{"id":28677,"date":"2025-02-12T20:45:57","date_gmt":"2025-02-12T12:45:57","guid":{"rendered":"https:\/\/www.1ai.net\/?p=28677"},"modified":"2025-02-12T20:45:57","modified_gmt":"2025-02-12T12:45:57","slug":"openai-%e6%9c%80%e6%96%b0%e8%ae%ba%e6%96%87%ef%bc%9ao3-%e5%9c%a8-ioi-2024-%e4%b8%a5%e6%a0%bc%e8%a7%84%e5%88%99%e4%b8%8b%e6%8b%bf%e5%88%b0-395-64-%e5%88%86%e8%be%be%e6%88%90%e9%87%91%e7%89%8c%e6%88%90","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/28677.html","title":{"rendered":"OpenAI's latest paper: o3 achieves gold medal with 395.64 points under the strict rules of IOI 2024"},"content":{"rendered":"<p>2 \u6708 12 \u65e5\u665a\u95f4\u6d88\u606f\uff0c\u5728\u4e2d\u56fd AI \u516c\u53f8\u7684\u5f71\u54cd\u4e0b\uff0c<a href=\"https:\/\/www.1ai.net\/en\/tag\/openai\" title=\"\u3010\u67e5\u770b\u542b\u6709[OpenAI]\u6807\u7b7e\u7684\u6587\u7ae0\u3011\" target=\"_blank\" rel=\"noopener\">OpenAI<\/a> \u516c\u5f00\u4e86 O \u7cfb\u5217\u5f3a\u5316\u5b66\u4e60\u7684\u79d8\u5bc6\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-28678\" title=\"930efbb9j00srkm3q008ed000fa00mvp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2025\/02\/930efbb9j00srkm3q008ed000fa00mvp.jpg\" alt=\"930efbb9j00srkm3q008ed000fa00mvp\" width=\"550\" height=\"823\" \/><\/p>\n<p>\u4eca\u5929\uff082 \u6708 12 \u65e5\uff09\uff0cOpenAI \u53d1\u5e03\u4e86\u5173\u4e8e\u63a8\u7406\u6a21\u578b\u5728\u7ade\u6280\u7f16\u7a0b\u4e2d\u5e94\u7528\u7684\u7814\u7a76<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e8%ae%ba%e6%96%87\" title=\"\u3010\u67e5\u770b\u542b\u6709[\u8bba\u6587]\u6807\u7b7e\u7684\u6587\u7ae0\u3011\" target=\"_blank\" rel=\"noopener\">\u8bba\u6587<\/a>\u62a5\u544a\u300aCompetitive Programming with Large Reasoning Models\u300b\uff0c\u6587\u4e2d\u653e\u51fa\u4e86 OpenAI \u4e09\u4e2a\u63a8\u7406\u6a21\u578b\uff1ao1\u3001o1-ioi\u3001o3 \u5728 IOI\uff08\u56fd\u9645\u4fe1\u606f\u5b66\u5965\u6797\u5339\u514b\u7ade\u8d5b\uff09\u548c CodeForces\uff08\u5168\u7403\u77e5\u540d\u5728\u7ebf\u7f16\u7a0b\u7ade\u8d5b\uff09\u4e2d\u7684\u6210\u7ee9\u3002<\/p>\n<p>\u8bba\u6587\u663e\u793a\uff0c\u5728 IOI 2024 \u4e2d\uff0co3 \u5728\u4e25\u683c\u89c4\u5219\u4e0b\u62ff\u5230 395.64 \u5206\uff0c\u8fbe\u6210\u91d1\u724c\u6210\u5c31\uff0c\u5e76\u4e14\u5728 CodeForces \u4e0a\u7684\u8868\u73b0\u4e0e\u4eba\u7c7b\u7cbe\u82f1\u9009\u624b\u76f8\u5f53\u3002<\/p>\n<p>\u8bba\u6587\u4e2d\u8fd8\u63d0\u5230\uff0c\u4e2d\u56fd\u7684 DeepSeek-R1 \u548c Kimi\u00a0k1.5 \u901a\u8fc7\u72ec\u7acb\u7814\u7a76\u663e\u793a\uff0c\u5229\u7528\u601d\u7ef4\u94fe\u5b66\u4e60\uff08COT\uff09\u65b9\u6cd5\uff0c\u53ef\u663e\u8457\u63d0\u5347\u6a21\u578b\u5728\u6570\u5b66\u89e3\u9898\u4e0e\u7f16\u7a0b\u6311\u6218\u4e2d\u7684\u7efc\u5408\u8868\u73b0\u3002R1\u3001k1.5 \u662f DeepSeek \u548c Kimi \u5728 1 \u6708 20 \u65e5\u540c\u65f6\u53d1\u5e03\u7684\u65b0\u578b\u63a8\u7406\u6a21\u578b\u3002<\/p>\n<p>\u8be5\u8bba\u6587\u901a\u8fc7\u5f3a\u5316\u5b66\u4e60\uff08RL\uff09\u8bad\u7ec3\u7684\u5927\u578b\u8bed\u8a00\u6a21\u578b\u5728\u590d\u6742\u7f16\u7801\u548c\u63a8\u7406\u4efb\u52a1\u4e0a\u7684\u6027\u80fd\u63d0\u5347\uff0c\u6bd4\u8f83\u4e86\u901a\u7528\u63a8\u7406\u6a21\u578b\u4e0e\u9488\u5bf9\u7279\u5b9a\u9886\u57df\u4f18\u5316\u7684\u7cfb\u7edf\u5728\u7ade\u6280\u7f16\u7a0b\u4e2d\u7684\u8868\u73b0\u3002\u7814\u7a76\u7ed3\u679c\u8868\u660e\uff0c\u589e\u52a0\u5f3a\u5316\u5b66\u4e60\u8bad\u7ec3\u8ba1\u7b97\u548c\u6d4b\u8bd5\u65f6\u8ba1\u7b97\u53ef\u663e\u8457\u63d0\u5347\u6a21\u578b\u6027\u80fd\uff0c\u4f7f\u5176\u63a5\u8fd1\u4e16\u754c\u9876\u5c16\u4eba\u7c7b\u9009\u624b\uff0c\u8fd9\u4e9b\u6a21\u578b\u5c06\u5728\u79d1\u5b66\u3001\u7f16\u7801\u3001\u6570\u5b66\u7b49\u9886\u57df\u7684 AI \u5e94\u7528\u4e2d\u89e3\u9501\u65b0\u7684\u5e94\u7528\u4f53\u9a8c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>February 12 evening news, under the influence of Chinese AI companies, OpenAI has disclosed the secrets of O-series reinforcement learning. Today (February 12), OpenAI released a research paper on the application of inference models in competitive programming, \"Competitive Programming with Large Reasoning Models\", which released the results of OpenAI's three inference models: o1, o1-ioi, and o3 in the IOI (International Olympiad of Informatics) and CodeForces (the world's leading online programming competition). Olympiad (International Olympiad in Informatics) and CodeForces (a globally recognized online programming competition). According to the paper, in IOI 2024, o3 got the highest score under strict rules.<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[146],"tags":[190,911],"collection":[],"class_list":["post-28677","post","type-post","status-publish","format-standard","hentry","category-news","tag-openai","tag-911"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/28677","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=28677"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/28677\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=28677"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=28677"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=28677"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=28677"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}