{"id":6245,"date":"2025-10-09T11:51:13","date_gmt":"2025-10-09T03:51:13","guid":{"rendered":"https:\/\/nullthought.net\/?p=6245"},"modified":"2025-10-09T11:54:27","modified_gmt":"2025-10-09T03:54:27","slug":"rd-agent%ef%bc%9a%e7%b3%bb%e7%bb%9f%e5%8c%96%e3%80%81%e5%8f%af%e6%89%a9%e5%b1%95%e3%80%81%e8%a7%a3%e8%80%a6%e5%90%88%e7%9a%84mle%ef%bc%88machine-learning-engineering%ef%bc%8c%e6%9c%ba%e5%99%a8","status":"publish","type":"post","link":"https:\/\/nullthought.net\/?p=6245","title":{"rendered":"R&amp;D-Agent\uff1a\u7cfb\u7edf\u5316\u3001\u53ef\u6269\u5c55\u3001\u89e3\u8026\u5408\u7684MLE\uff08Machine Learning Engineering\uff0c\u673a\u5668\u5b66\u4e60\u5de5\u7a0b\uff09\u667a\u80fd\u4f53\u67b6\u6784"},"content":{"rendered":"\n<p>\u968f\u7740\u4eba\u5de5\u667a\u80fd\uff08AI\uff09\u548c\u673a\u5668\u5b66\u4e60\uff08ML\uff09\u7684\u8fc5\u731b\u53d1\u5c55\uff0c\u6570\u636e\u79d1\u5b66\u9886\u57df\u53d6\u5f97\u4e86\u663e\u8457\u8fdb\u5c55\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u673a\u5668\u7ffb\u8bd1\u3001\u63a8\u8350\u7cfb\u7edf\u3001\u793e\u4f1a\u6a21\u62df\u548c\u533b\u5b66\u8bca\u65ad\u7b49\u9886\u57df\u3002\u7136\u800c\uff0c\u4f34\u968f\u6570\u636e\u65e5\u76ca\u5f02\u6784\u4e0e\u9ad8\u7ef4\u5316\uff0c\u6570\u636e\u79d1\u5b66\u4efb\u52a1\u7684\u590d\u6742\u6027\u4e5f\u6301\u7eed\u4e0a\u5347\uff0c\u9700\u8981\u66f4\u9ad8\u6c34\u5e73\u7684\u4e13\u4e1a\u77e5\u8bc6\u548c\u5de5\u7a0b\u80fd\u529b\u3002\u5c3d\u7ba1Kaggle\u7b49\u4f17\u5305\u5e73\u53f0\u90e8\u5206\u7f13\u89e3\u4e86\u8fd9\u4e00\u6311\u6218\uff0c\u4f46\u9ad8\u6c34\u5e73\u673a\u5668\u5b66\u4e60\u5de5\u7a0b\uff08MLE\uff09\u4efb\u52a1\u4ecd\u7136\u6781\u5ea6\u4f9d\u8d56\u624b\u5de5\u8fed\u4ee3\u4e0e\u7ecf\u9a8c\u8c03\u4f18\uff0c\u6548\u7387\u8f83\u4f4e\u3002<\/p>\n\n\n\n<p>\u5927\u8bed\u8a00\u6a21\u578b\uff08LLMs\uff09\u5177\u5907\u5f3a\u5927\u7684\u4ee3\u7801\u751f\u6210\u4e0e\u63a8\u7406\u80fd\u529b\uff0c\u88ab\u8ba4\u4e3a\u80fd\u6709\u6548\u63a2\u7d22\u5927\u89c4\u6a21\u8bbe\u8ba1\u7a7a\u95f4\uff0c\u81ea\u52a8\u5316MLE\u6d41\u7a0b\u3002\u4f46\u5b9e\u9645\u60c5\u51b5\u662f\uff0c\u5373\u4f7f\u6700\u5148\u8fdb\u7684LLMs\u5728MLE-bench\u8fd9\u7c7b\u73b0\u5b9e\u7ade\u8d5b\u57fa\u51c6\u4e2d\uff0c\u4e5f\u4ec5\u80fd\u5b9e\u73b0\u4eba\u7c7b\u4e13\u5bb6\u8868\u73b0\u7684\u4e00\u5c0f\u90e8\u5206\u3002\u4e3a\u89e3\u51b3\u8fd9\u4e00\u5dee\u8ddd\uff0c\u8bba\u6587<strong><a href=\"https:\/\/arxiv.org\/abs\/2505.14738\" target=\"_blank\" rel=\"noreferrer noopener\">R&amp;D-AGENT: AN LLM-AGENT FRAMEWORK TOWARDS AUTONOMOUS DATA SCIENCE<\/a><\/strong>\u63d0\u51fa\u4e86R&amp;D-Agent\u6846\u67b6\uff0c\u4e00\u4e2a\u7cfb\u7edf\u5316\u3001\u53ef\u6269\u5c55\u3001\u89e3\u8026\u5408\u7684MLE\u667a\u80fd\u4f53\u67b6\u6784\uff0c\u8be5\u67b6\u6784\u53ef\u5c06AI\u8f85\u52a9\u7684\u673a\u5668\u5b66\u4e60\u5de5\u7a0b\u63a8\u8fdb\u81f3\u66f4\u9ad8\u7684\u81ea\u52a8\u5316\u5c42\u6b21\u3002<\/p>\n\n\n\n<p>\u8bba\u6587\u4f5c\u8005\u4e3aXu Yang, Xiao Yang, Shikai Fang, Yifei Zhang, Jian Wang, Bowen Xian, Qizheng Li, Jingyuan Li, Minrui Xu, Yuante Li, Haoran Pan, Yuge Zhang, Weiqing Liu, Yelong Shen, Weizhu Chen, Jiang Bian\uff0c\u6765\u81ea\u5fae\u8f6f\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"747\" src=\"https:\/\/nullthought.net\/wp-content\/uploads\/2025\/10\/image-1-1024x747.png\" alt=\"\" class=\"wp-image-6246\" srcset=\"https:\/\/nullthought.net\/wp-content\/uploads\/2025\/10\/image-1-1024x747.png 1024w, https:\/\/nullthought.net\/wp-content\/uploads\/2025\/10\/image-1-300x219.png 300w, https:\/\/nullthought.net\/wp-content\/uploads\/2025\/10\/image-1-768x560.png 768w, https:\/\/nullthought.net\/wp-content\/uploads\/2025\/10\/image-1.png 1061w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\"><strong><a href=\"https:\/\/arxiv.org\/abs\/2505.14738\" target=\"_blank\" rel=\"noreferrer noopener\">R&amp;D-AGENT: AN LLM-AGENT FRAMEWORK TOWARDS AUTONOMOUS DATA SCIENCE<\/a><\/strong><\/figcaption><\/figure>\n\n\n\n<p>\u4e00\u3001R&amp;D-Agent \u6846\u67b6\u6982\u8ff0<\/p>\n\n\n\n<p>R&amp;D-Agent \u501f\u9274\u4eba\u7c7b\u6570\u636e\u79d1\u5b66\u5bb6\u7684\u5de5\u4f5c\u6d41\u7a0b\uff0c\u5c06\u6574\u4e2a\u673a\u5668\u5b66\u4e60\u5de5\u7a0b\u8fc7\u7a0b\u5212\u5206\u4e3a\u4e24\u4e2a\u6838\u5fc3\u9636\u6bb5\u548c\u516d\u4e2a\u6a21\u5757\u7ec4\u4ef6\uff1a<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>\u7814\u7a76\u9636\u6bb5\uff08Research Phase\uff09\n<ul class=\"wp-block-list\">\n<li>\u2776 \u89c4\u5212\uff08Planning\uff09\uff1a\u52a8\u6001\u8c03\u6574\u63a2\u7d22\u7b56\u7565\u4e0e\u8d44\u6e90\u5206\u914d\u3002<\/li>\n\n\n\n<li>\u2777 \u63a2\u7d22\u8def\u5f84\u7ed3\u6784\uff08Exploration Path Structuring\uff09\uff1a\u51b3\u5b9a\u63a2\u7d22\u7b56\u7565\u662f\u94fe\u5f0f\u8fd8\u662f\u6811\u72b6\u7ed3\u6784\u3002<\/li>\n\n\n\n<li>\u2778 \u63a8\u7406\u6d41\u7a0b\uff08Reasoning Pipeline\uff09\uff1a\u7cfb\u7edf\u5730\u751f\u6210\u4e0e\u7b5b\u9009\u5019\u9009\u7814\u7a76\u601d\u8def\u3002<\/li>\n\n\n\n<li>\u2779 \u8bb0\u5fc6\u4e0a\u4e0b\u6587\uff08Memory Context\uff09\uff1a\u7ec4\u7ec7\u5386\u53f2\u77e5\u8bc6\u4ee5\u4f9b\u91cd\u590d\u5229\u7528\u4e0e\u77e5\u8bc6\u8fc1\u79fb\u3002<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li>\u5f00\u53d1\u9636\u6bb5\uff08Development Phase\uff09\n<ul class=\"wp-block-list\">\n<li>\u277a \u7f16\u7801\u6d41\u7a0b\uff08Coding Workflow\uff09\uff1a\u5c06\u601d\u8def\u8f6c\u5316\u4e3a\u53ef\u8fd0\u884c\u3001\u53ef\u8c03\u8bd5\u7684\u4ee3\u7801\u3002<\/li>\n\n\n\n<li>\u277b \u8bc4\u4f30\u7b56\u7565\uff08Evaluation Strategy\uff09\uff1a\u63d0\u4f9b\u9c81\u68d2\u3001\u7a33\u5b9a\u7684\u4e00\u81f4\u6027\u8bc4\u4f30\u624b\u6bb5\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\n\n<p>\u6574\u4e2a\u6d41\u7a0b\u5f62\u6210\u4e00\u4e2aR&amp;D\u5faa\u73af\uff0c\u7814\u7a76\u4ee3\u7406\u751f\u6210\u60f3\u6cd5\uff0c\u5f00\u53d1\u4ee3\u7406\u5b9e\u73b0\u3001\u9a8c\u8bc1\u4e0e\u53cd\u9988\uff0c\u4ece\u800c\u9010\u6b65\u903c\u8fd1\u6700\u4f18\u89e3\u3002<\/p>\n\n\n\n<p>\u4e8c\u3001\u5173\u952e\u6a21\u5757\u8bbe\u8ba1\uff08\u4eba\u7c7b\u4e13\u5bb6\u542f\u53d1\uff09<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>\u89c4\u5212\u6a21\u5757\uff08MD-Planning\uff09\uff1a\u501f\u9274\u4eba\u7c7b\u7684\u79d1\u5b66\u63a2\u7d22\u8def\u5f84\uff0c\u524d\u671f\u805a\u7126\u8f7b\u91cf\u63a2\u7d22\u4e0e\u65b0\u9896\u65b9\u6848\uff0c\u540e\u671f\u9010\u6e10\u5f15\u5165\u590d\u6742\u65b9\u6cd5\uff08\u5982\u96c6\u6210\u3001\u4ea4\u53c9\u9a8c\u8bc1\uff09\u4ee5\u7cbe\u7ec6\u4f18\u5316\u3002<\/li>\n\n\n\n<li>\u63a2\u7d22\u8def\u5f84\u7ed3\u6784\uff08MD-Exploration Path\uff09\uff1a\u91c7\u7528DAG\u7ed3\u6784\u591a\u8def\u5f84\u5e76\u884c\u63a2\u7d22\uff0c\u524d\u671f\u5e7f\u5ea6\u6700\u5927\u5316\uff0c\u540e\u671f\u5c40\u90e8\u8d2a\u5fc3\u4f18\u5316\uff0c\u540c\u65f6\u5408\u5e76\u4f18\u8d28\u89e3\u3002<\/li>\n\n\n\n<li>\u591a\u6b65\u63a8\u7406\u6d41\u7a0b\uff08MD-Scientific Reasoning Pipeline\uff09\uff1a\u6a21\u4eff\u79d1\u5b66\u63a8\u7406\uff0c\u5148\u5206\u6790\u95ee\u9898\uff0c\u518d\u63d0\u51fa\u5047\u8bbe\u3001\u8bc4\u4f30\u4ee3\u4ef7\u4e0e\u6536\u76ca\uff0c\u5e76\u5f15\u5165LLM\u8bc4\u5206\u7684\u865a\u62df\u8bc4\u4f30\u6765\u7b5b\u9009\u6700\u4f18idea\u3002<\/li>\n\n\n\n<li>\u534f\u4f5c\u5f0f\u8bb0\u5fc6\u4e0a\u4e0b\u6587\uff08MD-Memory Context\uff09\uff1a\u8de8\u5206\u652f\u5171\u4eab\u9ad8\u8d28\u91cf\u601d\u8def\uff0c\u901a\u8fc7\u6982\u7387\u5185\u6838\u4e0eLLM\u51b3\u7b56\u8fdb\u884c\u4fe1\u606f\u4ea4\u4e92\uff0c\u65e2\u4fc3\u8fdb\u6536\u655b\u53c8\u4fdd\u7559\u591a\u6837\u6027\u3002<\/li>\n\n\n\n<li>\u9ad8\u6548\u7f16\u7801\u6d41\u7a0b\uff08MD-Coding Workflow\uff09\uff1a\u5148\u5728\u91c7\u6837\u5b50\u96c6\u4e0a\u5feb\u901f\u539f\u578b\u8fed\u4ee3\uff0c\u9a8c\u8bc1\u901a\u8fc7\u540e\u518d\u7528\u5168\u6570\u636e\u8bad\u7ec3\uff0c\u6781\u5927\u7f29\u77ed\u5f00\u53d1\u5468\u671f\uff0c\u63d0\u5347\u8bd5\u9a8c\u6548\u7387\u3002<\/li>\n\n\n\n<li>\u805a\u5408\u8bc4\u4f30\u7b56\u7565\uff08MD-Evaluation Strategy\uff09\uff1a\u7edf\u4e00\u9a8c\u8bc1\u96c6\u5212\u5206\u3001\u805a\u5408\u591a\u4e2a\u5b50\u8def\u5f84\u7ed3\u679c\uff0c\u4fdd\u8bc1\u8bc4\u4f30\u4e00\u81f4\u6027\u4e0e\u516c\u5e73\u6027\uff0c\u51cf\u5c11\u5076\u7136\u8bef\u5dee\u4e0e\u8fc7\u62df\u5408\u98ce\u9669\u3002<\/li>\n<\/ol>\n\n\n\n<p>\u4e09\u3001\u5b9e\u9a8c\u8bbe\u8ba1\u4e0e\u7ed3\u679c\u5206\u6790<\/p>\n\n\n\n<p>\u5b9e\u9a8c\u4f7f\u7528MLE-Bench\u57fa\u51c6\uff08\u57fa\u4e8e\u771f\u5b9eKaggle\u7ade\u8d5b\uff09\u8fdb\u884c\u8bc4\u4f30\uff0c\u5bf9\u6bd4\u4e3b\u6d41\u7cfb\u7edf\u5982ML-Master\u3001AIDE\u7b49\uff0c\u5e76\u4f7f\u7528\u4e24\u79cd\u914d\u7f6e\uff1aGPT-5\u72ec\u7acb\u90e8\u7f72\u4e0eo3\uff08Research\uff09+ GPT-4.1\uff08Development\uff09\u6df7\u5408\u67b6\u6784\u3002<\/p>\n\n\n\n<p>\u4e3b\u8981\u53d1\u73b0\u5982\u4e0b\uff1a<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>R&amp;D-Agent\uff08GPT-5\uff09\u5728\u201c\u4efb\u610f\u5956\u724c\u7387\uff08Any Medal Rate\uff09\u201d\u4e0a\u8fbe35.1%\uff0c\u663e\u8457\u4f18\u4e8eML-Master\uff0829.3%\uff09\uff0c\u5b9e\u73b0SOTA\u3002<\/li>\n\n\n\n<li>\u6df7\u5408\u6a21\u578b\u914d\u7f6e\u5728\u8d44\u6e90\u53d7\u9650\u60c5\u51b5\u4e0b\u4e5f\u8fbe\u5230\u4e8629.7%\uff0c\u8bf4\u660e\u6846\u67b6\u8bbe\u8ba1\u4f18\u4e8e\u7eaf\u6a21\u578b\u66ff\u6362\u3002<\/li>\n\n\n\n<li>\u5728\u4efb\u52a1\u590d\u6742\u5ea6\u4e0d\u540c\u7684\u5206\u7ec4\u4e2d\uff0cR&amp;D-Agent\u5728\u6240\u6709\u590d\u6742\u5ea6\uff08Low==Lite, Medium, High\uff09\u4e0a\u5747\u8868\u73b0\u4f18\u5f02\u3002<\/li>\n<\/ol>\n\n\n\n<p>\u56db\u3001\u6d88\u878d\u5b9e\u9a8c<\/p>\n\n\n\n<p>\u4e3a\u9a8c\u8bc1\u5404\u7ec4\u4ef6\u7684\u8d21\u732e\uff0c\u4f5c\u8005\u5bf9\u516d\u4e2a\u6838\u5fc3\u6a21\u5757\u9010\u4e00\u8fdb\u884c\u6d88\u878d\u5b9e\u9a8c\u3002\u7814\u7a76\u9636\u6bb5\u56db\u4e2a\u6a21\u5757\u4e2d\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u53bb\u9664\u89c4\u5212\u6a21\u5757\uff1a\u4efb\u610f\u5956\u724c\u7387\u4e0b\u964d24%\uff0c\u56e0\u7f3a\u4e4f\u65f6\u95f4\u52a8\u6001\u9002\u5e94\u3002<\/li>\n\n\n\n<li>\u53bb\u9664\u63a2\u7d22\u8def\u5f84\u7ed3\u6784\uff1a\u4e0b\u964d28%\uff0cR&amp;D\u5faa\u73af\u6b21\u6570\u51cf\u534a\uff0c\u8868\u660e\u6811\u72b6\u7ed3\u6784\u662f\u5173\u952e\u3002<\/li>\n\n\n\n<li>\u53bb\u9664\u591a\u6b65\u63a8\u7406\uff1a\u4e0b\u964d24%\uff0c\u6539\u8fdb\u7387\u9aa4\u964d\u81f323%\uff0c\u663e\u793a\u7b80\u5355\u5355\u6b65\u63a8\u7406\u65e0\u6cd5\u6709\u6548\u63d0\u5347\u8d28\u91cf\u3002<\/li>\n\n\n\n<li>\u53bb\u9664\u534f\u4f5c\u8bb0\u5fc6\uff1a\u4e0b\u964d\u8f83\u5c0f\uff089%\uff09\uff0c\u4f46\u4ecd\u5f71\u54cd\u8de8\u5206\u652f\u77e5\u8bc6\u5171\u4eab\u4e0e\u5feb\u901f\u6536\u655b\u3002<\/li>\n<\/ul>\n\n\n\n<p>\u5f00\u53d1\u9636\u6bb5\u4e2d\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u53bb\u9664\u7f16\u7801\u6d41\u7a0b\uff1a\u5956\u724c\u7387\u4e0b\u964d\u81f324%\uff0c\u91c7\u6837\u8c03\u8bd5\u6cd5\u4f18\u4e8e\u5168\u6570\u636e\u8c03\u8bd5\u662f\u5173\u952e\u7a81\u7834\u3002<\/li>\n\n\n\n<li>\u53bb\u9664\u8bc4\u4f30\u7b56\u7565\uff1a\u521d\u671f\u65e0\u660e\u663e\u5f71\u54cd\uff0c\u4f46\u968f\u7740\u590d\u6742\u89e3\u7684\u51fa\u73b0\uff0c\u5956\u724c\u7387\u4e0b\u964d\u81f330.7%\uff0c\u8868\u660e\u7cfb\u7edf\u6027\u8bc4\u4f30\u5728\u540e\u671f\u81f3\u5173\u91cd\u8981\u3002<\/li>\n<\/ul>\n\n\n\n<p>\u4e94\u3001\u989d\u5916\u5206\u6790\u4e0e\u6269\u5c55\u7ed3\u679c<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>\u4e0d\u540cLLM\u914d\u7f6e\u5bf9\u6bd4\uff1a\u6df7\u5408\u67b6\u6784\uff08o3+GPT-4.1\uff09\u4f18\u4e8e\u5355\u6a21\u578b\u914d\u7f6e\uff0c\u9a8c\u8bc1\u201c\u7814\u7a76-\u5f00\u53d1\u201d\u5206\u5de5\u4e0b\u7684\u6a21\u578b\u4e92\u8865\u6027\u3002<\/li>\n\n\n\n<li>RAG\uff08\u68c0\u7d22\u589e\u5f3a\uff09\u5f15\u5165\u53cd\u800c\u5728\u7b80\u5355\u4efb\u52a1\u4e0a\u964d\u4f4e\u8868\u73b0\uff0c\u8868\u660e\u201c\u662f\u5426\u4f7f\u7528RAG\u201d\u5e94\u6309\u4efb\u52a1\u96be\u5ea6\u9009\u62e9\uff0cRAG\u975e\u4e07\u80fd\u589e\u76ca\u3002<\/li>\n\n\n\n<li>\u4e0e\u95ed\u6e90\u7cfb\u7edf\u5bf9\u6bd4\uff1aR&amp;D-Agent\u5728\u4f4e\u8d44\u6e90\u4e0b\uff0812h\u30011\u00d7V100 GPU\uff09\u8fbe\u5230\u4e0eInternAgent\u7b49\u95ed\u6e90\u7cfb\u7edf\u76f8\u8fd1\u6216\u66f4\u4f18\u7684\u6c34\u5e73\uff0c\u51f8\u663e\u5176\u5b9e\u7528\u6027\u4e0e\u53ef\u590d\u73b0\u6027\u3002<\/li>\n\n\n\n<li>\u8ba1\u7b97\u6548\u7387\uff1a\u5e73\u5747\u6bcf\u573a\u6bd4\u8d5b\u82b1\u8d39\u7ea621\u7f8e\u5143\uff0c\u8fdc\u4f4e\u4e8e\u4f20\u7edfAutoML\u65b9\u6848\uff0c\u5177\u5907\u6781\u9ad8\u6027\u4ef7\u6bd4\u3002<\/li>\n<\/ol>\n\n\n\n<p>\u516d\u3001\u7ed3\u8bba<\/p>\n\n\n\n<p>R&amp;D-Agent\u5c06\u673a\u5668\u5b66\u4e60\u5de5\u7a0b\u4ee3\u7406\u7684\u8bbe\u8ba1\u4ece\u624b\u5de5\u8c03\u53c2\u8f6c\u53d8\u4e3a\u7cfb\u7edf\u6027\u3001\u53ef\u590d\u73b0\u3001\u53ef\u7ec4\u5408\u7684\u63a2\u7d22\u8fc7\u7a0b\u3002\u5176\u6838\u5fc3\u521b\u65b0\u5728\u4e8e\u660e\u786e\u533a\u5206\u7814\u7a76\u4e0e\u5f00\u53d1\u4e24\u4e2a\u9636\u6bb5\uff0c\u5e76\u5728\u516d\u4e2a\u7ec4\u4ef6\u4e0a\u7cfb\u7edf\u5316\u67b6\u6784\u8bbe\u8ba1\uff0c\u7ed3\u5408\u4eba\u7c7b\u4e13\u5bb6\u5de5\u4f5c\u6d41\uff0c\u5f62\u6210\u9ad8\u6548\u5e76\u884c\u3001\u53ef\u590d\u7528\u3001\u6613\u8c03\u8bd5\u7684\u901a\u7528\u6846\u67b6\u3002<\/p>\n\n\n\n<p>\u8bba\u6587\u901a\u8fc7\u8be6\u5c3d\u7684\u5b9e\u9a8c\u4e0e\u6d88\u878d\u5206\u6790\uff0c\u9a8c\u8bc1\u4e86\u6bcf\u4e2a\u6a21\u5757\u5728\u63d0\u5347\u6548\u7387\u3001\u51c6\u786e\u6027\u4e0e\u6cdb\u5316\u80fd\u529b\u4e0a\u7684\u4f5c\u7528\u3002R&amp;D-Agent\u4e0d\u4ec5\u5237\u65b0\u4e86MLE-Bench\u57fa\u51c6\u7684\u6700\u4f73\u6210\u7ee9\uff0c\u4e5f\u4e3a\u540e\u7eed\u7814\u7a76\u8005\u63d0\u4f9b\u4e86\u6807\u51c6\u5316\u3001\u6a21\u5757\u5316\u7684\u5f00\u53d1\u5e73\u53f0\uff0c\u6781\u5927\u964d\u4f4e\u4e86\u8fdb\u5165\u95e8\u69db\uff0c\u6709\u671b\u63a8\u52a8AI\u81ea\u4e3b\u79d1\u5b66\u7814\u7a76\u7684\u53d1\u5c55\u3002<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p>R&amp;D-Agent on GitHub: <a href=\"https:\/\/github.com\/microsoft\/RD-Agent\" target=\"_blank\" rel=\"noreferrer noopener\">https:\/\/github.com\/microsoft\/RD-Agent<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u968f\u7740\u4eba\u5de5\u667a\u80fd\uff08AI\uff09\u548c\u673a\u5668\u5b66\u4e60\uff08ML\uff09\u7684\u8fc5\u731b\u53d1\u5c55\uff0c\u6570\u636e\u79d1\u5b66\u9886\u57df\u53d6\u5f97\u4e86\u663e\u8457\u8fdb\u5c55\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u673a\u5668\u7ffb\u8bd1\u3001\u63a8\u8350\u7cfb\u7edf\u3001\u793e [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[8,36],"tags":[39,94,86],"class_list":["post-6245","post","type-post","status-publish","format-standard","hentry","category-tech","category-36","tag-ai","tag-agentic-ai","tag-llm"],"rttpg_featured_image_url":null,"rttpg_author":{"display_name":"NullThought","author_link":"https:\/\/nullthought.net\/?author=1"},"rttpg_comment":0,"rttpg_category":"<a href=\"https:\/\/nullthought.net\/?cat=8\" rel=\"category\">Tech<\/a> <a href=\"https:\/\/nullthought.net\/?cat=36\" rel=\"category\">\u79d1\u5b66<\/a>","rttpg_excerpt":"\u968f\u7740\u4eba\u5de5\u667a\u80fd\uff08AI\uff09\u548c\u673a\u5668\u5b66\u4e60\uff08ML\uff09\u7684\u8fc5\u731b\u53d1\u5c55\uff0c\u6570\u636e\u79d1\u5b66\u9886\u57df\u53d6\u5f97\u4e86\u663e\u8457\u8fdb\u5c55\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u673a\u5668\u7ffb\u8bd1\u3001\u63a8\u8350\u7cfb\u7edf\u3001\u793e&hellip;","_links":{"self":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/6245","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=6245"}],"version-history":[{"count":2,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/6245\/revisions"}],"predecessor-version":[{"id":6249,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/6245\/revisions\/6249"}],"wp:attachment":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6245"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6245"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6245"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}