{"id":6319,"date":"2025-03-03T06:10:00","date_gmt":"2025-03-02T21:10:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6319"},"modified":"2025-03-03T06:10:00","modified_gmt":"2025-03-02T21:10:00","slug":"claude-3-7-gpt-4-5-phi-4","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6319","title":{"rendered":"Claude 3.7, GPT-4.5, Phi-4, Selene"},"content":{"rendered":"\n<p>\u5148\u9031\u3082\u5927\u304d\u306a\u30cb\u30e5\u30fc\u30b9\u304c\u591a\u304f\u3001Anthropic\u306eClaude 3.7 sonnet\u3001OpenAI\u306eGPT-4.5\u306a\u3069\u30d5\u30e9\u30b0\u30b7\u30c3\u30d7\u3068\u547c\u3079\u308b\u30e2\u30c7\u30eb\u306e\u767a\u8868\u304c\u76f8\u6b21\u3044\u3060\u3002<\/p>\n\n\n\n<p>Claude 3.7\u306fLLM&amp;LRM\u3068\u3044\u3046\u3088\u3046\u306a\u30e2\u30c7\u30eb\u3067\u30b3\u30fc\u30c9\u751f\u6210\u3067\u9ad8\u3044\u6027\u80fd\u3092\u767a\u63ee\u3057\u3066\u3044\u308b\u3002<a href=\"https:\/\/www.anthropic.com\/news\/claude-3-7-sonnet\">Claude 3.7 Sonnet and Claude Code \\ Anthropic<\/a><\/p>\n\n\n\n<p>GPT-4.5\u306f\u5de8\u5927\u30fb\u9ad8\u6027\u80fd\u306aLLM\u3068\u3044\u3046\u5370\u8c61<a href=\"https:\/\/openai.com\/ja-JP\/index\/introducing-gpt-4-5\/\">GPT-4.5 \u304c\u767b\u5834 | OpenAI<\/a>\u3002LRM\u3067\u306f\u89e3\u304d\u306b\u304f\u3044\u9818\u57df\u3067\u306f\u3068\u3066\u3082\u6709\u52b9\u305d\u3046\u3002\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u500b\u5225\u3067\u306f\u540c\u3058LLM\u306eDeepseek V3\u306b\u8ca0\u3051\u3066\u3044\u308b\u3082\u306e\u304c\u3042\u308a\uff08<a href=\"https:\/\/github.com\/deepseek-ai\/DeepSeek-V3?tab=readme-ov-file#5-chat-website--api-platform\">GitHub &#8211; deepseek-ai\/DeepSeek-V3<\/a>\u306eAIME 2024\u3084SWE Verified\uff09\u3001OpenAI\u4e00\u5f37\u6642\u4ee3\u306e\u7d42\u308f\u308a\u3092\u611f\u3058\u3055\u305b\u308b\u7d50\u679c\u306b\u306a\u3063\u3066\u3044\u308b\u3002<\/p>\n\n\n\n<p>\u3053\u306e\u3088\u3046\u306a\u4e2d\u3001Microsoft\u306ePhi-4\u30b7\u30ea\u30fc\u30ba\u3067\u3082\u65b0\u305f\u306a\u30e2\u30c7\u30eb\u304c\u516c\u958b\u3055\u308c\u3066\u3044\u308b<a href=\"https:\/\/techcommunity.microsoft.com\/blog\/educatordeveloperblog\/welcome-to-the-new-phi-4-models---microsoft-phi-4-mini--phi-4-multimodal\/4386037\">Welcome to the new Phi-4 models &#8211; Microsoft Phi-4-mini &amp; Phi-4-multimodal<\/a>\u3002\u5c0f\u578b\u30e2\u30c7\u30eb\u3067\u3082\u5341\u5206\u306a\u6027\u80fd\u304c\u51fa\u3066\u3044\u308b\u3088\u3046\u306b\u898b\u3048\u308b\u3002<\/p>\n\n\n\n<p><a href=\"https:\/\/www.atla-ai.com\/post\/selene-1\">Frontier AI needs frontier evaluators. Meet Selene.<\/a>\u306a\u3069\u3001\u5f37\u529b\u306aevaluator\u306a\u3069LLM\u3084LRM\u3092\u88dc\u5b8c\u3059\u308b\u52d5\u304d\u3082\u8208\u5473\u6df1\u3044\u3002<\/p>\n\n\n\n<p>LLM, LRM, SLM\u3084\u30c1\u30e5\u30fc\u30cb\u30f3\u30b0\u3001\u30cf\u30a4\u30d6\u30ea\u30c3\u30c9\u69cb\u6210\u306a\u3069\u69d8\u3005\u306a\u30a2\u30d7\u30ed\u30fc\u30c1\u304c\u3042\u308a\u3001\u30e2\u30c7\u30eb\u306e\u9078\u629e\u80a2\u3082\u5897\u3048\u3001\u4f55\u3092\u9078\u629e\u3057\u3066\u3044\u304f\u3079\u304d\u304b\u60a9\u3080\u6642\u4ee3\u306b\u306a\u3063\u305f\u306e\u304b\u306a\u3068\u3044\u3046\u5370\u8c61\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Atla Selene Mini: A General Purpose Evaluation Model&nbsp;<\/strong>[2.9]<br>\u6211\u3005\u306fSLMJ(Small-as-a-judge)\u306e\u6700\u5148\u7aef\u306e\u5c0f\u578b\u8a00\u8a9e\u3067\u3042\u308bAtla Selene Mini\u3092\u7d39\u4ecb\u3057\u305f\u3002 Selene Mini\u306f\u3001\u5168\u4f53\u7684\u306a\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u306b\u304a\u3044\u3066\u6700\u9ad8\u306eSLMJ\u3068GPT-4o-mini\u3088\u308a\u512a\u308c\u305f\u6c4e\u7528\u8a55\u4fa1\u5668\u3067\u3042\u308b\u3002 RewardBench\u3067\u6700\u3082\u9ad8\u30448B\u751f\u6210\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2501.17195v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2501.17195v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Mon, 27 Jan 2025 15:09:08 GMT)<\/li>\n\n\n\n<li>\u4e0a\u8ff0\u306eEvaluater\u30c1\u30fc\u30e0\u306e\u8ad6\u6587<\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Phi-4-Mini Technical Report: Compact yet Powerful MultimodalLanguage Models via Mixture-of-LoRAs<\/strong><br>Phi-4Mini\u3068Phi-4-Multimodal\u3001\u30b3\u30f3\u30d1\u30af\u30c8\u3067\u9ad8\u6a5f\u80fd\u306a\u8a00\u8a9e\u3068\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u30e2\u30c7\u30eb\u3092\u7d39\u4ecb\u3057\u307e\u3059\u3002Phi-4-Mini\u306f\u3001\u9ad8\u54c1\u8cea\u306a\u30a6\u30a7\u30d6\u304a\u3088\u3073\u5408\u6210\u30c7\u30fc\u30bf\u306b\u57fa\u3065\u3044\u3066\u8a13\u7df4\u3055\u308c\u305f3.8\u30d3\u30ea\u30aa\u30f3\u30d1\u30e9\u30e1\u30fc\u30bf\u8a00\u8a9e\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002Phi-4-Multimodal\u306f\u3001\u30c6\u30ad\u30b9\u30c8\u3001\u8996\u899a\u3001\u97f3\u58f0\/\u97f3\u58f0\u5165\u529b\u30e2\u30c0\u30ea\u30c6\u30a3\u3092\u5358\u4e00\u306e\u30e2\u30c7\u30eb\u306b\u7d71\u5408\u3059\u308b\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002<\/li>\n\n\n\n<li><a href=\"https:\/\/huggingface.co\/microsoft\/Phi-4-multimodal-instruct\/blob\/main\/phi_4_mm.tech_report.02252025.pdf\">phi_4_mm.tech_report.02252025.pdf \u00b7 microsoft\/Phi-4-multimodal-instruct at main<\/a><\/li>\n<\/ul>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>OpenAI GPT-4.5 System Card<\/strong><br>GPT-4.5\u306f\u4e8b\u524d\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3092\u3055\u3089\u306b\u30b9\u30b1\u30fc\u30eb\u3057\u3001\u5f37\u529b\u306aSTEM\u7126\u70b9\u63a8\u8ad6\u30e2\u30c7\u30eb\u3088\u308a\u3082\u6c4e\u7528\u7684\u306b\u8a2d\u8a08\u3055\u308c\u3066\u3044\u308b\u3002\u5e45\u5e83\u3044\u77e5\u8b58\u30d9\u30fc\u30b9\u3001\u30e6\u30fc\u30b6\u30fc\u306e\u610f\u56f3\u3068\u306e\u3088\u308a\u5f37\u56fa\u306a\u9023\u643a\u3001\u611f\u60c5\u7684\u77e5\u6027\u306e\u5411\u4e0a\u306f\u3001\u57f7\u7b46\u3001\u30d7\u30ed\u30b0\u30e9\u30df\u30f3\u30b0\u3001\u5b9f\u7528\u7684\u306a\u554f\u984c\u89e3\u6c7a\u3068\u3044\u3063\u305f\u30bf\u30b9\u30af\u306b\u9069\u3057\u3066\u3044\u308b\u3002<\/li>\n\n\n\n<li><a href=\"https:\/\/openai.com\/index\/gpt-4-5-system-card\/\">OpenAI GPT-4.5 System Card | OpenAI<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u5148\u9031\u3082\u5927\u304d\u306a\u30cb\u30e5\u30fc\u30b9\u304c\u591a\u304f\u3001Anthropic\u306eClaude 3.7 sonnet\u3001OpenAI\u306eGPT-4.5\u306a\u3069\u30d5\u30e9\u30b0\u30b7\u30c3\u30d7\u3068\u547c\u3079\u308b\u30e2\u30c7\u30eb\u306e\u767a\u8868\u304c\u76f8\u6b21\u3044\u3060\u3002 Claude 3.7\u306fLLM&amp;LRM\u3068\u3044\u3046\u3088\u3046\u306a\u30e2 &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=6319\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;Claude 3.7, GPT-4.5, Phi-4, Selene&#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":[223,232,365,368],"class_list":["post-6319","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-lrm","tag-slm","tag-small-model"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6319","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=6319"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6319\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6319"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6319"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6319"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}