{"id":34176,"date":"2022-05-04T20:05:55","date_gmt":"2022-05-04T12:05:55","guid":{"rendered":"https:\/\/swarma.org\/?p=34176"},"modified":"2022-05-04T20:05:55","modified_gmt":"2022-05-04T12:05:55","slug":"%e2%80%8bnat-hum-behav-%e9%80%9f%e9%80%92%ef%bc%9a%e8%af%ad%e4%b9%89%e6%8a%95%e5%bd%b1%e4%bb%8e%e8%af%8d%e5%b5%8c%e5%85%a5%e4%b8%ad%e8%bf%98%e5%8e%9f%e4%b8%b0%e5%af%8c%e7%9a%84%e8%af%ad%e5%a2%83","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=34176","title":{"rendered":"\u200bNat. Hum. Behav. \u901f\u9012\uff1a\u8bed\u4e49\u6295\u5f71\u4ece\u8bcd\u5d4c\u5165\u4e2d\u8fd8\u539f\u4e30\u5bcc\u7684\u8bed\u5883\u77e5\u8bc6"},"content":{"rendered":"
\n
\n
\n
\n
\n

<\/p>\n<\/section>\n

\n
\n
\n
\n
\n
\n

<\/section>\n<\/section>\n
\n
\n

\u6458\u8981<\/strong><\/p>\n<\/section>\n<\/section>\n

\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n

\u5173\u4e8e\u8bcd\u4e49\u7684\u77e5\u8bc6\u5982\u4f55\u5728\u5fc3\u7406\u8bcd\u5178\uff08mental lexicon\uff09\u4e2d\u8868\u793a\uff1f\u5f53\u524d\u7684\u8ba1\u7b97\u6a21\u578b\u4ece\u8bcd\u6c47\u5171\u73b0\u6a21\u5f0f\u4e2d\u63a8\u65ad\u8bcd\u4e49\uff0c\u5b66\u4e60\u5982\u4f55\u5c06\u5355\u8bcd\u8868\u793a\u4e3a\u591a\u7ef4\u7a7a\u95f4\u4e2d\u7684\u5411\u91cf\uff0c\u4f7f\u5f97\u5728\u66f4\u76f8\u4f3c\u8bed\u5883\u4e2d\u4f7f\u7528\uff08\u5373\u8bed\u4e49\u76f8\u5173\u6027\u66f4\u9ad8\uff09\u7684\u5355\u8bcd\u4e4b\u5411\u91cf\u66f4\u9760\u8fd1\u3002\u7136\u800c\uff0c\u8bcd\u95f4\u63a5\u8fd1\u5ea6\u4ec5\u6355\u83b7\u6574\u4f53\u76f8\u5173\u6027\uff0c\u4eba\u7c7b\u5224\u65ad\u5374\u9ad8\u5ea6\u4f9d\u8d56\u4e8e\u8bed\u5883\u3002\u4f8b\u5982\uff0c\u6d77\u8c5a\u548c\u77ed\u543b\u9cc4\u4f53\u578b\u76f8\u4f3c\uff0c\u4f46\u5371\u9669\u6027\u4e0d\u540c\u3002\u8fd1\u65e5\u53d1\u8868\u5728 Nature Human Behaviour<\/em> \u4e0a\u7684\u6700\u65b0\u7814\u7a76\uff0c\u4f7f\u7528\u4e00\u79cd\u9886\u57df\u901a\u7528\u65b9\u6cd5\u4ece\u8bcd\u5d4c\u5165\uff08word embeddings\uff0c\u5373\u8bcd\u5411\u91cf\uff09\u4e2d\u63d0\u53d6\u4e0e\u8bed\u5883\u76f8\u5173\u7684\u5173\u7cfb\uff1a\u5c06\u8bcd\u5411\u91cf\u201c\u8bed\u4e49\u6295\u5f71\u201d\u5230\u8868\u793a\u8bf8\u5982\u5927\u5c0f\uff08\u8fde\u63a5\u8bcd\u201c\u5c0f\u201d\u548c\u201c\u5927\u201d\u7684\u7ebf\uff09\u6216\u5371\u9669\uff08\u201c\u5b89\u5168\u201d\u5230\u201c\u5371\u9669\u201d\uff09\u7b49\u7279\u5f81\u7684\u7ebf\u4e0a\uff0c\u7c7b\u4f3c\u4e8e\u201c\u5fc3\u7406\u91cf\u8868\u201d\u3002\u8be5\u65b9\u6cd5\u8fd8\u539f\u4e86\u4eba\u7c7b\u5bf9\u5404\u79cd\u5bf9\u8c61\u7c7b\u522b\u548c\u5c5e\u6027\u7684\u5224\u65ad\u3002\u56e0\u6b64\uff0c\u8bcd\u5d4c\u5165\u7684\u51e0\u4f55\u5f62\u72b6\u660e\u786e\u5730\u4ee3\u8868\u4e86\u4e30\u5bcc\u7684\u8bed\u5883\u76f8\u5173\u7684\u4e16\u754c\u77e5\u8bc6\u3002<\/span><\/strong><\/span><\/p>\n

 <\/p>\n

\u7814\u7a76\u9886<\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong>\u57df\uff1a<\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong>\u8ba1\u7b97\u8bed\u8a00\u5b66\uff0c\u5fc3\u7406\u8bed\u8a00\u5b66<\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n
\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
<\/strong><\/section>\n<\/section>\n
\n

\u5f90\u6069\u5ce4<\/span> <\/span>| \u4f5c\u8005<\/strong><\/span><\/p>\n

\u9093\u4e00\u96ea<\/span> | \u7f16\u8f91<\/strong><\/span><\/strong><\/span><\/p>\n


<\/p>\n<\/section>\n


<\/p>\n

<\/section>\n
\n
\n
\n
\u8bba\u6587\u9898\u76ee\uff1aSemantic projection recovers rich human knowledge of multiple object features from word embeddings<\/span><\/section>\n
\u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/www.nature.com\/articles\/s41562-022-01316-8<\/span><\/section>\n<\/section>\n<\/section>\n<\/blockquote>\n


<\/p>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

1. \u8bed\u4e49\u6295\u5f71\u4f5c\u4e3a\u7279\u5f81\u6807\u5c3a<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n

\n
\n

<\/section>\n<\/section>\n<\/h3>\n
\u4e00\u7ec4\u8bcd\u5411\u91cf\u80fd\u591f\u8574\u542b\u4e30\u5bcc\u7684\u8bed\u5883\u4fe1\u606f\u3002\u5982\u679c\u4e24\u4e2a\u8bcd\u8bed\u5728\u7a7a\u95f4\u4e0a\u7684\u4f4d\u7f6e\u66f4\u52a0\u63a5\u8fd1\uff0c\u5219\u80fd\u591f\u8bf4\u660e\u4ed6\u4eec\u5728\u6574\u4e2a\u8bed\u6599\u4e2d\u66f4\u503e\u5411\u4e8e\u51fa\u73b0\u5728\u7c7b\u4f3c\u7684\u8bed\u5883\u5f53\u4e2d\u3002\u4f46\u662f\u6709\u4e00\u4e2a\u95ee\u9898\uff0c\u5f53\u6211\u4eec\u5728\u8bf4\u4e24\u4e2a\u8bcd\u8bed\u76f8\u4f3c\u7684\u65f6\u5019\uff0c\u6211\u4eec\u5b9e\u9645\u4e0a\u5728\u8bf4\u4ec0\u4e48\uff1f\u7531\u4e8e\u73b0\u5b9e\u4e16\u754c\u4e8b\u7269\u7684\u4e30\u5bcc\u6027\uff0c\u4e00\u4e2a\u5bf9\u8c61\u96be\u514d\u4f1a\u6709\u591a\u4e2a\u7ef4\u5ea6\u7684\u4e0d\u540c\u7279\u5f81\u3002\u4f8b\u5982\uff0c\u6d77\u8c5a\u4e0e\u77ed\u543b\u9cc4\u5728\u4f53\u578b\u4e0a\u76f8\u8fd1\uff0c\u4f46\u662f\u5371\u9669\u6027\u5374\u5927\u6709\u4e0d\u540c\u3002\u672c\u6587\u4f5c\u8005\u521b\u65b0\u6027\u5730\u8bbe\u8ba1\u4e86\u4e00\u4e2a\u79f0\u4e4b\u4e3a\u8bed\u4e49\u6295\u5f71<\/strong><\/span>\uff08semantic projection\uff09<\/span>\u7684\u65b9\u6cd5\uff0c\u4ece\u8bcd\u5411\u91cf\u4e2d\u6316\u6398\u591a\u4e2a\u7ef4\u5ea6\u7684\u4fe1\u606f\u3002<\/span><\/section>\n

<\/span><\/section>\n
\u56fe1\u4e2d\u7a7f\u8fc7\u5c0f<\/span>\uff08SMALL<\/span>\uff09\u548c\u5927<\/span>\uff08LARGE\uff09<\/span>\u4e24\u4e2a\u8bcd\u7684\u7ea2\u7ebf\u4ee3\u8868\u4e86\u4e00\u628a\u6807\u5c3a\u3002\u84dd\u70b9\u6240\u4ee3\u8868\u7684\u52a8\u7269\u8d8a\u9760\u8fd1SMALL\u4e00\u4fa7\uff0c\u5219\u66f4\u53ef\u80fd\u5177\u6709\u5c0f\u7684\u4f53\u578b\uff1b\u66f4\u9760\u8fd1LARGE\u4e00\u4fa7\uff0c\u5219\u53cd\u4e4b\u3002\u53ef\u4ee5\u770b\u5230SMALL\u4e00\u4fa7\u6709\u6bd4\u8f83\u591a\u7684\u8bf8\u5982\u8681<\/span>\uff08Ant\uff09<\/span>\u3001\u9f20<\/span>\uff08Mouse\uff09<\/span>\u3001\u871c\u8702<\/span>\uff08Bee\uff09<\/span>\u7b49\u52a8\u7269\uff1b\u800cLARGE\u4e00\u4fa7\u5219\u6709\u9cb8\u9c7c<\/span>\uff08Whale\uff09<\/span>\u3001\u8c61<\/span>\uff08Elephant\uff09<\/span>\u7b49\u52a8\u7269\u2014\u2014\u8fd9\u8bf4\u660e\u8bed\u4e49\u6295\u5f71\u7684\u65b9\u6cd5\u80fd\u591f\u6bd4\u8f83\u597d\u5730\u6355\u6349\u5230\u4eba\u7c7b\u4e16\u754c\u77e5\u8bc6\u4e2d\u8fd9\u4e9b\u52a8\u7269\u7684\u5927\u5c0f\u5173\u7cfb\u3002<\/span><\/section>\n

<\/section>\n
<\/section>\n

\u56fe1. \u7528PCA\u964d\u7ef4\u81f33D\u7684\u8bcd\u5411\u91cf\uff08\u6e90\u4e8eGloVe\u65b9\u6cd5\u751f\u6210\u7684300D\u8bcd\u5411\u91cf\uff09\u6240\u5c55\u793a\u8bed\u4e49\u6295\u5f71\u65b9\u6cd5\uff0c\u4ee5\u52a8\u7269\u7c7b\u522b\u7684\u5c3a\u5bf8\u7279\u5f81\u4e3a\u4f8b\u3002\u7ea2\u70b9\u4e3a\u4e00\u5bf9\u7279\u5f81\u8bcd\uff08SMALL\u548cLARGE\uff09\uff0c\u76f8\u8fde\u6210\u7ea2\u7ebf\u5373\u4e3a\u7279\u5f81\u6807\u5c3a\uff1b\u84dd\u70b9\u4e3a\u4e0d\u540c\u7684\u52a8\u7269\uff0c\u5f15\u5230\u7ea2\u7ebf\u4e0a\u7684\u6de1\u84dd\u7ebf\u4e3a\u6b63\u4ea4\u6295\u5f71\u7ebf\u3002<\/span><\/p>\n


<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

2. \u8bed\u4e49\u6295\u5f71\u8fd8\u539f\u591a\u6837\u4e16\u754c\u77e5\u8bc6<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n

\n
\n

<\/section>\n<\/section>\n<\/h3>\n
\u8fdb\u4e00\u6b65\u7684\u7814\u7a76\u53d1\u73b0\uff0c\u8bed\u4e49\u6295\u5f71\u65b9\u6cd5\u80fd\u591f\u5728\u7279\u5f81\u6216\u8005\u7c7b\u522b\u53d1\u751f\u53d8\u52a8\u65f6\uff0c\u8f83\u597d\u9884\u6d4b\u4eba\u7c7b\u5173\u4e8e\u67d0\u4e9b\u7279\u5f81\u7684\u5224\u65ad\u3002\u89c1\u56fe2a\u7ea2\u70b9\u6807\u51fa\u7684\u6d77\u8c5a<\/span>\uff08Dolphin\uff09<\/span>\u4e0e\u864e<\/span>\uff08Tiger\uff09<\/span>\u8fd9\u4e24\u79cd\u52a8\u7269\uff0c\u76f4\u89c9\u544a\u8bc9\u6211\u4eec\uff1a\u864e\u8fdc\u8fdc\u6bd4\u6d77\u8c5a\u5371\u9669\uff0c\u4f46\u5b83\u4eec\u5927\u5c0f\u76f8\u4f3c\uff0c\u800c\u6d77\u8c5a\u56e0\u751f\u6d3b\u5728\u6c34\u4e2d\u800c\u6709\u66f4\u9ad8\u7684\u6e7f\u5ea6\u2014\u2014\u8fd9\u4e0e\u8bed\u4e49\u6295\u5f71\u7684\u9884\u6d4b\u76f8\u7b26\u3002\u56fe2b\u4e0e\u56fe2a\u7c7b\u4f3c\uff0c\u4f46\u662f\u53d1\u751f\u53d8\u5316\u7684\u662f\u7c7b\u522b\uff0c\u8fd9\u91cc\u4e0d\u518d\u8d58\u8ff0\u3002\u56fe\u4e2d\u5176\u4ed6\u6563\u70b9\u4e5f\u663e\u793a\uff0c\u4ece\u8bcd\u5411\u91cf\u4e2d\u8fd8\u539f\u51fa\u7684\u77e5\u8bc6\u4e0e\u4eba\u7c7b\u5224\u65ad\u662f\u6b63\u76f8\u5173\u7684\u3002<\/strong><\/span>
<\/section>\n

<\/span><\/section>\n
<\/section>\n

\u56fe2. \u8bed\u4e49\u6295\u5f71\u9884\u6d4b\u4eba\u7c7b\u5224\u65ad\u7684\u6848\u4f8b\uff0c\u6bcf\u4e2a\u7ec4\u5408\u5747\u670925\u4e2a\u53c2\u4e0e\u8005\u63d0\u4f9b\u4eba\u7c7b\u8bc4\u5206\u3002a \u56fa\u5b9a\u7c7b\u522b\uff08\u52a8\u7269\uff09\uff0c\u53d8\u52a8\u7279\u5f81\uff08\u5371\u9669\u6027\u3001\u5c3a\u5bf8\u3001\u6e7f\u5ea6\uff09\uff1bb \u53d8\u52a8\u7c7b\u522b\uff08\u52a8\u7269\u57ce\u5e02\uff0c\u6587\u7269\uff0c\u795e\u8bdd\u751f\u7269\uff09\uff0c\u56fa\u5b9a\u7279\u5f81\uff08\u5371\u9669\u6027\uff09\u3002<\/span><\/p>\n


<\/span><\/section>\n
\u7814\u7a76\u8005\u968f\u540e\u5728\u66f4\u5e7f\u6cdb\u7684\u7c7b\u522b\u4e0e\u7279\u5f81\u7ec4\u5408\u4e0a\u5c55\u793a\u4e86\u7ed3\u679c\u7684\u6709\u6548\u6027\u3002\u56fe3\u5c55\u793a\u4e86\u300c9\u79cd\u7c7b\u522b\u300dx\u300c17\u79cd\u7279\u5f81\u300d\u7ec4\u5408\u7684\u9884\u6d4b\u60c5\u51b5\u3002\u4ee5\u7b2c\u4e03\u5217\u4f4d\u7f6e<\/span>\uff08Location\uff0c\u6807\u5c3a\u4e3aindoors-outdoors\uff0c\u5373\u5ba4\u5185-\u6237\u5916\uff09<\/span>\u4e3a\u4f8b\uff0c\u8fd0\u52a8<\/span>\uff08Sports\uff09<\/span>\u4e0e\u4f4d\u7f6e\u663e\u8457\u76f8\u5173<\/span>\uff08\u5ba4\u5185\u3001\u6237\u5916\u8fd0\u52a8\uff09<\/span>\uff0c\u800c\u804c\u4e1a<\/span>\uff08Professions\uff09<\/span>\u4e0e\u4f4d\u7f6e\u7684\u5173\u8054\u76f8\u5bf9\u66f4\u5f31\uff0c\u8fd9\u5927\u4f53\u7b26\u5408\u4eba\u7c7b\u5173\u4e8e\u4f4d\u7f6e\u7684\u77e5\u8bc6\uff1b\u800c\u5bf9\u4e8e\u6700\u540e\u4e00\u884c\u5929\u6c14<\/span>\uff08Weather\uff09<\/span>\u6765\u8bf4\uff0c\u5371\u9669<\/span>\uff08Danger\uff09<\/span>\u4e0e\u5426\u548c\u6e29\u5ea6<\/span>\uff08Temperature\uff09<\/span>\u3001\u6e7f\u5ea6<\/span>\uff08Wetness\uff09<\/span>\u786e\u5b9e\u90fd\u662f\u4eba\u7c7b\u5173\u5fc3\u7684\uff0c\u5747\u6709\u663e\u8457\u5173\u8054\u3002\u8bfb\u8005\u53ef\u4ee5\u5c1d\u8bd5\u81ea\u884c\u9a8c\u8bc1\u5176\u5b83\u7684\u7ec4\u5408\uff0c\u5176\u4e2d\u5927\u90e8\u5206\u7684\u7ec4\u5408\u5747\u80fd\u8f83\u597d\u7b26\u5408\u4eba\u7c7b\u76f4\u89c9\u3002<\/span><\/section>\n

<\/span><\/section>\n
<\/section>\n

\u56fe3. \u8bed\u4e49\u6295\u5f71\u9884\u6d4b\u591a\u79cd\u7c7b\u522b\u4e0e\u7279\u5f81\u4e0a\u7684\u4eba\u7c7b\u8bc4\u5206\u6548\u679c\u3002\u56fe\u4e2d\u989c\u8272\u5bf9\u5e94Pearson\u76f8\u5173\u6027\uff08\u7070\u8272\u662f\u672a\u505a\u5b9e\u9a8c\u7684\u7ec4\u5408\uff09\uff0c\u884c\u3001\u5217\u5206\u522b\u4ee3\u8868\u7c7b\u522b\u4e0e\u7279\u5f81\uff0c\u884c\u3001\u5217\u7684\u4e2d\u4f4d\u6570\u5206\u522b\u5728\u6700\u540e\u4e00\u5217\u4e0e\u6700\u540e\u4e00\u884c\u6807\u51fa\u3002\u7edf\u8ba1\u663e\u8457\u7684\u62df\u5408\u7ed3\u679c\u683c\u5b50\u88ab\u7c97\u6846\u56f4\u7ed5\u3002<\/span><\/p>\n


<\/span><\/section>\n

<\/span><\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

3. \u8fd8\u539f\u4e30\u5bcc\u8bcd\u5411\u91cf\u4fe1\u606f\u7684\u65b0\u65b9\u6cd5<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n

\n
\n

<\/section>\n<\/section>\n<\/h3>\n
\u8bed\u4e49\u6295\u5f71\u65b9\u6cd5\u7684\u4f7f\u7528\u8bc1\u660e\uff0c\u8bcd\u5411\u91cf\u4e2d\u5b9e\u9645\u4e0a\u786e\u5b9e\u660e\u786e\u5730\u7f16\u7801\u4e86\u6bd4\u6211\u4eec\u539f\u5148\u60f3\u8c61\u66f4\u52a0\u4e30\u5bcc\u7684\u8bed\u5883\u77e5\u8bc6<\/strong>\uff0c\u76ee\u524d\u6211\u4eec\u4ece\u5176\u4e2d\u6316\u6398\u5230\u7684\u4fe1\u606f\u53ef\u80fd\u53ea\u89e6\u53ca\u5176\u8868\u9762\uff0c\u56e0\u6b64\u540e\u7eed\u53ef\u4ee5\u57fa\u4e8e\u6b64\u601d\u8def\u5f00\u53d1\u66f4\u4f18\u7684\u6a21\u578b\u3001\u6316\u6398\u66f4\u591a\u7684\u4fe1\u606f\u3002<\/span>
<\/section>\n

<\/span><\/section>\n
\u53e6\u5916\uff0c\u9664\u4e86\u6709\u4f18\u5f02\u7684\u8fd8\u539f\u6548\u679c\uff0c\u4e0e\u5148\u524d\u4ece\u81ea\u7136\u8bed\u8a00\u4f7f\u7528\u6a21\u5f0f\u4e2d\u63d0\u53d6\u8bed\u4e49\u77e5\u8bc6\u7684\u5c1d\u8bd5\u76f8\u6bd4\uff0c\u8fd9\u79cd\u65b9\u6cd5\u9700\u8981\u7684\u4eba\u5de5\u76d1\u7763\u548c\/\u6216\u8bed\u6599\u5e93\u6ce8\u91ca\u8981\u5c11\u5f97\u591a\uff0c\u56e0\u6b64\u4ece\u5e94\u7528\u4e0a\u6765\u8bf4\uff0c\u8be5\u65b9\u6cd5\u6709\u671b\u4e3a\u5404\u79cd\u8bed\u6599\u7075\u6d3b\u5730\u6062\u590d\u5404\u79cd\u8bed\u4e49\u7279\u5f81\uff0c\u751a\u81f3\u80fd\u591f\u628a\u4e0d\u540c\u5b66\u79d1\u7684\u8bed\u6599\u8f6c\u6362\u4e3a\u77e5\u8bc6\u542b\u91cf\u4e30\u5bcc\u3001\u66f4\u52a0\u6613\u7528\u7684\u6570\u636e\u3002<\/span><\/section>\n


<\/p>\n

\n
\n
\n

\u8bba\u6587 Abstract<\/strong><\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n

How is knowledge about word meaning represented in the mental lexicon? Current computational models infer word meanings from lexical co-occurrence patterns. They learn to represent words as vectors in a multidimensional space, wherein words that are used in more similar linguistic contexts\u2014that is, are more semantically related\u2014are located closer together. However, whereas inter-word proximity captures only overall relatedness, human judgements are highly context dependent. For example, dolphins and alligators are similar in size but differ in dangerousness. Here, we use a domain-general method to extract context-dependent relationships from word embeddings: \u2018semantic projection\u2019 of word-vectors onto lines that represent features such as size (the line connecting the words \u2018small\u2019 and \u2018big\u2019) or danger (\u2018safe\u2019 to \u2018dangerous\u2019), analogous to \u2018mental scales\u2019. This method recovers human judgements across various object categories and properties. Thus, the geometry of word embeddings explicitly represents a wealth of context-dependent world knowledge.<\/span><\/section>\n

<\/section>\n
 <\/span><\/section>\n
\n
\n
\n
\n

\u590d\u6742\u79d1\u5b66\u6700\u65b0\u8bba\u6587<\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n


<\/p>\n<\/section>\n

\u96c6\u667a\u6591\u56fe\u9876\u520a\u8bba\u6587\u901f\u9012\u680f\u76ee\u4e0a\u7ebf\u4ee5\u6765\uff0c\u6301\u7eed\u6536\u5f55\u6765\u81eaNature\u3001Science<\/span>\u7b49\u9876\u520a\u7684\u6700\u65b0\u8bba\u6587\uff0c\u8ffd\u8e2a\u590d\u6742\u7cfb\u7edf\u3001\u7f51\u7edc\u79d1\u5b66\u3001\u8ba1\u7b97\u793e\u4f1a\u79d1\u5b66\u7b49\u9886\u57df\u7684\u524d\u6cbf\u8fdb\u5c55\u3002\u73b0\u5728\u6b63\u5f0f\u63a8\u51fa\u8ba2\u9605\u529f\u80fd\uff0c\u6bcf\u5468\u901a\u8fc7\u5fae\u4fe1\u670d\u52a1\u53f7\u300c\u96c6\u667a\u6591\u56fe\u300d\u63a8\u9001\u8bba\u6587\u4fe1\u606f\u3002\u626b\u63cf\u4e0b\u65b9\u4e8c\u7ef4\u7801\u5373\u53ef\u4e00\u952e\u8ba2\u9605\uff1a<\/span>
<\/p>\n

<\/p>\n

\n


<\/p>\n<\/section>\n


<\/span><\/p>\n

\n

\u63a8\u8350\u9605\u8bfb<\/span><\/strong><\/strong><\/p>\n