30d6\u7814\u7a76\u8ab2\u984c\u306b\u5bfe\u3059\u308b\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u8a55\u4fa1\u3092\u884c\u3063\u305f\u3002 \u6211\u3005\u306f\u3001o1-preview\u3001GPT-4o\u3001Claude-3.5 Sonnet\u3001Llama 3.1 (405b)\u3001GPT-4o-mini\u3068\u3044\u3063\u305f\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u3092\u69cb\u7bc9\u3057\u3001\u30c6\u30b9\u30c8\u3057\u305f\u3002 LLM\u5168\u4f53\u3067\u306f\u3001\u30b5\u30d6\u30bf\u30b9\u30af\u3092\u30b5\u30d6\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306b\u59d4\u8b72\u3059\u308b\u6a5f\u80fd\u3092\u5099\u3048\u305fReAct\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u304c\u6700\u3082\u3088\u304f\u6a5f\u80fd\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2409.14913v2\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2409.14913v2\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Wed, 25 Sep 2024 08:52:49 GMT)<\/li>\n\n\n\n<li>\u8907\u6570\u306e\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u306b\u3088\u308b\u8a55\u4fa1\u3001\u7dcf\u5408\u7684\u306bo1\u306f\u5f37\u529b\u3067\u306f\u3042\u308b\u304c\u3001\u30bf\u30b9\u30af\u3084\u4f7f\u3044\u65b9\u306b\u3088\u308b\u5dee\u7570\u306f\u5927\u304d\u305d\u3046\u306b\u898b\u3048\u308b\u3002<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Can GPT-O1 Kill All Bugs? An Evaluation of GPT-Family LLMs on QuixBugs&nbsp;<\/strong>[2.2]<br>\u3053\u306e\u4f5c\u696d\u306f\u3001\u6700\u8fd1\u306eGPT-o1\u30e2\u30c7\u30eb\u306e\u516c\u958b\u30ea\u30ea\u30fc\u30b9\u306b\u30a4\u30f3\u30b9\u30d1\u30a4\u30a2\u3055\u308c\u3066\u3044\u308b\u3002 \u81ea\u52d5\u30d7\u30ed\u30b0\u30e9\u30e0\u4fee\u5fa9(APR)\u306b\u304a\u3051\u308bGPT\u30d5\u30a1\u30df\u30ea\u30fc\u30e2\u30c7\u30eb\u306e\u7570\u306a\u308b\u30d0\u30fc\u30b8\u30e7\u30f3\u306e\u6709\u52b9\u6027\u306e\u6bd4\u8f03\u3092\u884c\u3063\u305f\u3002 O1\u306e\u4fee\u5fa9\u6a5f\u80fd\u306f\u3001\u4ee5\u524d\u306eGPT\u30d5\u30a1\u30df\u30ea\u30fc\u30e2\u30c7\u30eb\u3088\u308a\u3082\u512a\u308c\u3066\u304a\u308a\u3001\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u306740\u306e\u30d0\u30b0\u3092\u4fee\u6b63\u3067\u304d\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2409.10033v2\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2409.10033v2\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Tue, 17 Sep 2024 01:49:17 GMT)<\/li>\n\n\n\n<li>\u30d0\u30b0\u4fee\u6b63\u306b\u304a\u3051\u308bo1\u306e\u8a55\u4fa1\u3002GPT-4o\u3092\u8d85\u3048\u3066\u3044\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/Tomsawyerhu\/GPT-O1-on-QuixBugs\">GitHub &#8211; Tomsawyerhu\/GPT-O1-on-QuixBugs: Evaluating GPT-o1 on QuixBugs benchmark.<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>OpenAI o1\u306e\u8a55\u4fa1\u7d50\u679c\u304c\u69d8\u3005\u51fa\u3066\u3044\u308b\u3002\u533b\u7642\u30b7\u30ca\u30ea\u30aa\u3067\u306e\u8a55\u4fa1\u306f\u7279\u306b\u8208\u5473\u6df1\u3044\u3002Gemini \u306e\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3082\u3042\u308a\u3001Claude 3.5 Opus\u306e\u5642\u3082\u3042\u308a\u3001\u5546\u7528\u30e2\u30c7\u30eb\u306e\u7af6\u4e89\u3082\u6fc0\u3057\u3044\u3002 Updated productio &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=5526\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;OpenAI o1\u306e\u8a55\u4fa1\u3001A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor?&#8221; \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[232,283],"class_list":["post-5526","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-lrm","tag-o1"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5526","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=5526"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5526\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5526"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5526"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5526"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}