{"id":310,"date":"2025-09-09T14:07:57","date_gmt":"2025-09-09T06:07:57","guid":{"rendered":"http:\/\/159.223.199.105\/?p=310"},"modified":"2025-09-15T14:11:47","modified_gmt":"2025-09-15T06:11:47","slug":"%e5%85%a8%e7%90%83%e5%a4%a7%e6%a8%a1%e5%9e%8b%e7%bc%96%e7%a8%8b%e8%af%84%e6%b5%8b%ef%bc%81%e5%b7%a5%e7%a8%8b%e8%83%bd%e5%8a%9b%e6%89%8d%e6%98%af%e5%85%b3%e9%94%ae%ef%bc%8c%e5%88%ab%e8%a2%ab%e3%80%8c","status":"publish","type":"post","link":"http:\/\/www.applicationsofllm.com\/index.php\/2025\/09\/09\/%e5%85%a8%e7%90%83%e5%a4%a7%e6%a8%a1%e5%9e%8b%e7%bc%96%e7%a8%8b%e8%af%84%e6%b5%8b%ef%bc%81%e5%b7%a5%e7%a8%8b%e8%83%bd%e5%8a%9b%e6%89%8d%e6%98%af%e5%85%b3%e9%94%ae%ef%bc%8c%e5%88%ab%e8%a2%ab%e3%80%8c\/","title":{"rendered":"\u5168\u7403\u5927\u6a21\u578b\u7f16\u7a0b\u8bc4\u6d4b\uff01\u5de5\u7a0b\u80fd\u529b\u624d\u662f\u5173\u952e\uff0c\u522b\u88ab\u300c\u5237\u699c\u6210\u7ee9\u300d\u9a97\u4e86"},"content":{"rendered":"\n<p>\u8fd1\u4e24\u5e74\uff0c\u5927\u8bed\u8a00\u6a21\u578b\u7684\u7f16\u7a0b\u80fd\u529b\u53d1\u5c55\u98de\u5feb\uff0c\u4f8b\u5982 DeepMind \u7684 AlphaCode \u66fe\u5ba3\u79f0\u8fbe\u5230\u4eba\u7c7b\u7ade\u6280\u7f16\u7a0b\u9009\u624b\u7684\u6c34\u5e73\uff1bOpenAI \u7684\u9876\u5c16\u6a21\u578b\u5c61\u5c61\u88ab\u62a5\u9053\u80fd\u901a\u8fc7\u8c37\u6b4c\u9ad8\u7ea7\u7f16\u7a0b\u9762\u8bd5\uff0c\u5e76\u5728&nbsp;LeetCode&nbsp;\u6311\u6218\u4e2d\u8868\u73b0\u51fa\u8f83\u9ad8\u80fd\u529b\u3002\u7136\u800c\uff0c\u5f53\u6211\u4eec\u5c06\u76ee\u5149\u4ece\u5927\u6a21\u578b\u5237\u699c\u8f6c\u5411\u89e3\u51b3\u771f\u5b9e\u7684\u3001\u590d\u6742\u7684\u5de5\u7a0b\u9879\u76ee\u65f6\uff0c\u4e00\u4e2a\u6838\u5fc3\u95ee\u9898\u968f\u4e4b\u6d6e\u73b0\uff1a<strong>\u8fd9\u4e9b\u53f7\u79f0\u64c5\u957f\u7f16\u7a0b\u7684\u6a21\u578b\u5728\u9762\u5bf9\u771f\u5b9e\u5de5\u7a0b\u573a\u666f\u65f6\uff0c\u5176\u51c6\u786e\u6027\u3001\u5065\u58ee\u6027\u548c\u5b9e\u9645\u5e94\u7528\u4ef7\u503c\u7a76\u7adf\u5982\u4f55\uff1f<\/strong><\/p>\n\n\n\n<p>\u7814\u7a76\u53d1\u73b0\uff0c\u73b0\u6709\u7684\u4ee3\u7801\u57fa\u51c6\u6570\u636e\u96c6\u5728\u9762\u5bf9\u590d\u6742\u7684\u5de5\u7a0b\u573a\u666f\u65f6\u666e\u904d\u5b58\u5728\u7f3a\u4e4f\u591a\u6837\u6027\u548c\u53ef\u63a7\u6027\u7684\u53cc\u91cd\u95ee\u9898\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u5de5\u7a0b\u5f00\u53d1\u73af\u8282\u8986\u76d6\u6709\u9650\uff1a<\/strong>\u5c3d\u7ba1\u73b0\u6709\u57fa\u51c6\uff08\u5982 FullStackBench\u3001SWEBench\uff09\u5728\u9886\u57df\u548c\u8bed\u8a00\u591a\u6837\u6027\u4e0a\u53d6\u5f97\u8fdb\u5c55\uff0c\u4f46\u5176\u4efb\u52a1\u7c7b\u578b\u4ecd\u4e3b\u8981\u96c6\u4e2d\u4e8e\u5355\u6bb5\u843d\u4ee3\u7801\u751f\u6210\u3002\u800c\u771f\u5b9e\u5de5\u7a0b\u5b9e\u8df5\u901a\u5e38\u6d89\u53ca\u8de8\u6587\u4ef6\u3001\u8de8\u6a21\u5757\u7684\u534f\u540c\uff0c\u4ee5\u53ca\u4ee3\u7801\u4fee\u590d\u3001\u6d4b\u8bd5\u9a71\u52a8\u5f00\u53d1\u3001\u591a\u51fd\u6570\u534f\u4f5c\u7b49\u590d\u6742\u4efb\u52a1\uff0c\u8fd9\u4e9b\u90fd\u5e94\u88ab\u5de5\u7a0b\u7ea7\u57fa\u51c6\u5168\u9762\u8986\u76d6\u3002<strong><br><br><\/strong><\/li>\n\n\n\n<li><strong>\u6570\u636e\u6784\u5efa\u65b9\u6cd5\u5c40\u9650\uff1a<\/strong>\u5927\u591a\u6570\u6570\u636e\u96c6\u91c7\u7528\u968f\u673a\u6316\u7a7a\u6216\u4ece\u4ee3\u7801\u4ed3\u5e93\u7684\u5386\u53f2\u00a0PR\u00a0\u8bb0\u5f55\u4e2d\u63d0\u53d6\u4fee\u6539\u70b9\uff08\u5982 GitHub \u7684 Pull Request\uff09\u3002\u524d\u8005\u5bb9\u6613\u5ffd\u7565\u9879\u76ee\u7684\u6838\u5fc3\u903b\u8f91\u4ee3\u7801\u6bb5\uff0c\u540e\u8005\u4e0d\u4ec5\u6570\u636e\u91cf\u7a00\u5c11\uff0c\u8fd8\u9700\u6295\u5165\u5927\u91cf\u4eba\u5de5\u8fdb\u884c\u6570\u636e\u6e05\u6d17\u548c\u6807\u6ce8\uff0c\u96be\u4ee5\u89c4\u6a21\u5316\u6784\u5efa\u9ad8\u8d28\u91cf\u7684\u8bc4\u6d4b\u9898\u76ee\u3002<\/li>\n<\/ul>\n\n\n\n<p><strong>\u603b\u4e4b\uff0c\u6211\u4eec\u53d1\u73b0\u73b0\u6709\u7684\u4ee3\u7801\u57fa\u51c6\u6d4b\u8bd5\u5927\u591a\u201c\u504f\u79d1\u201d\u4e86\u3002<\/strong>\u5b83\u4eec\u8981\u4e48\u53ea\u5173\u6ce8\u5355\u4e2a\u51fd\u6570\u7684\u8865\u5168\uff0c\u5ffd\u7565\u4e86\u5f00\u53d1\u8005\u4fee\u590d Bug \u3001\u6839\u636e\u5355\u5143\u6d4b\u8bd5\u53cd\u5411\u5f00\u53d1\u7684\u771f\u5b9e\u573a\u666f\uff1b\u8981\u4e48\u91c7\u7528\u968f\u673a\u6316\u7a7a\u7684\u65b9\u5f0f\uff0c\u96be\u4ee5\u89e6\u53ca\u4ee3\u7801\u7684\u6838\u5fc3\u903b\u8f91\u3002\u8fd9\u5bfc\u81f4\u6211\u4eec\u65e0\u6cd5\u79d1\u5b66\u3001\u5b8c\u6574\u3001\u5168\u9762\u5730\u6d4b\u8bc4&nbsp;LLM&nbsp;\u5728\u771f\u5b9e\u5de5\u7a0b\u573a\u666f\u4e2d\u7684\u4ee3\u7801\u80fd\u529b\uff0c\u5c24\u5176\u662f\u5728\u53ef\u9760\u6027\u548c\u9002\u7528\u6027\u65b9\u9762\uff0c\u6211\u4eec\u4e9f\u9700\u4e00\u4e2a\u80fd\u89e3\u51b3\u6b64\u96be\u9898\u7684\u65b9\u6848\u3002<\/p>\n\n\n\n<p>\u4e3a\u4e86\u5e94\u5bf9\u4e0a\u8ff0\u6311\u6218\uff0c<strong>&nbsp;Meituan-M17 \u56e2\u961f\u3001\u4e0a\u6d77\u4ea4\u5927\u8054\u5408\u53d1\u5e03\u4e86\u4e00\u4e2a\u5168\u65b0\u7684\u5927\u6a21\u578b\u5de5\u7a0b\u7ea7\u522b\u4ee3\u7801\u57fa\u51c6\u6d4b\u8bd5&nbsp;CoreCodeBench \u6570\u636e\u96c6\uff0c\u6258\u7ba1\u5230 AGI-Eval \u793e\u533a\u3002<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>CoreCodeBench\u699c\u5355\u5730\u5740\uff1a<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-embed\"><div class=\"wp-block-embed__wrapper\">\nhttps:\/\/agi-eval.cn\/evaluation\/detail?id=64\n<\/div><\/figure>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u5fae\u4fe1\u5c0f\u7a0b\u5e8f\uff1a<strong><a href=\"https:\/\/mp.weixin.qq.com\/s?__biz=Mzk2NDE2MjMxOA==&amp;mid=2247485852&amp;idx=1&amp;sn=23cd7a551e6c1710f0ac82c426033495&amp;chksm=c5e6614231dfb4e67e0da90f62527ce9386f55576301d6471d12f39b87714600022ccd018eb4&amp;scene=126&amp;sessionid=1757915777&amp;subscene=7&amp;clicktime=1757916413&amp;enterid=1757916413&amp;click_id=53&amp;key=daf9bdc5abc4e8d018eea76685abc34f2a0aa8b19d04885a4a4c1f3aeb84727702875532bc27c71bbf0f0a8ac5764f6a977b35e7720325d033ab970c47b9a862a4380ea32f24b27ca6a2fe2961ad3adbe1109c32731f773dee326f8ab963e9aa35b05b1636e3ecb79835521436c41012864dda92125b6fe4d2e8ea387b118fcd&amp;ascene=1&amp;uin=MjY5NDMzMDE1&amp;devicetype=UnifiedPCWindows&amp;version=f2541022&amp;lang=zh_CN&amp;countrycode=CN&amp;exportkey=n_ChQIAhIQlsL5bVqMz0jp41wIa6zuyhLcAQIE97dBBAEAAAAAAGjTJJ47U8YAAAAOpnltbLcz9gKNyK89dVj04htO%2BnbyIc6an7Z1PrRKBDr5abPb2P9omaZfOAL4e3pRxji01nQt%2FeYXZo5ZIBsvWJFnKzOusNPl8n3ZdzQ%2Fox40yb6hfq7PoUE6sMTVkXD3MNru%2FwTJIh6ic2i61v2zcof%2BfDueJui%2BkQ%2FsplAvJb8%2FnzBBzeo1gNAwjD4pbJs6dnPm0C%2Fsp2CNVxGHfNU1a7gGpB9R9PT4pm4QJTqTeGnEr0Vqn1N3xQ0JSdlKtYYtVkPuYQ0%3D&amp;acctmode=0&amp;pass_ticket=MTMa%2BqC0Qvp96ftUQyQnIpL7VYz4dR9MsQM4xwjGWuiN4TPAHu4AOOKwYyr0DEIf&amp;wx_header=0&amp;fasttmpl_type=0&amp;fasttmpl_fullversion=7906304-zh_CN-html&amp;from_xworker=1\">AGI-Eval\u6a21\u578b\u8bc4\u6d4b<\/a><\/strong><\/li>\n<\/ul>\n\n\n\n<p>\u76ee\u524d\u8be5\u7814\u7a76\u5df2\u7ecf\u53d1\u8868\u6210\u8bba\u6587\uff1a\u300aCoreCodeBench: A Configurable Multi-Scenario Repository-Level Benchmark\u300b\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"462\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-50-1024x462.jpg\" alt=\"\" class=\"wp-image-311\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-50-1024x462.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-50-300x135.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-50-768x347.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-50.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u8bba\u6587\u5730\u5740\uff1a<em><u><br><br><\/u><\/em><\/li>\n<\/ul>\n\n\n\n<p>http:\/\/arxiv.org\/abs\/2507.05281<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>GitHub\u5730\u5740\uff1a<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-embed\"><div class=\"wp-block-embed__wrapper\">\nhttps:\/\/github.com\/AGI-Eval-Official\/CoreCodeBench\n<\/div><\/figure>\n\n\n\n<p>\u5b83\u4e13\u6ce8\u4e8e\u8bc4\u4f30\u5927\u8bed\u8a00\u6a21\u578b\u5728\u771f\u5b9e\u5de5\u7a0b\u9879\u76ee\u4e2d\u7684\u7efc\u5408\u4ee3\u7801\u80fd\u529b\uff0c\u8986\u76d6\u4e86\u4ece\u4ee3\u7801\u5f00\u53d1\u5230\u4ee3\u7801\u4fee\u6b63\u7684\u591a\u4e2a\u6838\u5fc3\u9636\u6bb5\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"306\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-52-1024x306.jpg\" alt=\"\" class=\"wp-image-313\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-52-1024x306.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-52-300x90.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-52-768x229.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-52.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3 \u56fe 1: CoreCodeBench \u9898\u578b\u5c55\u793a<\/em><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"373\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-51-1024x373.jpg\" alt=\"\" class=\"wp-image-312\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-51-1024x373.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-51-300x109.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-51-768x280.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-51.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3 \u56fe 2: CoreCodeBench \u6a21\u578b\u80fd\u529b\u699c\u5355<\/em><\/p>\n\n\n\n<p>\u901a\u8fc7\u5728 CoreCodeBench \u4e0a\u5bf9\u5f53\u524d\u4e3b\u6d41\u5927\u8bed\u8a00\u6a21\u578b\u7684\u5168\u9762\u8bc4\u6d4b\uff0c\u6211\u4eec\u5f97\u51fa\u4e86\u4ee5\u4e0b<strong>\u5173\u952e\u7ed3\u8bba\uff1a<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u5927\u6a21\u578b\u7f16\u7a0b\u80fd\u529b\u8fed\u4ee3\u8fdb\u6b65\u663e\u8457\uff0c\u4f46\u53d1\u5c55\u4e0d\u5747\u8861\uff1a<\/strong>\u8f83\u65b0\u7684\u6a21\u578b {\u5982 Claude 3.7 \u3001o4 mini\uff08high\uff09} \u76f8\u8f83\u4e8e\u524d\u4ee3\u4ea7\u54c1\u8868\u73b0\u51fa\u660e\u663e\u8fdb\u6b65\u3002\u7136\u800c\uff0c\u53d7\u6d4b\u6a21\u578b\u5728\u4ee3\u7801\u4fee\u6b63\uff08BugFix\uff09\u4efb\u52a1\u4e0a\u8868\u73b0\u6b20\u4f73\uff0c\u5c24\u5176\u662f\u5355\u51fd\u6570\u4efb\u52a1\u573a\u666f\u4e0b\uff0c\u4fee\u6b63\u4efb\u52a1\u7684\u6210\u529f\u7387\u5168\u90e8\u4f4e\u4e8e\u5f00\u53d1\u4efb\u52a1\uff0c\u8fd9\u63ed\u793a\u4e86\u5f53\u524d LLM \u5728\u7406\u89e3\u548c\u4fee\u590d\u6df1\u5c42\u903b\u8f91\u9519\u8bef\u65b9\u9762\u5b58\u5728\u7684\u666e\u904d\u77ed\u677f\u3002<\/li>\n\n\n\n<li><strong>\u591a\u51fd\u6570\u534f\u4f5c\u662f\u5f53\u524d\u5927\u6a21\u578b\u7f16\u7a0b\u573a\u666f\u7684\u4e3b\u8981\u74f6\u9888\uff1a<\/strong>\u51e0\u4e4e\u6240\u6709\u6a21\u578b\u5728\u5904\u7406\u591a\u51fd\u6570\u4efb\u52a1\u65f6\u7684\u8868\u73b0\u90fd\u663e\u8457\u52a3\u4e8e\u5355\u51fd\u6570\u4efb\u52a1\u3002\u8fd9\u8868\u660e\uff0c\u5f53\u9700\u8981\u540c\u65f6\u5904\u7406\u591a\u4e2a\u51fd\u6570\u95f4\u7684\u4f9d\u8d56\u5173\u7cfb\u3001\u8c03\u7528\u903b\u8f91\u548c\u534f\u540c\u5b9e\u73b0\u65f6\uff0c\u5f53\u524d\u5927\u6a21\u578b\u7684\u8de8\u51fd\u6570\u63a8\u7406\u548c\u89c4\u5212\u80fd\u529b\u5c1a\u663e\u4e0d\u8db3\u3002<\/li>\n\n\n\n<li><strong>\u5927\u6a21\u578b\u7f16\u7a0b\u573a\u666f\u666e\u904d\u7f3a\u4e4f\u7075\u6d3b\u7684\u89c4\u5212\u4e0e\u5206\u5c42\u63a8\u7406\u80fd\u529b\uff1a<\/strong>\u5728\u591a\u51fd\u6570\u4ee3\u7801\u751f\u6210\u4efb\u52a1\u4e2d\uff0c\u6211\u4eec\u89c2\u5bdf\u5230\u5927\u591a\u6570\u6a21\u578b\u4e25\u683c\u9075\u5faa\u8f93\u5165\u63d0\u793a\u4e2d\u7684\u51fd\u6570\u987a\u5e8f\u751f\u6210\u4ee3\u7801\uff0c\u800c\u975e\u50cf\u4eba\u7c7b\u5de5\u7a0b\u5e08\u90a3\u6837\uff0c\u57fa\u4e8e\u529f\u80fd\u4f9d\u8d56\uff08\u5982\u5148\u5b9e\u73b0\u88ab\u8c03\u7528\u7684\u5de5\u5177\u51fd\u6570\uff09\u8fdb\u884c\u4f18\u5316\u3002\u8fd9\u4e00\u73b0\u8c61\u53cd\u6620\u4e86\u5f53\u524d\u6a21\u578b\u5728\u9762\u5bf9\u590d\u6742\u4efb\u52a1\u65f6\uff0c\u503e\u5411\u4e8e\u91c7\u7528\u9ed8\u8ba4\u7684\u987a\u5e8f\u7b56\u7565\uff0c\u7f3a\u4e4f\u4e3b\u52a8\u89c4\u5212\u7684\u610f\u8bc6\u3002<\/li>\n<\/ul>\n\n\n\n<p><strong>\u4e0b\u9762\u6211\u4eec\u5c06\u8be6\u7ec6\u4ecb\u7ecd CoreCodeBench \u662f\u5982\u4f55\u6784\u5efa\u7684\uff0c\u4ee5\u53ca\u53d6\u5f97\u7684\u5b9e\u9a8c\u7ed3\u679c\u548c\u6df1\u5ea6\u5206\u6790\u3002<\/strong><\/p>\n\n\n\n<p><strong>01. \u57fa\u51c6\u6784\u5efa\u65b9\u6cd5\u4e0e\u5b9e\u9a8c\u5206\u6790<\/strong><\/p>\n\n\n\n<p><strong>\u6570\u636e\u96c6\u6784\u5efa\u65b9\u6cd5\uff1aCorePipe \u6d41\u7a0b<\/strong><\/p>\n\n\n\n<p>\u4e3a\u4e86\u6784\u5efa\u4e00\u4e2a\u65e2\u5173\u6ce8\u591a\u6837\u5316\u5de5\u7a0b\u95ee\u9898\uff0c\u53c8\u805a\u7126\u4e8e\u6838\u5fc3\u4ee3\u7801\u7684\u57fa\u51c6\uff0cCoreCodeBench \u4e2d\u8bbe\u8ba1\u4e86\u4ece\u5de5\u7a0b\u4ee3\u7801\u4ed3\u5e93\u5230\u591a\u79cd\u51fd\u6570\u9898\u578b\u7684<strong>\u5168\u81ea\u52a8\u5316<\/strong>\u6784\u5efa\u6d41\u7a0bCorePipe\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"380\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-53-1024x380.jpg\" alt=\"\" class=\"wp-image-314\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-53-1024x380.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-53-300x111.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-53-768x285.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-53.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3 \u56fe 3: CorePipe \u81ea\u52a8\u5316\u751f\u4ea7\u6570\u636e\u6d41\u7a0b\u793a\u610f\u56fe<\/em><br><br><\/p>\n\n\n\n<p>\u5982\u4e0a\u56fe\u6240\u793a\uff0cCorePipe \u57fa\u4e8e\u51fd\u6570\u8c03\u7528\u6811\uff0c\u7cfb\u7edf\u5316\u5730\u751f\u6210\u8986\u76d6\u4e09\u5927\u6838\u5fc3\u573a\u666f\u7684\u5355\u51fd\u6570\u4e0e\u591a\u51fd\u6570\u9898\u76ee\uff0c\u786e\u4fdd\u6bcf\u4e00\u9053\u9898\u76ee\u90fd\u76f4\u51fb\u201c\u8981\u5bb3\u201d\uff1a<\/p>\n\n\n\n<p><strong>\u7cbe\u9009\u771f\u5b9e\u9879\u76ee\uff1a<\/strong>\u6211\u4eec\u4ece PyPI \u5bf9\u5e94\u7684 GitHub \u4ed3\u5e93\u4e2d\u7b5b\u9009\u51fa\u9ad8\u6d3b\u8dc3\u5ea6\u3001\u9ad8\u6d4b\u8bd5\u8986\u76d6\u7387\u548c\u9ad8\u6280\u672f\u590d\u6742\u5ea6\u7684\u9876\u7ea7\u5f00\u6e90\u9879\u76ee\u3002<strong>\u5b9a\u4f4d\u6838\u5fc3\u4ee3\u7801\uff1a<\/strong>\u901a\u8fc7\u52a8\u6001\u548c\u9759\u6001\u8ffd\u8e2a\u4ee3\u7801\u7684\u6267\u884c\uff0c\u6211\u4eec\u9996\u5148\u6784\u5efa\u51fd\u6570\u8c03\u7528\u56fe\uff0c\u518d\u5229\u7528\u62bd\u8c61\u8bed\u6cd5\u6811\uff08AST\uff09\u62bd\u53d6\u51fa\u5173\u952e\u51fd\u6570\u4e2d\u7684\u6838\u5fc3\u4ee3\u7801\uff0c\u7cbe\u51c6\u5b9a\u4f4d\u9879\u76ee\u4e2d\u90a3\u4e9b\u201c\u7275\u4e00\u53d1\u800c\u52a8\u5168\u8eab\u201d\u7684\u6838\u5fc3\u4ee3\u7801\u5757\u3002\u6211\u4eec\u80fd\u7cbe\u51c6\u5b9a\u4f4d\u9879\u76ee\u4e2d\u90a3\u4e9b\u201c\u7275\u4e00\u53d1\u800c\u52a8\u5168\u8eab\u201d\u7684\u6838\u5fc3\u51fd\u6570\u3002<\/p>\n\n\n\n<p><strong>\u6a21\u62df\u4e09\u5927\u771f\u5b9e\u573a\u666f\uff1a<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u76f4\u63a5\u5f00\u53d1 (Development)\uff1a<\/strong>\u00a0\u4e0d\u4ec5\u4ec5\u662f\u586b\u7a7a\uff0c\u6211\u4eec\u5229\u7528 GPT-4o \u751f\u6210\u9ad8\u8d28\u91cf\u7684\u51fd\u6570\u529f\u80fd\u63cf\u8ff0\uff0c\u5e76\u7531 Claude 3.5 Sonnet \u8fdb\u884c\u201c\u6311\u523a\u201d\u548c\u5ba1\u6838\uff0c\u786e\u4fdd\u6a21\u578b\u662f\u5728\u7406\u89e3\u771f\u6b63\u9700\u6c42\u7684\u524d\u63d0\u4e0b\u8fdb\u884c\u5f00\u53d1\u3002<\/li>\n\n\n\n<li><strong>\u4ee3\u7801\u4fee\u6b63 (BugFix)\uff1a<\/strong>\u544a\u522b\u7b80\u5355\u7684\u8bed\u6cd5\u9519\u8bef\uff0c\u8f6c\u800c\u4f7f\u7528 LLM \u751f\u6210\u66f4\u9690\u853d\u3001\u66f4\u590d\u6742\u7684\u903b\u8f91\u9519\u8bef\uff0c\u771f\u5b9e\u6a21\u62df\u4e86\u5f00\u53d1\u4e2d\u90a3\u4e9b\u4ee4\u4eba\u5934\u75bc\u7684 Bug \u3002<\/li>\n\n\n\n<li><strong>\u6d4b\u8bd5\u9a71\u52a8\u5f00\u53d1 (TDD)\uff1a<\/strong>\u63d0\u4f9b\u5b8c\u6574\u7684\u5355\u5143\u6d4b\u8bd5\uff0c\u8981\u6c42\u6a21\u578b\u6839\u636e\u6d4b\u8bd5\u7528\u4f8b\u53cd\u5411\u5f00\u53d1\u529f\u80fd\u4ee3\u7801\uff0c\u8003\u5bdf\u5176\u9075\u5faa\u73b0\u4ee3\u5f00\u53d1\u8303\u5f0f\u7684\u80fd\u529b\u3002<\/li>\n<\/ul>\n\n\n\n<p><strong>\u5f15\u5165\u591a\u51fd\u6570\u96be\u9898\uff1a<\/strong>&nbsp;<\/p>\n\n\n\n<p>\u6211\u4eec\u5c06\u4e0a\u8ff0\u5355\u51fd\u6570\u95ee\u9898\u6309\u7167\u771f\u5b9e\u7684\u51fd\u6570\u8c03\u7528\u5173\u7cfb\u7ec4\u5408\u8d77\u6765\uff0c\u521b\u9020\u51fa\u66f4\u590d\u6742\u7684<strong>\u591a\u51fd\u6570\u9898\u76ee\uff0c<\/strong>\u5168\u9762\u8003\u9a8c\u6a21\u578b\u5728\u5b8f\u89c2\u5c42\u9762\u7684\u4ee3\u7801\u7ec4\u7ec7\u548c\u89c4\u5212\u80fd\u529b\u3002<\/p>\n\n\n\n<p><strong>02. \u5b9e\u9a8c\u7ed3\u679c\u4e0e\u6df1\u5ea6\u5206\u6790<\/strong><\/p>\n\n\n\n<p>\u4e3a\u786e\u4fdd\u8bc4\u6d4b\u7684\u79d1\u5b66\u6027\uff0c\u6211\u4eec\u91c7\u7528\u4e86\u4fe1\u606f\u589e\u76ca\u5206\u6570\uff08IG Score\uff09\u4f5c\u4e3a\u6838\u5fc3\u6307\u6807\uff0c\u5e76\u901a\u8fc7&nbsp;<strong>IG Filter<\/strong><strong>\uff08\u4fe1\u606f\u589e\u76ca\u8fc7\u6ee4\uff09<\/strong>\u548c<strong>\u4e13\u4e1a\u5de5\u7a0b\u5e08\u4eba\u5de5\u5ba1\u6838\uff08\u6700\u7ec8\u5408\u683c\u738778.55%\uff09<\/strong>\u5bf9\u9898\u76ee\u8d28\u91cf\u8fdb\u884c\u5145\u5206\u7684\u76d1\u6d4b\uff0c<strong>\u517c\u5177\u53ef\u8bfb\u6027\u3001\u51c6\u786e\u6027\u548c\u5b8c\u6574\u6027\u3002<\/strong><\/p>\n\n\n\n<p><strong>2.1 \u5355\u51fd\u6570\u4e0e\u591a\u51fd\u6570\u4efb\u52a1\u5206\u6790<\/strong><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"886\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-4-1024x886.jpeg\" alt=\"\" class=\"wp-image-315\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-4-1024x886.jpeg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-4-300x260.jpeg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-4-768x665.jpeg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-4.jpeg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3&nbsp;\u8868 1: CoreCodeBench \u5355\u51fd\u6570\u548c\u591a\u51fd\u6570\u4efb\u52a1\u699c\u5355<\/em><br><br><\/p>\n\n\n\n<p>\u5982\u4e0a\u56fe\u53ef\u4ee5\u770b\u51fa\uff0c\u5b9e\u9a8c\u7ed3\u679c\u6709\u529b\u5730\u652f\u6301\u4e86\u6211\u4eec\u7684\u6838\u5fc3\u7ed3\u8bba\uff0c Claude 3.7 \u5728\u6240\u6709\u4efb\u52a1\u4e2d\u8868\u73b0\u7a81\u51fa\u3002<\/p>\n\n\n\n<p>\u4f46\u6240\u6709\u6a21\u578b\u5728\u591a\u51fd\u6570\u4efb\u52a1\u4e0a\u7684\u666e\u904d\u8868\u73b0\u4e0b\u6ed1\u5dee\u4e8e\u5355\u6a21\u578b\u4efb\u52a1\uff0c\u8fd9\u53ef\u80fd\u662f\u56e0\u4e3a\u591a\u51fd\u6570\u4efb\u52a1\u9700\u540c\u65f6\u5904\u7406\u591a\u4e2a\u51fd\u6570\u95f4\u7684\u4f9d\u8d56\u5173\u7cfb\u3001\u8c03\u7528\u903b\u8f91\u548c\u534f\u540c\u5b9e\u73b0\uff0c\u5bf9\u5927\u8bed\u8a00\u6a21\u578b\u7684\u8de8\u51fd\u6570\u63a8\u7406\u548c\u89c4\u5212\u80fd\u529b\u8981\u6c42\u66f4\u9ad8\uff0c\u4ee5\u53ca\u5728 BugFix \u4efb\u52a1\u4e0a\u7684\u96c6\u4f53\u77ed\u677f\uff0c\u6e05\u6670\u5730\u52fe\u52d2\u51fa\u5f53\u524d\u6280\u672f\u7684\u80fd\u529b\u8fb9\u754c\u3002<\/p>\n\n\n\n<p><strong>2.2 \u6a21\u578b\u89c4\u5212\u80fd\u529b\u6d1e\u5bdf<br><br><\/strong><\/p>\n\n\n\n<p>\u591a\u51fd\u6570\u4efb\u52a1\u7684\u5b9e\u9a8c\u5206\u6790\u63ed\u793a\uff0c<strong>\u6a21\u578b\u7f3a\u4e4f\u5bf9\u5b9e\u73b0\u987a\u5e8f\u7684\u89c4\u5212\u80fd\u529b\u3002<\/strong><\/p>\n\n\n\n<p>\u5927\u591a\u6570\u6a21\u578b\u4e25\u683c\u9075\u5faa\u8f93\u5165\u63d0\u793a\u4e2d\u7684\u51fd\u6570\u987a\u5e8f\u751f\u6210\u4ee3\u7801\u3002\u5f53\u524d\u6a21\u578b\u5728\u591a\u51fd\u6570\u4ee3\u7801\u751f\u6210\u65f6\u7f3a\u4e4f\u7075\u6d3b\u89c4\u5212\u80fd\u529b\u4e0e\u5206\u5c42\u63a8\u7406\u80fd\u529b\uff0c\u5f80\u5f80\u91c7\u7528\u9ed8\u8ba4\u7684\u987a\u5e8f\u8f93\u51fa\u7b56\u7565\uff0c\u800c\u975e\u57fa\u4e8e\u903b\u8f91\u6216\u529f\u80fd\u4f9d\u8d56\u8fdb\u884c\u4f18\u5316\u3002<\/p>\n\n\n\n<p>\u8fd9\u79cd\u201c\u987a\u5e8f\u6267\u884c\u201d\u800c\u975e\u201c\u903b\u8f91\u6267\u884c\u201d\u7684\u7b56\u7565\uff0c\u662f\u5176\u4e0e\u4eba\u7c7b\u5de5\u7a0b\u5e08\u5728\u89e3\u51b3\u590d\u6742\u95ee\u9898\u601d\u8def\u4e0a\u7684\u4e00\u5927\u5dee\u5f02\u3002<\/p>\n\n\n\n<p><strong>2.3 \u6781\u9650\u6311\u6218<br><br><\/strong><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"610\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-54-1024x610.jpg\" alt=\"\" class=\"wp-image-316\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-54-1024x610.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-54-300x179.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-54-768x457.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-54.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3 \u56fe 4: CoreCodeBench-Difficult \u6570\u636e\u96c6\u7684\u6a21\u578b\u7ed3\u679c<\/em><\/p>\n\n\n\n<p>\u6211\u4eec\u901a\u8fc7\u653e\u5bbd\u591a\u51fd\u6570\u95ee\u9898\u7684\u590d\u6742\u5ea6\u9650\u5236\uff0c\u6784\u5efa\u4e86&nbsp;CoreCodeBench-Difficult&nbsp;\u6570\u636e\u96c6\u3002<\/p>\n\n\n\n<p>\u5728\u8be5\u6d4b\u8bd5\u4e2d\uff0c<strong>\u6240\u6709\u6a21\u578b\u7684\u901a\u8fc7\u7387\u5747\u4f4e\u4e8e 30 %\uff0c<\/strong>\u8fd9\u4e0d\u4ec5\u5370\u8bc1\u4e86\u8be5\u57fa\u51c6\u5728\u63ed\u793a\u6a21\u578b\u5c40\u9650\u6027\u65b9\u9762\u7684\u6709\u6548\u6027\uff0c\u4e5f\u4e3a\u672a\u6765\u6280\u672f\u7684\u7a81\u7834\u63d0\u4f9b\u4e86\u4e25\u82db\u7684\u6d4b\u8bd5\u5e73\u53f0\u3002<\/p>\n\n\n\n<p><strong>2.4 LLM \u4ee3\u7801\u80fd\u529b\u5168\u666f\u96f7\u8fbe\u56fe<\/strong><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"654\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-55-1024x654.jpg\" alt=\"\" class=\"wp-image-317\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-55-1024x654.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-55-300x192.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-55-768x491.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-55.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3 \u56fe 5: \u524d\u6cbf LLM \u4ee3\u7801\u80fd\u529b\u96f7\u8fbe\u56fe<\/em><\/p>\n\n\n\n<p>\u6211\u4eec\u5c06\u6a21\u578b\u7684<strong>\u516d\u4e2a\u6838\u5fc3\u573a\u666f<\/strong>\u8868\u73b0\u7ed8\u5236\u6210\u96f7\u8fbe\u56fe\uff0c\u53ef\u4ee5\u76f4\u89c2\u5730\u770b\u5230\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u6ca1\u6709\u4e00\u4e2a\u6a21\u578b\u80fd\u5728\u6240\u6709\u573a\u666f\u4e2d\u72ec\u5360\u9ccc\u5934\uff0c\u8bc1\u660e\u4e86 CoreCodeBench\u00a0<strong>\u8bc4\u4f30\u7ef4\u5ea6\u7684\u5168\u9762\u6027\u3002<\/strong><\/li>\n\n\n\n<li>\u5f00\u53d1\uff08Development\uff09\u548c\u6d4b\u8bd5\u9a71\u52a8\u5f00\u53d1\uff08TDD\uff09\u4efb\u52a1\u4e2d\uff0c\u5355\/\u591a\u51fd\u6570\u8868\u73b0\u5e76\u4e0d\u5b8c\u5168\u76f8\u5173\uff0c\u8bf4\u660e<strong>\u5f00\u53d1\u591a\u5173\u8054\u51fd\u6570\u9700\u8981\u989d\u5916\u7684\u89c4\u5212\u80fd\u529b\u3002<\/strong><\/li>\n\n\n\n<li>\u4ee3\u7801\u4fee\u6b63\uff08BugFix\uff09\u4efb\u52a1\u4e2d\uff0c\u5355\/\u591a\u51fd\u6570\u8868\u73b0\u9ad8\u5ea6\u76f8\u5173\uff0c\u8fd9\u8bf4\u660e<strong>\u8c03\u8bd5\u66f4\u4f9d\u8d56\u4e8e\u4e00\u79cd\u901a\u7528\u7684\u3001\u5c40\u90e8\u7684\u9519\u8bef\u4fee\u6b63\u6280\u80fd\u3002<\/strong><\/li>\n<\/ul>\n\n\n\n<p>\u4e3a\u8fdb\u4e00\u6b65\u5206\u6790\u5404\u8bc4\u6d4b\u7ef4\u5ea6\u4e4b\u95f4\u7684\u5173\u7cfb\uff0c\u6211\u4eec\u8ba1\u7b97\u4e86\u6240\u6709\u6a21\u578b\u5728\u516d\u4e2a\u7ef4\u5ea6\u4e0a\u7684\u76ae\u5c14\u900a\u76f8\u5173\u7cfb\u6570\u5e76\u7ed8\u5236\u70ed\u529b\u56fe\u3002<strong><br><br><\/strong><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"878\" src=\"http:\/\/159.223.199.105\/wp-content\/uploads\/2025\/09\/image-56-1024x878.jpg\" alt=\"\" class=\"wp-image-318\" srcset=\"http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-56-1024x878.jpg 1024w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-56-300x257.jpg 300w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-56-768x658.jpg 768w, http:\/\/www.applicationsofllm.com\/wp-content\/uploads\/2025\/09\/image-56.jpg 1108w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><em>\u25b3 \u56fe 6: \u4ee3\u7801\u80fd\u529b\u9879\u76f8\u5173\u5ea6\u5206\u6790<\/em><\/p>\n\n\n\n<p>\u5982\u4e0a\u56fe\u6240\u793a\u53ef\u4ee5\u89c2\u5bdf\u5f97\u5230\uff0c\u76f8\u5173\u7cfb\u6570\u7684\u6d4b\u7b97\u7ed3\u679c\u8868\u660e\uff0cCoreCodeBench \u7684\u516d\u4e2a\u6838\u5fc3\u573a\u666f\u4e4b\u95f4\u65e2\u5b58\u5728\u4e00\u5b9a\u7684<strong>\u76f8\u5173\u6027\uff0c<\/strong>\u4e5f\u4f53\u73b0\u51fa\u5404\u81ea\u7684<strong>\u5dee\u5f02\u6027\uff1a<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Single-function \u4efb\u52a1\u4e4b\u95f4\u76f8\u5173\u6027\u8f83\u9ad8\uff0c\u8868\u73b0\u51fa<strong>\u5355\u51fd\u6570\u4efb\u52a1\u5728\u57fa\u7840\u7f16\u7a0b\u3001\u7406\u89e3\u548c\u5b9e\u73b0\u80fd\u529b\u4e0a\u7684\u5171\u6027\u3002<\/strong><\/li>\n\n\n\n<li>Multi-TDD \u548c Multi-Development \u5b58\u5728\u4e00\u5b9a\u7684\u76f8\u5173\u6027\uff0c\u8fd9\u662f\u56e0\u4e3a\u00a0<strong>Multi-function <\/strong><strong>\u4efb\u52a1\u901a\u5e38\u8003\u5bdf\u6a21\u578b\u5728\u66f4\u590d\u6742\u573a\u666f\u4e0b\u7684\u7efc\u5408\u80fd\u529b\uff0c<\/strong>\u5305\u62ec\u591a\u6b65\u63a8\u7406\u3001\u5b9e\u73b0\u89c4\u5212\u7b49\uff0c\u4e0e\u5355\u51fd\u6570\u4efb\u52a1\u6240\u9700\u7684\u57fa\u7840\u80fd\u529b\u5b58\u5728\u660e\u663e\u533a\u522b\u3002<\/li>\n\n\n\n<li>Multi-BugFix\u00a0\u867d\u7136\u5c5e\u4e8e\u591a\u51fd\u6570\u4efb\u52a1\uff0c\u4f46\u5b83\u548c\u5355\u51fd\u6570\u4efb\u52a1\u76f8\u5173\u6027\u9ad8\uff0c\u53cd\u800c\u548c Multi-TDD\u3001Multi-Development \u76f8\u5173\u6027\u4f4e\u3002\u8fd9\u662f\u56e0\u4e3a<strong>\u00a0Multi-BugFix\u00a0\u4efb\u52a1\u7684\u672c\u8d28\u66f4\u63a5\u8fd1\u4e8e\u201c\u5355\u70b9\u6392\u67e5\u201d\uff0c\u5b83\u66f4\u4fa7\u91cd\u4e8e\u5177\u4f53\u7ec6\u8282\u6216\u67d0\u4e00\u5c40\u90e8\u7684\u80fd\u529b\u8003\u5bdf\uff0c\u800c\u4e0e\u9700\u8981\u5168\u5c40\u7efc\u5408\u80fd\u529b\u7684\u591a\u51fd\u6570\u4efb\u52a1\u5b58\u5728\u5dee\u5f02\u3002<\/strong><\/li>\n<\/ul>\n\n\n\n<p><strong>03. \u603b\u7ed3\u4e0e\u5c55\u671b<\/strong><\/p>\n\n\n\n<p>CoreCodeBench&nbsp;\u7684\u6784\u5efa\u4e0e\u5e94\u7528\uff0c\u65e8\u5728\u4e3a\u5927\u8bed\u8a00\u6a21\u578b\u7684\u4ee3\u7801\u80fd\u529b\u8bc4\u4f30\u63d0\u4f9b\u4e00\u628a\u66f4\u79d1\u5b66\u3001\u66f4\u5168\u9762\u3001\u66f4\u8d34\u8fd1\u771f\u5b9e\u7684\u201c\u5de5\u7a0b\u6807\u5c3a\u201d\u3002\u56de\u987e\u6211\u4eec\u7684\u7814\u7a76\uff0c\u6211\u4eec\u7cfb\u7edf\u6027\u5730\u63ed\u793a\u4e86\u5f53\u524d\u9876\u5c16 LLM \u5728\u771f\u5b9e\u5de5\u7a0b\u573a\u666f\u4e2d\u7684\u6838\u5fc3\u77ed\u677f\uff1a<strong>\u65e0\u8bba\u662f\u591a\u4e48\u5148\u8fdb\u7684\u6a21\u578b\uff0c\u90fd\u5728\u903b\u8f91\u9519\u8bef\u4fee\u590d\u65b9\u9762\u6b65\u5c65\u7ef4\u8270\uff1b\u5728\u9762\u5bf9\u591a\u51fd\u6570\u534f\u540c\u4efb\u52a1\u65f6\uff0c\u5176\u8de8\u51fd\u6570\u63a8\u7406\u4e0e\u89c4\u5212\u80fd\u529b\u90fd\u663e\u5f97\u6349\u895f\u89c1\u8098\uff1b\u5e76\u4e14\uff0c\u5b83\u4eec\u666e\u904d\u7f3a\u4e4f\u4eba\u7c7b\u5de5\u7a0b\u5e08\u6240\u5177\u5907\u7684\u7075\u6d3b\u89c4\u5212\u4e0e\u5206\u5c42\u63a8\u7406\u80fd\u529b\u3002<\/strong><\/p>\n\n\n\n<p>\u7136\u800c\uff0c\u8fd9\u4e9b\u88ab\u63ed\u793a\u7684\u5c40\u9650\u6027\u5e76\u975e\u6280\u672f\u7684\u7ec8\u70b9\uff0c\u800c\u662f\u4e3a\u4e0b\u4e00\u4ee3\u5927\u8bed\u8a00\u6a21\u578b\u7684\u53d1\u5c55\u6307\u660e\u4e86\u6e05\u6670\u7684\u4f18\u5316\u65b9\u5411\u3002\u6211\u4eec\u76f8\u4fe1\uff0c\u901a\u8fc7\u5728 CoreCodeBench \u8fd9\u7c7b\u66f4\u8d34\u8fd1\u771f\u5b9e\u5de5\u7a0b\u9700\u6c42\u7684\u57fa\u51c6\u4e0a\u8fdb\u884c\u8bad\u7ec3\u548c\u8fed\u4ee3\uff0c\u5927\u8bed\u8a00\u6a21\u578b\u5c06\u80fd\u66f4\u5feb\u5730\u4ece\u4e00\u4e2a\u201c\u4ee3\u7801\u7247\u6bb5\u751f\u6210\u5668\u201d\uff0c\u8fdb\u5316\u6210\u4e00\u4e2a\u771f\u6b63\u5177\u5907\u5206\u6790\u3001\u89c4\u5212\u548c\u89e3\u51b3\u590d\u6742\u5de5\u7a0b\u95ee\u9898\u7684\u201c\u865a\u62df\u8f6f\u4ef6\u5de5\u7a0b\u5e08\u201d\uff0c\u4ece\u800c\u5728\u8f6f\u4ef6\u5f00\u53d1\u9886\u57df\u91ca\u653e\u51fa\u66f4\u6df1\u8fdc\u7684\u53d8\u9769\u529b\u91cf\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u8fd1\u4e24\u5e74\uff0c\u5927\u8bed\u8a00\u6a21\u578b\u7684\u7f16\u7a0b\u80fd\u529b\u53d1\u5c55\u98de\u5feb\uff0c\u4f8b\u5982 DeepMind \u7684 AlphaCode \u66fe\u5ba3\u79f0\u8fbe\u5230\u4eba\u7c7b\u7ade\u6280\u7f16\u7a0b\u9009 [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":319,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[],"class_list":["post-310","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-8"],"_links":{"self":[{"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/posts\/310","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/comments?post=310"}],"version-history":[{"count":1,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/posts\/310\/revisions"}],"predecessor-version":[{"id":320,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/posts\/310\/revisions\/320"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/media\/319"}],"wp:attachment":[{"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/media?parent=310"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/categories?post=310"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.applicationsofllm.com\/index.php\/wp-json\/wp\/v2\/tags?post=310"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}