{"id":43442,"date":"2025-09-18T11:33:00","date_gmt":"2025-09-18T03:33:00","guid":{"rendered":"https:\/\/www.1ai.net\/?p=43442"},"modified":"2025-09-18T11:33:00","modified_gmt":"2025-09-18T03:33:00","slug":"deepseek-r1-%e8%ae%ba%e6%96%87%e7%99%bb%e4%b8%8a%e3%80%8a%e8%87%aa%e7%84%b6%e3%80%8b%e5%b0%81%e9%9d%a2%ef%bc%8c%e9%80%9a%e8%ae%af%e4%bd%9c%e8%80%85%e4%b8%ba%e6%a2%81%e6%96%87%e9%94%8b","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/43442.html","title":{"rendered":"DeepSeek-R1 Thesis is on the cover of Nature, which is written by Liang Wenbing"},"content":{"rendered":"<p>Message of September 18, by <a href=\"https:\/\/www.1ai.net\/en\/tag\/deepseek\" title=\"[View articles tagged with [DeepSeek]]\" target=\"_blank\" >DeepSeek<\/a> Teamwork, teamwork<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e6%a2%81%e6%96%87%e9%94%8b\" title=\"[See articles with labels]\" target=\"_blank\" >Leung Man Fung<\/a>DeepSeek-R1 Logic Model Study as Communication Author<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e8%ae%ba%e6%96%87\" title=\"[See articles with [paper] labels]\" target=\"_blank\" >paper<\/a>It's on the international authoritative journal Nature<a href=\"https:\/\/www.1ai.net\/en\/tag\/nature\" title=\"[read articles with [Nature] label]\" target=\"_blank\" >Nature<\/a>No. 645 on the cover. Compared to the first edition of DeepSeek-R1 published in January this year, this paper reveals more details of model training\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-43443\" title=\"ab218981j00t2rltx001ed000m200bfp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2025\/09\/ab218981j00t2rltx001ed000m200bfp.jpg\" alt=\"ab218981j00t2rltx001ed000m200bfp\" width=\"794\" height=\"411\" \/><\/p>\n<p>It is reported that,<strong>DeepSeek-R1 is also the first globally peer-reviewed mainstream language model<\/strong>I don't know. Nature assesses that almost all major mainstream models have not yet been independently peer-reviewed, and that gap \u201chas finally been broken by DeepSeek\u201d\u3002<\/p>\n<p>The summary of the paper shows that common reasoning has been a long and challenging challenge in the AI area. In recent years, technological breakthroughs represented by large language models (LLMs) and COTs (CoTs) have achieved remarkable success in basic reasoning tasks. I don't know<strong>This success relies heavily on manual demonstration data and the model ' s ability to deal with more complex issues remains inadequate<\/strong>.<\/p>\n<p>STUDIES HAVE SHOWN THAT THE REASONING OF LARGE LANGUAGE MODELS CAN BE STIMULATED BY PURE ENHANCED LEARNING (RL), WITHOUT RELYING ON ARTIFICIALLY MARKED REASONING TRACKS. THE PROPOSED ENHANCED LEARNING FRAMEWORK PROMOTES THE AUTONOMY OF ADVANCED REASONING MODELS<strong>For example, self-reflection, validation and dynamic strategy adjustments<\/strong>.<\/p>\n<p>THUS, TRAINED MODELS PRESENT MORE SUPERIOR PERFORMANCES IN VERIFIABLE TASKS SUCH AS MATHEMATICS, PROGRAMMING COMPETITIONS AND STEM (SCIENCE, TECHNOLOGY, ENGINEERING, MATHEMATICS) THAN SIMILAR MODELS TRAINED THROUGH TRADITIONAL SUPERVISORY LEARNING (BASED ON MANUAL DEMONSTRATION DATA). MOREOVER, THE AUTONOMOUS MODE OF REASONING PRESENTED BY THESE LARGE MODELS CAN BE USED SYSTEMATICALLY TO GUIDE AND ENHANCE THE REASONING CAPACITY OF SMALL MODELS\u3002<\/p>\n<p><strong>1AI WITH PAPER LINKS:<\/strong><\/p>\n<p>https:\/\/www.nature.com\/articles\/s41586-025-09422-z<\/p>","protected":false},"excerpt":{"rendered":"<p>The 18 September message, which was jointly completed by the DeepSeek team and published as a communication author by Liang Wenfeng, contained a DeepSeek-R1 research paper on the reasoning model, on the cover of issue 645 of the international authoritative journal Nature. Compared to the first edition of DeepSeek-R1 published in January this year, this paper reveals more details of model training. DeepSeek-R1 is also known to be the first globally peer-reviewed dominant language model. Nature assesses that almost all major mainstream models have not yet been independently peer-reviewed, and that gap \u201chas finally been broken by DeepSeek\u201d. The dissertation summary shows that the generic reasoning has always been AI-based<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[3606,1824,5762,911],"collection":[],"class_list":["post-43442","post","type-post","status-publish","format-standard","hentry","category-news","tag-deepseek","tag-nature","tag-5762","tag-911"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/43442","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=43442"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/43442\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=43442"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=43442"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=43442"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=43442"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}