{"id":9476,"date":"2025-11-11T16:31:32","date_gmt":"2025-11-11T08:31:32","guid":{"rendered":"\/?p=9476"},"modified":"2025-11-11T16:45:01","modified_gmt":"2025-11-11T08:45:01","slug":"text2sql-nl2sql%ef%bc%88%e4%b8%80%ef%bc%89%e6%95%b0%e6%8d%ae%e9%9b%86%ef%bc%9a%e8%87%aa%e7%84%b6%e8%af%ad%e8%a8%80%e4%b8%8e%e6%95%b0%e6%8d%ae%e5%ba%93%e7%9a%84%e7%bf%bb%e8%af%91%e6%a1%a5","status":"publish","type":"post","link":"\/?p=9476","title":{"rendered":"Text2SQL \/ NL2SQL\uff08\u4e00\uff09\u6570\u636e\u96c6\uff1a\u81ea\u7136\u8bed\u8a00\u4e0e\u6570\u636e\u5e93\u7684\u201c\u7ffb\u8bd1\u6865\u6881\u201d"},"content":{"rendered":"<h2>\u80cc\u666f<\/h2>\n<p>\ud83c\udfaf\u81ea\u7136\u8bed\u8a00\u4e0e\u6570\u636e\u5e93\u7684\u6c9f\u901a\u56f0\u5883<\/p>\n<p>\u5728\u5f53\u4eca\u6570\u5b57\u5316\u65f6\u4ee3\uff0c\u6570\u636e\u5e93\u4f5c\u4e3a\u6570\u636e\u5b58\u50a8\u548c\u7ba1\u7406\u7684\u6838\u5fc3\u5de5\u5177\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u5404\u4e2a\u9886\u57df\u3002\u65e0\u8bba\u662f\u4f01\u4e1a\u7684\u6570\u636e\u7ba1\u7406\u3001\u79d1\u7814\u7684\u6570\u636e\u7814\u7a76\uff0c\u8fd8\u662f\u65e5\u5e38\u7684\u4fe1\u606f\u67e5\u8be2\uff0c\u6211\u4eec\u90fd\u79bb\u4e0d\u5f00\u6570\u636e\u5e93\u3002\u7136\u800c\uff0c\u4f20\u7edf\u7684\u6570\u636e\u5e93\u67e5\u8be2\u8bed\u8a00 \u2014\u2014SQL\uff0c\u5374\u50cf\u4e00\u9053\u95e8\u69db\uff0c\u6a2a\u4e98\u5728\u666e\u901a\u7528\u6237\u4e0e\u6570\u636e\u5e93\u4e4b\u95f4\u3002<\/p>\n<p><img src=\"https:\/\/lemon-guess.oss-cn-hangzhou.aliyuncs.com\/img\/640.png\" alt=\"\u56fe\u7247\" \/><\/p>\n<p>\u60f3\u8c61\u4e00\u4e0b\uff0c\u4f60\u662f\u4e00\u4f4d\u4f01\u4e1a\u7684\u5e02\u573a\u5206\u6790\u5e08\uff0c\u60f3\u8981\u4ece\u516c\u53f8\u7684\u6570\u636e\u5e93\u4e2d\u83b7\u53d6\u8fc7\u53bb\u4e00\u4e2a\u6708\u5185\uff0c\u8d2d\u4e70\u6b21\u6570\u8d85\u8fc7 5 \u6b21\u4e14\u6d88\u8d39\u91d1\u989d\u5927\u4e8e 1000 \u5143\u7684\u5ba2\u6237\u4fe1\u606f\u3002\u9762\u5bf9 SQL \u590d\u6742\u7684\u8bed\u6cd5\u89c4\u5219\uff0c\u4f60\u53ef\u80fd\u4f1a\u611f\u5230\u65e0\u4ece\u4e0b\u624b\u3002\u7f16\u5199 SQL \u8bed\u53e5\u9700\u8981\u719f\u6089\u5404\u79cd\u5173\u952e\u5b57\u3001\u51fd\u6570\u548c\u8bed\u6cd5\u7ed3\u6784\uff0c\u7a0d\u6709\u4e0d\u614e\u5c31\u4f1a\u51fa\u9519\u3002\u5bf9\u4e8e\u4e0d\u5177\u5907\u4e13\u4e1a\u7f16\u7a0b\u77e5\u8bc6\u7684\u4eba\u6765\u8bf4\uff0c\u8fd9\u65e0\u7591\u662f\u4e00\u9879\u8270\u5de8\u7684\u4efb\u52a1\u3002<\/p>\n<p>\u5373\u4f7f\u662f\u5bf9\u4e8e\u6709\u4e00\u5b9a\u7f16\u7a0b\u57fa\u7840\u7684\u4eba\uff0c\u7f16\u5199\u590d\u6742\u7684 SQL \u67e5\u8be2\u4e5f\u5e76\u975e\u6613\u4e8b\u3002\u5f53\u6d89\u53ca\u5230\u591a\u4e2a\u8868\u7684\u5173\u8054\u67e5\u8be2\u3001\u590d\u6742\u7684\u6761\u4ef6\u7b5b\u9009\u548c\u6570\u636e\u805a\u5408\u65f6\uff0cSQL \u8bed\u53e5\u4f1a\u53d8\u5f97\u5197\u957f\u800c\u590d\u6742\uff0c\u96be\u4ee5\u7406\u89e3\u548c\u7ef4\u62a4\u3002\u800c\u4e14\uff0c\u4e0d\u540c\u7684\u6570\u636e\u5e93\u7cfb\u7edf\u53ef\u80fd\u8fd8\u5b58\u5728\u4e00\u4e9b\u8bed\u6cd5\u5dee\u5f02\uff0c\u8fd9\u8fdb\u4e00\u6b65\u589e\u52a0\u4e86\u5b66\u4e60\u548c\u4f7f\u7528\u7684\u96be\u5ea6\u3002<\/p>\n<p><strong>\ud83c\udfafText2SQL \/ NL2SQL \u662f\u4ec0\u4e48<\/strong><\/p>\n<p>Text2SQL \/ NL2SQL\uff0c\u5373 \u201cText-to-SQL\u201d \u6216 \u201cNatural Language to SQL\u201d \uff0c\u987e\u540d\u601d\u4e49\uff0c\u5c31\u662f\u5c06\u81ea\u7136\u8bed\u8a00\uff08Natural Language\uff0cNL\uff09\u95ee\u9898\u8f6c\u5316\u4e3a\u5728\u5173\u7cfb\u578b\u6570\u636e\u5e93\u4e2d\u53ef\u4ee5\u6267\u884c\u7684\u7ed3\u6784\u5316\u67e5\u8be2\u8bed\u8a00\uff08Structured Query Language\uff0cSQL\uff09\uff0c\u5b83\u4e5f\u662f\u4f20\u7edf NLP \u7684\u4efb\u52a1\u4e4b\u4e00\u3002\u5b83\u7684\u6838\u5fc3\u5728\u4e8e\u8ba9\u7528\u6237\u80fd\u591f\u7528\u65e5\u5e38\u7684\u8bed\u8a00\u4e0e\u6570\u636e\u5e93\u8fdb\u884c\u4ea4\u4e92\uff0c\u800c\u65e0\u9700\u7f16\u5199\u590d\u6742\u7684 SQL \u8bed\u53e5\u3002<\/p>\n<p><strong>\u5728\u5b9e\u9645\u4ea7\u4e1a\u754c\u5e94\u7528\u4e2d\uff0cText2SQL\u5f80\u5f80\u4f5c\u4e3a\u6838\u5fc3\u5f15\u64ce\uff0c\u63d0\u4f9b\u7528\u6237\u6570\u636e\u6316\u6398\u3001\u62a5\u544a\u5904\u7406\u7684\u81ea\u7136\u4ea4\u4e92\u65b9\u5f0f<\/strong>\uff0c\u901a\u8fc7\u4ece\u751f\u6210\u7684 SQL \u8bed\u53e5\uff0c\u62ff\u5230\u751f\u4ea7\u7684\u5b9e\u9645\u6570\u636e\uff0c\u8f93\u51fa\u65b9\u5f0f\u53ef\u80fd\u4e3a\u8868\u683c\u3001\u56fe\u8868\u3001\u62a5\u544a\u7b49\u3002<\/p>\n<p><img src=\"https:\/\/lemon-guess.oss-cn-hangzhou.aliyuncs.com\/img\/640-20251111162849793.png\" alt=\"\u56fe\u7247\" \/><\/p>\n<p>\u865a\u7ebf\u6846\u5185\uff0c\u662f\u81ea\u52a8\u5904\u7406\u7684\u8fc7\u7a0b\uff0c\u7528\u6237\u65e0\u611f\u77e5\uff0c\u7528\u6237\u53ea\u4f1a\u4f53\u4f1a\u4fbf\u662f\uff1a<\/p>\n<p>\u8f93\u5165\u5de6\u4fa7\u7684\u8981\u6c42\uff0c\u4fbf\u5f97\u5230\u6700\u53f3\u4fa7\u7684\u56fe\u8868\u3002<\/p>\n<p><strong>\ud83c\udfaf\u6570\u636e\u96c6<\/strong><\/p>\n<p>\u4f17\u6240\u5468\u77e5\uff0c\u6570\u636e\u96c6\u662f AI \u5e94\u7528\u7684\u539f\u6599\uff0c\u5904\u4e8e\u7b97\u529b\u3001\u7b97\u6cd5\u3001\u6570\u636e\u4e09\u8981\u7d20\u4e4b\u4e00\u3002\u5c24\u5176\u662f\u73b0\u5728\uff0c\u7b97\u529b\u3001\u7b97\u6cd5\u5df2\u7ecf\u53d6\u5f97\u4e86\u91cd\u5927\u8fdb\u5c55\uff0c\u4e14\u53ef\u4ee5\u76f4\u63a5\u8c03\u7528 API \u80fd\u529b\uff0c\u90a3\u9488\u5bf9\u66f4\u591a\u5782\u76f4\u9886\u57df\uff0c\u8981\u53d1\u5c55 AI \u5e94\u7528\uff0c<strong>\u6570\u636e\u4fbf\u662f\u6700\u91cd\u8981\u7684\u8d44\u6e90<\/strong>\u3002<\/p>\n<p><strong>Text2SQL \u6570\u636e\u96c6\u5206\u7c7b\uff1a<\/strong><\/p>\n<ul>\n<li>\n<p>\u6839\u636e\u5305\u542b\u9886\u57df\u6570\u91cf\uff0c\u6570\u636e\u96c6\u5206\u4e3a\u5355\u9886\u57df\u548c\u591a\u9886\u57df\u3002<\/p>\n<\/li>\n<li>\n<p>\u6839\u636e\u6bcf\u4e2a\u6570\u636e\u5e93\u5305\u542b\u8868\u7684\u6570\u91cf\uff0c\u6570\u636e\u96c6\u5206\u4e3a\u5355\u8868\u548c\u591a\u8868\u6a21\u5f0f\u3002\u5728\u591a\u8868\u6a21\u5f0f\u4e2d\uff0cSQL\u751f\u6210\u6d89\u53ca\u5230\u8868\u683c\u7684\u9009\u62e9\u3002<\/p>\n<\/li>\n<li>\n<p>\u6839\u636e\u95ee\u9898\u590d\u6742\u5ea6\uff0c\u6570\u636e\u96c6\u5206\u4e3a\u7b80\u5355\u95ee\u9898\u548c\u590d\u6742\u95ee\u9898\u6a21\u5f0f\uff0c\u5176\u4e2d\u95ee\u9898\u590d\u6742\u5ea6\u7531SQL\u67e5\u8be2\u8bed\u53e5\u6d89\u53ca\u5230\u7684\u5173\u952e\u8bcd\u6570\u91cf\u3001\u5d4c\u5957\u5c42\u6b21\u3001\u5b50\u53e5\u6570\u91cf\u7b49\u786e\u5b9a\u3002<\/p>\n<\/li>\n<li>\n<p>\u6839\u636e\u5b8c\u6574SQL\u751f\u6210\u6240\u9700\u8f6e\u6570\uff0c\u6570\u636e\u96c6\u5206\u4e3a\u5355\u8f6e\u548c\u591a\u8f6e\u3002<\/p>\n<\/li>\n<li>\n<p>\u82e5SQL\u751f\u6210\u878d\u8fdb\u6e10\u8fdb\u5f0f\u5bf9\u8bdd\uff0c\u5219\u6570\u636e\u96c6\u589e\u52a0\u201c\u7ed3\u5408\u5bf9\u8bdd\u201d\u6807\u8bb0\u3002\u5f53\u524d\u53ea\u6709CoSQL\u6570\u636e\u96c6\u662f\u878d\u8fdb\u5bf9\u8bdd\u7684\u6570\u636e\u96c6\u3002<\/p>\n<\/li>\n<\/ul>\n<p><img src=\"https:\/\/lemon-guess.oss-cn-hangzhou.aliyuncs.com\/img\/640-20251111162849911.png\" alt=\"\u56fe\u7247\" \/><\/p>\n<p><strong>\u4e1a\u754c\u5e38\u7528\u6570\u636e\u96c6\uff1a<\/strong><\/p>\n<h2>WikiSQL<\/h2>\n<ul>\n<li>\n<p>\u63d0\u51fa\u65b9\uff1aSalesforce<\/p>\n<\/li>\n<li>\n<p>2017<\/p>\n<\/li>\n<li>\n<p>\u5305\u542b\u4e86 24,241 \u5f20\u8868\uff0c80,645\u6761\u81ea\u7136\u8bed\u8a00\u95ee\u53e5\u53ca\u76f8\u5e94\u7684SQL\u8bed\u53e5<\/p>\n<\/li>\n<li>\n<p><a href=\"https:\/\/github.com\/salesforce\/WikiSQL\">https:\/\/github.com\/salesforce\/WikiSQL<\/a><\/p>\n<\/li>\n<li>\n<p><a href=\"https:\/\/arxiv.org\/pdf\/1709.00103\">https:\/\/arxiv.org\/pdf\/1709.00103<\/a><\/p>\n<\/li>\n<\/ul>\n<p>We evaluate using the execution accuracy metric Accex = Nex \/ N . One downside of Accex is that it is possible to construct a SQL query that does not correspond to the question but nevertheless obtains the same result. <\/p>\n<p>For example, the two queries SELECT COUNT(name) WHERE SSN = 123 and SELECT COUNT(SSN) WHERE SSN = 123 produce the same result if no two people with different names share the SSN 123. Hence, we also use the logical form accuracy Acclf = Nlf \/ N . However, as we showed in Section 2.2, Acclf incorrectly penalizes queries that achieve the correct result but do not have exact string match with the ground truth query. Due to these observations, we use both metrics to evaluate the models.  <\/p>\n<h2>Spider<\/h2>\n<ul>\n<li>Spider\uff1a<a href=\"https:\/\/arxiv.org\/pdf\/1809.08887\">https:\/\/arxiv.org\/pdf\/1809.08887<\/a><\/li>\n<li><a href=\"https:\/\/yale-lily.github.io\/spider\">https:\/\/yale-lily.github.io\/spider<\/a><\/li>\n<li><a href=\"https:\/\/spider2-sql.github.io\/\">https:\/\/spider2-sql.github.io\/<\/a><\/li>\n<li>2018\u5e749\u6708\uff0c\u8036\u9c81\u5927\u5b66\u63d0\u51fa\u7684\u591a\u6570\u636e\u5e93\u3001\u591a\u8868\u3001\u5355\u8f6e\u67e5\u8be2\u7684Text-to-SQL\u6570\u636e\u96c6\uff0c\u4e5f\u662f\u4e1a\u754c\u516c\u8ba4\u96be\u5ea6\u6700\u5927\u7684\u5927\u89c4\u6a21\u8de8\u9886\u57df\u8bc4\u6d4b\u699c\u5355\uff0c\u5305\u542b\u4e8610181\u4e2a\u81ea\u7136\u8bed\u8a00\u95ee\u9898\uff0c5693\u4e2aSQL\u8bed\u53e5\uff0c\u6d89\u53ca138\u4e2a\u4e0d\u540c\u9886\u57df\u7684200\u591a\u4e2a\u6570\u636e\u5e93\uff0c\u96be\u6613\u7a0b\u5ea6\u5206\u4e3a\uff1a\u7b80\u5355\u3001\u4e2d\u7b49\u3001\u56f0\u96be\u3001\u7279\u522b\u56f0\u96be\u30022024\u5e742\u6708\uff0c\u8036\u9c81\u5927\u5b66\u5f00\u6e90\u4e86Spider1.0\u6392\u884c\u699c\u5355\u7684test\u6570\u636e\u96c6\uff0c\u5e76\u4e14\u4ed6\u4eec\u5c06\u57283\u6708\u5f00\u6e90<strong>Spider 2.0<\/strong>\u6570\u636e\u96c6\u3002<\/li>\n<li>Spider 2.0\uff0c\u5305\u542b\u6765\u81ea\u4f01\u4e1a\u6570\u636e\u5e93\u7684 632 \u4e2a\u771f\u5b9e\u4e16\u754c\u7684\u6587\u672c\u5230 SQL \u4efb\u52a1\u3002\u8fd9\u4e9b\u6570\u636e\u5e93\u901a\u5e38\u6709 1000 \u591a\u5217\uff0c\u6765\u81ea\u4e91\u6216\u672c\u5730\u7cfb\u7edf\uff0c\u5982 BigQuery\u3001Snowflake \u548c PostgreSQL\u3002\u89e3\u51b3\u8fd9\u4e9b\u4efb\u52a1\u9700\u8981\u6a21\u578b\u7406\u89e3\u6570\u636e\u5e93\u5143\u6570\u636e\u3001\u9879\u76ee\u7279\u5b9a\u8bed\u8a00\u548c\u9879\u76ee\u4ee3\u7801\uff0c\u5bfc\u822a\u590d\u6742\u7684 SQL \u73af\u5883\u5e76\u5904\u7406\u957f\u4e0a\u4e0b\u6587\u3002\u6a21\u578b\u5fc5\u987b\u6267\u884c\u9ad8\u7ea7\u63a8\u7406\u5e76\u751f\u6210\u591a\u6837\u5316\u7684 SQL \u67e5\u8be2\uff0c\u6709\u65f6\u8d85\u8fc7 100 \u884c\uff0c\u8d85\u8fc7\u4f20\u7edf\u7684\u6587\u672c\u5230 SQL \u6311\u6218\u3002<\/li>\n<li>\u4e0b\u56fe\u4e3aSpider 1\u96be\u6613\u5212\u5206\u4e3e\u4f8b\uff1a<\/li>\n<\/ul>\n<p><img src=\"https:\/\/lemon-guess.oss-cn-hangzhou.aliyuncs.com\/img\/640-20251111162850138.png\" alt=\"\u56fe\u7247\" \/><\/p>\n<h2>SParC<\/h2>\n<ul>\n<li><a href=\"https:\/\/yale-lily.github.io\/sparc\">https:\/\/yale-lily.github.io\/sparc<\/a><\/li>\n<li>Yale &amp; Salesforce<\/li>\n<\/ul>\n<p><strong><em>SParC<\/em><\/strong> is a dataset for cross-domain <strong>S<\/strong>emantic <strong>Par<\/strong>sing in <strong>C<\/strong>ontext. It is the context-dependent\/multi-turn version of the Spider task, a complex and cross-domain text-to-SQL challenge. SParC consists of 4,298 coherent question sequences (12k+ unique individual questions annotated with SQL queries annotated by 14 Yale students), obtained from user interactions with 200 complex databases over 138 domains.<\/p>\n<p>\u4e0b\u56fe\u4e3a\u6570\u636e\u96c6\u793a\u4f8b\uff1a<\/p>\n<p><img src=\"https:\/\/lemon-guess.oss-cn-hangzhou.aliyuncs.com\/img\/640-20251111162850295.png\" alt=\"\u56fe\u7247\" \/><\/p>\n<h2>CoSQL<\/h2>\n<p><strong>\u91cd\u70b9\u5728\u4e8e\uff0c\u7cfb\u7edf\u53ef\u4ee5\u4ece\u7528\u6237\u7684\u5bf9\u8bdd\uff0c\u4e0d\u65ad\u5f15\u5bfc\u786e\u8ba4\u7528\u6237\u7684\u9700\u6c42\uff0c\u800c\u975e\u53ea\u4ece\u7b80\u5355\u7684\u4e00\u53e5\u8bdd<\/strong><\/p>\n<ul>\n<li><a href=\"https:\/\/yale-lily.github.io\/cosql\">https:\/\/yale-lily.github.io\/cosql<\/a><\/li>\n<\/ul>\n<p><strong><em>CoSQL<\/em><\/strong> is a corpus for building cross-domain <strong>Co<\/strong>nversational text-to-<strong>SQL<\/strong> systems. It is the dialogue version of the Spider and SParC tasks. CoSQL consists of 30k+ turns plus 10k+ annotated SQL queries, obtained from a Wizard-of-Oz collection of 3k dialogues querying 200 complex databases spanning 138 domains. Each dialogue simulates a real-world DB query scenario with a crowd worker as a user exploring the database and a SQL expert retrieving answers with SQL, clarifying ambiguous questions, or otherwise informing of unanswerable questions.<\/p>\n<p>\u4e0b\u56fe\u4e3a\u6570\u636e\u96c6\u793a\u4f8b\uff1a<\/p>\n<p><img src=\"https:\/\/lemon-guess.oss-cn-hangzhou.aliyuncs.com\/img\/640-20251111162850387.png\" alt=\"\u56fe\u7247\" \/><\/p>\n<h2>BIRD - 202305<\/h2>\n<ul>\n<li><a href=\"https:\/\/bird-bench.github.io\/\">https:\/\/bird-bench.github.io\/<\/a><\/li>\n<li><a href=\"https:\/\/github.com\/AlibabaResearch\/DAMO-ConvAI\/tree\/main\/bird\">https:\/\/github.com\/AlibabaResearch\/DAMO-ConvAI\/tree\/main\/bird<\/a><\/li>\n<\/ul>\n<p>2023\u5e745\u6708\uff0c\u9999\u6e2f\u5927\u5b66\u548c\u963f\u91cc\u5df4\u5df4\u63d0\u51fa\u4e86\u4e00\u4e2a\u5927\u89c4\u6a21\u8de8\u57df\u6570\u636e\u96c6BIRD\uff0c\u5176\u4e2d\u5305\u542b\u8d85\u8fc712751\u4e2a\u72ec\u7279\u7684\u95ee\u9898 SQL\u300195\u4e2a\u5927\u6570\u636e\u5e93\uff0c\u603b\u5927\u5c0f\u4e3a33.4GB\u3002\u5b83\u8fd8<strong>\u6db5\u76d6\u533a\u5757\u94fe\u3001\u66f2\u68cd\u7403\u3001\u533b\u7597\u4fdd\u5065\u548c\u6559\u80b2\u7b49\u8d85\u8fc737\u4e2a\u4e13\u4e1a\u9886\u57df<\/strong>\u3002<\/p>\n<p>BIRD-SQL is the first cross-domain large-scale benchmark specifically designed to bridge the gap between academic research and real-world applications in the field of text-to-SQL parsing. While models such as Codex and ChatGPT have demonstrated remarkable performance, existing benchmarks such as Spider and WikiSQL concentrate primarily on database schema, leaving database contents largely unexplored. Realizing this limitation, we set out to create a comprehensive benchmark that delves deeper into database values, ultimately unveiling new challenges and opportunities for developments in the text-to-SQL domain.<\/p>\n<p>BIRD-SQL is distinguished by its large dataset, which includes <strong>12,751<\/strong> text-to-SQL pairs, <strong>95<\/strong> databases encompassing <strong>37<\/strong> professional domains, and a total size of <strong>33.4 GB<\/strong>. By highlighting database values, BIRD-SQL draws attention to new challenges, such as external knowledge, dirty data, and SQL efficiency in vast databases. In order to generate accurate SQL queries, models must not only conduct semantic parsing but also comprehend database values.<\/p>\n<p>Valid Efficiency Score (VES) Evaluation (time-mainly)<\/p>\n<p>\u8be5\u57fa\u51c6\u6d4b\u8bd5\u8fd8\u5f15\u5165\u4e86\u4e00\u4e2a\u65b0\u7684\u5ea6\u91cf\u6807\u51c6\u2014\u2014\u6709\u6548\u6027\u8bc4\u5206\uff08Valid Efficiency Score\uff0cVES\uff09\uff0c\u4ee5\u8bc4\u4f30\u751f\u6210\u7684SQL\u8bed\u53e5\u7684\u6267\u884c\u6548\u7387\u3002\u8fd9\u662f\u9996\u4e2a\u5c06\u6548\u7387\u7eb3\u5165\u8003\u8651\u7684\u6587\u672c\u5230SQL\u57fa\u51c6\u6d4b\u8bd5\uff0c\u4e3a\u5728\u5927\u89c4\u6a21\u548c\u5608\u6742\u6570\u636e\u5e93\u80cc\u666f\u4e0b\u5b9e\u73b0\u66f4\u9ad8\u6548\u7684\u67e5\u8be2\u65b9\u6cd5\u63d0\u4f9b\u63a8\u52a8\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u80cc\u666f \ud83c\udfaf\u81ea\u7136\u8bed\u8a00\u4e0e\u6570\u636e\u5e93\u7684\u6c9f\u901a\u56f0\u5883 \u5728\u5f53\u4eca\u6570\u5b57\u5316\u65f6\u4ee3\uff0c\u6570\u636e\u5e93\u4f5c\u4e3a\u6570\u636e\u5b58\u50a8\u548c\u7ba1\u7406\u7684\u6838\u5fc3\u5de5\u5177\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u5404\u4e2a\u9886\u57df\u3002\u65e0\u8bba\u662f\u4f01\u4e1a\u7684\u6570\u636e\u7ba1\u7406\u3001\u79d1\u7814   \u2026 &#8230;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0},"categories":[288],"tags":[],"_links":{"self":[{"href":"\/index.php?rest_route=\/wp\/v2\/posts\/9476"}],"collection":[{"href":"\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=9476"}],"version-history":[{"count":2,"href":"\/index.php?rest_route=\/wp\/v2\/posts\/9476\/revisions"}],"predecessor-version":[{"id":9483,"href":"\/index.php?rest_route=\/wp\/v2\/posts\/9476\/revisions\/9483"}],"wp:attachment":[{"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=9476"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=9476"},{"taxonomy":"post_tag","embeddable":true,"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=9476"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}