{"id":23901,"date":"2025-01-03T12:11:20","date_gmt":"2025-01-03T04:11:20","guid":{"rendered":"https:\/\/aif.amtbbs.org\/?p=23901"},"modified":"2025-01-03T12:11:20","modified_gmt":"2025-01-03T04:11:20","slug":"%e5%85%a8%e6%96%b0%e6%a8%a1%e5%9e%8brobovlms%e8%a7%a3%e9%94%81vla%e6%97%a0%e9%99%90%e5%8f%af%e8%83%bd%ef%bc%8c%e7%9c%9f%e5%ae%9e%e6%9c%ba%e5%99%a8%e4%ba%ba%e5%ae%9e%e9%aa%8c%e4%ba%a4%e5%87%ba%e6%bb%a1","status":"publish","type":"post","link":"https:\/\/aif.amtbbs.org\/index.php\/2025\/01\/03\/23901\/","title":{"rendered":"\u5168\u65b0\u6a21\u578bRoboVLMs\u89e3\u9501VLA\u65e0\u9650\u53ef\u80fd\uff0c\u771f\u5b9e\u673a\u5668\u4eba\u5b9e\u9a8c\u4ea4\u51fa\u6ee1\u5206\u7b54\u5377"},"content":{"rendered":"<div><img data-dominant-color=\"3584b8\" data-has-transparency=\"false\" style=\"--dominant-color: #3584b8;\" loading=\"lazy\" decoding=\"async\" class=\"not-transparent alignnone size-full wp-image-23903\" src=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/658b59286a6847be7b45756fbbf542d440b6f0-300x167-1.jpg\" width=\"300\" height=\"167\" alt=\"\" srcset=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/658b59286a6847be7b45756fbbf542d440b6f0-300x167-1.jpg 300w, https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/658b59286a6847be7b45756fbbf542d440b6f0-300x167-1-150x84.jpg 150w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/div>\n<div><\/div>\n<div class=\"article-desc\">\u867d\u7136 VLAs \u5728\u5404\u79cd\u4efb\u52a1\u548c\u573a\u666f\u4e2d\u8868\u73b0\u62a2\u773c\uff0c\u4f46\u5927\u5bb6\u5728\u6a21\u578b\u8bbe\u8ba1\u4e0a\u5374\u8d70\u4e86\u5f88\u591a\u4e0d\u540c\u7684\u8def\uff0c\u6bd4\u5982\u7528\u4ec0\u4e48\u67b6\u6784\u3001\u600e\u4e48\u9009\u6570\u636e\u3001\u600e\u4e48\u8c03\u8bad\u7ec3\u7b56\u7565\u7b49\u7b49\uff0c\u8fd9\u5bfc\u81f4\u9886\u57df\u5185\u5bf9 \u201c\u600e\u4e48\u505a\u597d\u4e00\u4e2a VLA\u201d \u8fd8\u6ca1\u6709\u7edf\u4e00\u7684\u7b54\u6848\u3002<\/div>\n<div id=\"postspictures\" class=\"article-content\">\n<div id=\"container\" class=\"container am-engine\" data-v-1d7a5742=\"\" data-element=\"root\">\n<p><strong>\u672c\u6587\u4f5c\u8005\u6765\u81ea\u6e05\u534e\u5927\u5b66\u3001\u5b57\u8282\u8df3\u52a8\u3001\u4e2d\u79d1\u9662\u81ea\u52a8\u5316\u6240\u3001\u4e0a\u6d77\u4ea4\u901a\u5927\u5b66\u548c\u65b0\u52a0\u5761\u56fd\u7acb\u5927\u5b66\u3002\u4f5c\u8005\u5217\u8868\uff1a\u674e\u5174\u822a\u3001\u674e\u6c9b\u8a00\u3001\u5218\u660e\u6853\u3001\u738b\u680b\u3001\u5218\u6d4e\u6995\u3001\u5eb7\u70b3\u6613\u3001\u9a6c\u9a81\u3001\u5b54\u6d9b\u3001\u5f20\u7ff0\u535a\u548c\u5218\u534e\u5e73\u3002\u7b2c\u4e00\u4f5c\u8005\u674e\u5174\u822a\u662f\u6e05\u534e\u5927\u5b66\u8ba1\u7b97\u673a\u7cfb\u535a\u58eb\u751f\u3002\u901a\u8baf\u4f5c\u8005\u662f\u5b57\u8282\u8df3\u52a8\u673a\u5668\u4eba\u7814\u7a76\u5458\u5b54\u6d9b\uff0c\u65b0\u52a0\u5761\u56fd\u7acb\u5927\u5b66\u535a\u58eb\u540e\u5f20\u7ff0\u535a\u548c\u6e05\u534e\u5927\u5b66\u8ba1\u7b97\u673a\u7cfb\u6559\u6388\u5218\u534e\u5e73\u3002<\/strong><\/p>\n<p>\u8fd1\u5e74\u6765\uff0c<strong>\u89c6\u89c9\u8bed\u8a00\u57fa\u7840\u6a21\u578b<\/strong>\uff08Vision Language Models, VLMs\uff09\u5927\u653e\u5f02\u5f69\uff0c\u5728\u591a\u6a21\u6001\u7406\u89e3\u548c\u63a8\u7406\u4e0a\u5c55\u73b0\u51fa\u4e86\u8d85\u5f3a\u80fd\u529b\u3002\u73b0\u5728\uff0c\u66f4\u52a0\u9177\u70ab\u7684<strong>\u89c6\u89c9\u8bed\u8a00\u52a8\u4f5c\u6a21\u578b<\/strong>\uff08Vision-Language-Action Models, VLAs\uff09\u6765\u4e86\uff01\u901a\u8fc7\u4e3a VLMs \u52a0\u4e0a\u52a8\u4f5c\u9884\u6d4b\u6a21\u5757\uff0cVLAs \u4e0d\u4ec5\u80fd \u201c\u770b\u201d \u61c2\u548c \u201c\u8bf4\u201d \u6e05\uff0c\u8fd8\u80fd \u201c\u52a8\u201d \u8d77\u6765\uff0c\u4e3a\u673a\u5668\u4eba\u9886\u57df\u5f00\u542f\u4e86\u65b0\u73a9\u6cd5\uff01<\/p>\n<p><img data-dominant-color=\"eff0f1\" data-has-transparency=\"true\" style=\"--dominant-color: #eff0f1;\" loading=\"lazy\" decoding=\"async\" class=\"has-transparency alignnone size-full wp-image-23904\" src=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/e34a5943623efbb01ef9739559fe039c83f703.png\" width=\"658\" height=\"371\" srcset=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/e34a5943623efbb01ef9739559fe039c83f703.png 658w, https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/e34a5943623efbb01ef9739559fe039c83f703-300x169.png 300w, https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/e34a5943623efbb01ef9739559fe039c83f703-150x85.png 150w\" sizes=\"auto, (max-width: 658px) 100vw, 658px\" \/><\/p>\n<p>\u867d\u7136 VLAs \u5728\u5404\u79cd\u4efb\u52a1\u548c\u573a\u666f\u4e2d\u8868\u73b0\u62a2\u773c\uff0c\u4f46\u5927\u5bb6\u5728\u6a21\u578b\u8bbe\u8ba1\u4e0a\u5374\u8d70\u4e86\u5f88\u591a\u4e0d\u540c\u7684\u8def\uff0c\u6bd4\u5982\u7528\u4ec0\u4e48\u67b6\u6784\u3001\u600e\u4e48\u9009\u6570\u636e\u3001\u600e\u4e48\u8c03\u8bad\u7ec3\u7b56\u7565\u7b49\u7b49\uff0c\u8fd9\u5bfc\u81f4\u9886\u57df\u5185\u5bf9 \u201c\u600e\u4e48\u505a\u597d\u4e00\u4e2a VLA\u201d \u8fd8\u6ca1\u6709\u7edf\u4e00\u7684\u7b54\u6848\u3002\u4e3a\u4e86\u7406\u6e05\u8fd9\u4e9b\u95ee\u9898\uff0c\u6211\u4eec\u901a\u8fc7\u4e00\u7cfb\u5217\u7684\u5b9e\u9a8c\uff0c\u63d0\u51fa\u4e86\u4e00\u4e2a\u5168\u65b0\u6a21\u578b \u2014\u2014<strong>RoboVLMs<\/strong>\u3002<\/p>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s3.51cto.com\/oss\/202501\/02\/47437475124d6e14f3a2477586bef64d9ffdb2.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<ul data-id=\"u738a58b-IiE5Y3lA\">\n<li data-id=\"ld70c578-91Wma1gQ\">\u8bba\u6587\u6807\u9898\uff1aTowards Generalist Robot Policies: What Matters in Building Vision-Language-Action Models<\/li>\n<li data-id=\"ld70c578-lLSbO1gU\">\u8bba\u6587\u5730\u5740\uff1ahttps:\/\/arxiv.org\/pdf\/2412.14058<\/li>\n<\/ul>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s9.51cto.com\/oss\/202501\/02\/c68af22350a0caad6d2982e0339f06a9c50e61.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<p>\u8fd9\u4e2a\u6a21\u578b\u8d85\u7ea7\u7b80\u5355\uff0c\u4f46\u6027\u80fd\u5374\u76f8\u5f53\u786c\u6838\uff01\u5b83\u4e0d\u4ec5\u5728\u4e09\u4e2a\u6a21\u62df\u4efb\u52a1\u4e2d\u53d6\u5f97\u4e86\u9ad8\u5206\uff0c\u8fd8\u5728\u771f\u5b9e\u673a\u5668\u4eba\u5b9e\u9a8c\u4e2d\u4ea4\u51fa\u4e86\u6ee1\u5206\u7b54\u5377\u3002\u8fd9\u7bc7\u6587\u7ae0\u5c31\u662f\u8981\u5e26\u4f60\u4e00\u8d77\u770b\u770b\uff0c\u6211\u4eec\u662f\u5982\u4f55\u7528 RoboVLMs \u89e3\u9501 VLA \u7684\u65e0\u9650\u53ef\u80fd\uff01<\/p>\n<h4>\u56db\u5927\u7075\u9b42\u62f7\u95ee\uff1aRoboVLMs \u662f\u600e\u4e48\u70bc\u6210\u7684\uff1f<\/h4>\n<p>\u6211\u4eec\u56f4\u7ed5\u56db\u4e2a\u5173\u952e\u95ee\u9898\uff0c\u5bf9 VLA \u7684\u8bbe\u8ba1\u5c55\u5f00\u4e86\u6df1\u5ea6\u63a2\u7d22\uff0c\u4e0b\u9762\u5c31\u5e26\u4f60\u770b\u770b\u7b54\u6848\uff01<\/p>\n<p><strong>1. \u4e3a\u4ec0\u4e48\u8981\u7528 VLA \u6a21\u578b\uff1f<\/strong><\/p>\n<p>\u7b80\u5355\u8bf4\uff0c\u901a\u8fc7\u5b9e\u9a8c\uff0c\u6211\u4eec\u53d1\u73b0\u8bbe\u8ba1\u5408\u7406\u7684 VLA \u4e0d\u4ec5\u80fd\u8f7b\u677e\u641e\u5b9a\u5e38\u89c1\u7684\u64cd\u4f5c\u4efb\u52a1\uff0c\u8fd8\u80fd\u5728\u964c\u751f\u573a\u666f\u4e2d\u7a33\u7a33\u53d1\u6325\u3002<\/p>\n<p><strong>\u4eff\u771f\u4efb\u52a1\u4e2d\u62ff\u4e0b\u9876\u5c16\u6210\u7ee9<\/strong><\/p>\n<p>\u5728 CALVIN \u548c SimplerEnv \u73af\u5883\u91cc\uff0cRoboVLMs \u53d6\u5f97\u4e86\u538b\u5012\u6027\u7684\u80dc\u5229\uff1a<\/p>\n<ul data-id=\"u738a58b-d7aNJFNM\">\n<li data-id=\"ld70c578-nn2VZ1Vi\">\u4efb\u52a1\u6210\u529f\u7387\uff1a\u8868\u73b0\u7a33\u5b9a\u4e14\u8d85\u8d8a\u4e3b\u6d41\u6a21\u578b\u3002<\/li>\n<li data-id=\"ld70c578-h42PmMad\">\u6cdb\u5316\u80fd\u529b\uff1a\u5373\u4f7f\u5728\u964c\u751f\u573a\u666f\u4e2d\uff0c\u8868\u73b0\u4f9d\u7136\u6297\u6253\uff01<\/li>\n<\/ul>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s5.51cto.com\/oss\/202501\/02\/8580b3a5546d0387cff065d9274e4fd5985efd.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<p>\u56fe 1 SimplerEnv \u4eff\u771f\u73af\u5883\u4e2d\u7684\u8bc4\u6d4b\u7ed3\u679c<\/p>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s3.51cto.com\/oss\/202501\/02\/d7af51826a1be2c3a43705083057da0b5b4256.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<p>\u56fe 2 \u9488\u5bf9\u89c6\u89c9\u8bed\u8a00\u9884\u8bad\u7ec3\u7684\u6d88\u878d\u5b9e\u9a8c\u7ed3\u679c<\/p>\n<p><img data-dominant-color=\"6b513b\" data-has-transparency=\"false\" style=\"--dominant-color: #6b513b;\" loading=\"lazy\" decoding=\"async\" class=\"not-transparent alignnone size-full wp-image-23905\" src=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/1714b79607b422606d6010f0053a81f5419596.gif\" width=\"1080\" height=\"216\" alt=\"\u56fe\u7247\" title=\"\u56fe\u7247\" \/><\/p>\n<p><img data-dominant-color=\"8d6b5b\" data-has-transparency=\"false\" style=\"--dominant-color: #8d6b5b;\" loading=\"lazy\" decoding=\"async\" class=\"not-transparent alignnone size-full wp-image-23906\" src=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/61506152529cad95a0a5036cd9fe8300163b41.gif\" width=\"1077\" height=\"202\" alt=\"\u56fe\u7247\" title=\"\u56fe\u7247\" \/><\/p>\n<h4>\u771f\u5b9e\u673a\u5668\u4eba\u5b9e\u9a8c\u4e5f\u4e0d\u8f93<\/h4>\n<p>\u5728\u771f\u5b9e\u73af\u5883\u4e2d\uff0cRoboVLMs \u9762\u5bf9\u66f4\u590d\u6742\u7684\u6311\u6218\uff0c\u4ecd\u7136\u6bd4\u5176\u4ed6\u6a21\u578b\u8868\u73b0\u66f4\u597d\u3002\u6bd4\u5982\uff0c\u5728\u679c\u852c\u5206\u7c7b\u4efb\u52a1\u4e2d\uff0c\u5b83\u4e0d\u4ec5\u80fd\u7cbe\u51c6\u8bc6\u522b\uff0c\u8fd8\u80fd\u5e94\u5bf9\u5e72\u6270\u73af\u5883\uff0c\u7a33\u7a33\u5b8c\u6210\u5206\u7c7b\u64cd\u4f5c\u3002\u65e0\u8bba\u662f\u5df2\u77e5\u573a\u666f\u8fd8\u662f\u65b0\u4efb\u52a1\uff0c\u5b83\u90fd\u80fd\u8f7b\u677e\u62ff\u4e0b\u3002<\/p>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s3.51cto.com\/oss\/202501\/02\/84429fb162f41a8efdb8841fe4f6f2bd151aff.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<p>\u56fe 3 \u771f\u5b9e\u73af\u5883\u4e0b\u7684\u8bc4\u6d4b\u7ed3\u679c<\/p>\n<p>\u5bf9\u4e8e\u672a\u89c1\u8fc7\u7684\u6280\u80fd\u63cf\u8ff0\u3001\u80cc\u666f\u3001\u5e72\u6270\u7269\u4f53\u548c\u76ee\u6807\u7269\u4f53\uff0cRoboVLMs \u5747\u80fd\u5f88\u597d\u7684\u5b8c\u6210\u4efb\u52a1\u3002<\/p>\n<p><img data-dominant-color=\"8f8a85\" data-has-transparency=\"false\" style=\"--dominant-color: #8f8a85;\" loading=\"lazy\" decoding=\"async\" class=\"not-transparent alignnone size-full wp-image-23907\" src=\"https:\/\/aiforumimage.oss-cn-shanghai.aliyuncs.com\/wp-content\/uploads\/2025\/01\/7756c4002309d012e0c920591ca5bdea24e163.gif\" width=\"640\" height=\"364\" alt=\"\u56fe\u7247\" title=\"\u56fe\u7247\" \/><\/p>\n<p><strong>2. \u600e\u4e48\u8bbe\u8ba1\u4e00\u4e2a\u9760\u8c31\u7684 VLA \u67b6\u6784\uff1f<\/strong><\/p>\n<p>\u8fd9\u91cc\u9762\u8bb2\u7a76\u53ef\u4e0d\u5c11\uff01\u6bd4\u5982\uff1a<\/p>\n<ul data-id=\"u738a58b-MO30NIIB\">\n<li data-id=\"ld70c578-mFRn6FIj\">\u52a8\u4f5c\u7a7a\u95f4\uff1a\u7528\u8fde\u7eed\u52a8\u4f5c\u7a7a\u95f4\u6bd4\u79bb\u6563\u7684\u597d\u5f88\u591a\u3002<\/li>\n<li data-id=\"ld70c578-gdNJan0l\">\u5386\u53f2\u4fe1\u606f\uff1a\u52a0\u591a\u6b65\u5386\u53f2\u4fe1\u606f\u540e\uff0c\u6a21\u578b\u7684\u64cd\u4f5c\u66f4\u7a33\u51c6\u72e0\u3002<\/li>\n<li data-id=\"ld70c578-aCbiqgSk\">\u5386\u53f2\u4fe1\u606f\u7ec4\u7ec7\u6a21\u5757\uff1a\u4e00\u4e2a\u4e13\u95e8\u7684\u6a21\u5757\u53ef\u4ee5\u8ba9\u6a21\u578b\u66f4\u61c2 \u201c\u4e0a\u4e0b\u6587\u201d\u3002<\/li>\n<\/ul>\n<p>\u7ecf\u8fc7\u4e00\u7cfb\u5217\u5b9e\u9a8c\uff0c\u6211\u4eec\u786e\u8ba4\u4e86\u8fd9\u4e9b\u8bbe\u8ba1\u9009\u62e9\u662f\u63d0\u5347\u6a21\u578b\u6027\u80fd\u548c\u6cdb\u5316\u80fd\u529b\u7684\u5173\u952e\u3002\u8fdb\u4e00\u6b65\u7684\u5b9e\u9a8c\u4e5f\u8868\u660e\uff0c<strong>\u6700\u4f18\u7684\u8bbe\u8ba1\u6765\u81ea\u4e8e\u57fa\u4e8e KosMos \u57fa\u5ea7\u6a21\u578b\u7684\u67b6\u6784\uff0c\u5e76\u4e14\u7ed3\u5408\u4e86\u4e13\u95e8\u7684\u5386\u53f2\u4fe1\u606f\u7ec4\u7ec7\u6a21\u5757<\/strong>\u3002\u8fd9\u6837\u7684\u8bbe\u8ba1\u5728 CALVIN \u4e2d\u5b9e\u73b0\u4e86\u51fa\u8272\u7684\u6cdb\u5316\u80fd\u529b\uff0c\u5728 zero-shot \u8bbe\u7f6e\u4e0b\u4ec5\u6709\u8f7b\u5fae\u7684\u6027\u80fd\u4e0b\u964d\uff0c\u800c\u5176\u4ed6\u8bbe\u8ba1\u5f62\u5f0f\u7684\u6a21\u578b\u5219\u51fa\u73b0\u4e86\u663e\u8457\u6389\u5206\u3002\u8fd9\u4e00\u7ed3\u8bba\u76f4\u63a5\u8bf4\u660e\uff0c<strong>\u67b6\u6784\u8bbe\u8ba1\u7684\u597d\u574f\u5bf9\u6a21\u578b\u7684\u6cdb\u5316\u80fd\u529b\u548c\u6548\u7387\u81f3\u5173\u91cd\u8981<\/strong>\u3002<\/p>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s6.51cto.com\/oss\/202501\/02\/07669a879183ec9d0ec97152cb1fb1f6f46d56.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<p><strong>3. \u9009\u4ec0\u4e48\u57fa\u5ea7\u6a21\u578b\u6700\u5408\u9002\uff1f<\/strong><\/p>\n<p>\u6211\u4eec\u5bf9\u6bd4\u4e86\u5f53\u524d\u4e3b\u6d41\u7684 8 \u79cd\u89c6\u89c9\u8bed\u8a00\u6a21\u578b\uff08VLM\uff09\uff0c\u7ed3\u679c\u53d1\u73b0\u00a0<strong>KosMos<\/strong>\u00a0\u548c\u00a0<strong>Paligemma<\/strong>\u00a0\u7684\u8868\u73b0\u9065\u9065\u9886\u5148\uff0c\u8f7b\u677e\u78be\u538b\u5176\u4ed6\u6a21\u578b\u3002\u65e0\u8bba\u662f\u4efb\u52a1\u5b8c\u6210\u7684\u7cbe\u786e\u5ea6\u8fd8\u662f\u6cdb\u5316\u80fd\u529b\uff0c\u5b83\u4eec\u90fd\u5c55\u73b0\u51fa\u4e86\u538b\u5012\u6027\u7684\u4f18\u52bf\u3002\u7a76\u5176\u539f\u56e0\uff0c\u4e3b\u8981\u5f97\u76ca\u4e8e\u5b83\u4eec\u7ecf\u8fc7\u4e86\u624e\u5b9e\u4e14\u5168\u9762\u7684\u89c6\u89c9\u8bed\u8a00\u9884\u8bad\u7ec3\uff0c\u4ece\u800c\u4e3a\u6a21\u578b\u63d0\u4f9b\u4e86\u5f3a\u5927\u7684\u5148\u9a8c\u77e5\u8bc6\u548c\u7406\u89e3\u80fd\u529b\u3002<\/p>\n<p>\u8fd9\u4e00\u53d1\u73b0\u8ba9\u6211\u4eec\u66f4\u52a0\u786e\u4fe1\uff1a<strong>\u9009\u5bf9\u57fa\u5ea7\u6a21\u578b\uff0c\u5c31\u662f\u8ba9 VLA \u6a21\u578b\u8d77\u98de\u7684\u5173\u952e\u4e00\u6b65<\/strong>\uff01\u60f3\u8981\u8ba9\u6a21\u578b\u5728\u591a\u6a21\u6001\u4efb\u52a1\u4e2d\u8868\u73b0\u60ca\u8273\uff0c\u4e00\u4e2a\u7ecf\u8fc7\u6df1\u5ea6\u9884\u8bad\u7ec3\u3001\u5177\u5907\u5f3a\u5927\u89c6\u89c9\u8bed\u8a00\u8868\u5f81\u80fd\u529b\u7684 VLM \u57fa\u5ea7\u663e\u7136\u80fd\u63d0\u4f9b\u65e0\u4e0e\u4f26\u6bd4\u7684\u52a9\u529b\u3002\u800c\u4e00\u65e6\u6253\u597d\u4e86\u8fd9\u4e2a\u57fa\u7840\uff0c\u540e\u7eed\u7684\u8bbe\u8ba1\u548c\u8bad\u7ec3\u624d\u80fd\u771f\u6b63\u53d1\u6325\u6700\u5927\u6f5c\u529b\u3002<\/p>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s7.51cto.com\/oss\/202501\/02\/937c4db26cc6e7db207680939fb9988b713f16.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<p><strong>4. \u8de8\u672c\u4f53\u6570\u636e\u4ec0\u4e48\u65f6\u5019\u52a0\u5165\u6700\u5408\u9002\uff1f<\/strong><\/p>\n<p>\u5b9e\u9a8c\u544a\u8bc9\u6211\u4eec\u4e00\u4e2a\u9ec4\u91d1\u6cd5\u5219\uff1a<strong>\u5728\u9884\u8bad\u7ec3\u9636\u6bb5\u5f15\u5165\u8de8\u672c\u4f53\u6570\u636e\uff08\u5982 Open-X Embodiment \u6570\u636e\u96c6\uff09\u53ef\u4ee5\u663e\u8457\u63d0\u5347\u6a21\u578b\u7684\u9c81\u68d2\u6027\u548c\u5c11\u6837\u672c\u573a\u666f\u4e0b\u7684\u8868\u73b0<\/strong>\u3002\u53cd\u4e4b\uff0c\u76f4\u63a5\u5c06\u8de8\u672c\u4f53\u6570\u636e\u548c\u5fae\u8c03\u6570\u636e\u6df7\u5408\u8bad\u7ec3\uff0c\u6548\u679c\u5c31\u6ca1\u90a3\u4e48\u663e\u8457\u4e86\u3002\u8fd9\u4e9b\u7ed3\u8bba\u4e3a\u672a\u6765 VLA \u6a21\u578b\u7684\u8bad\u7ec3\u7b56\u7565\u6307\u660e\u4e86\u65b9\u5411\u3002<\/p>\n<p>\u5177\u4f53\u5b9e\u9a8c\u4e2d\uff0c\u6211\u4eec\u5728 WidowX+Bridge \u548c Google Robot \u4e24\u5927\u73af\u5883\u4e0b\u5206\u522b\u8fdb\u884c\u4e86\u4e0d\u540c\u8bad\u7ec3\u7b56\u7565\u7684\u6d4b\u8bd5\uff1a<\/p>\n<p>WidowX+Bridge \u73af\u5883\uff1a<\/p>\n<ul data-id=\"u738a58b-c4D9D6PB\">\n<li data-id=\"ld70c578-rFKNSblf\">Bridge Finetune\uff1a\u76f4\u63a5\u5728\u5b8c\u6574\u7684 Bridge \u6570\u636e\u96c6\u4e0a\u5fae\u8c03\uff08\u6d4b\u8bd5\u4efb\u52a1\u4e0d\u5305\u62ec\u5728\u5185\uff09\u3002<\/li>\n<li data-id=\"ld70c578-9pn0XSh7\">OXE Pre-Train\uff1a\u5148\u7528 OXE \u6570\u636e\u96c6\u9884\u8bad\u7ec3\u6a21\u578b\u3002<\/li>\n<li data-id=\"ld70c578-fig6SVeI\">Post-Train\uff1a\u7528\u7ecf\u8fc7 OXE \u9884\u8bad\u7ec3\u7684\u6a21\u578b\u518d\u5728 Bridge \u6570\u636e\u96c6\u4e0a\u5fae\u8c03\u3002<\/li>\n<\/ul>\n<p>Google Robot \u73af\u5883\uff1a<\/p>\n<ul data-id=\"u738a58b-5g7amEE8\">\n<li data-id=\"ld70c578-frQPqDVg\">RT-Partial Finetune\uff1a\u4ec5\u5728\u7279\u5b9a\u7684 RT \u4efb\u52a1\u4e0a\u5fae\u8c03\u3002<\/li>\n<li data-id=\"ld70c578-JsxH8No8\">RT Finetune\uff1a\u5728\u5b8c\u6574\u7684 RT \u6570\u636e\u96c6\u4e0a\u5fae\u8c03\uff08\u5305\u62ec\u6d4b\u8bd5\u4efb\u52a1\uff09\u3002<\/li>\n<li data-id=\"ld70c578-7awb4WmD\">OXE Pre-Train\uff1a\u5148\u7528 OXE \u6570\u636e\u96c6\u9884\u8bad\u7ec3\u6a21\u578b\u3002<\/li>\n<li data-id=\"ld70c578-LPeEGkjk\">Post-Train\uff1a\u5728 OXE \u9884\u8bad\u7ec3\u57fa\u7840\u4e0a\u7528 RT \u6570\u636e\u96c6\u8fdb\u4e00\u6b65\u8bad\u7ec3\u3002<\/li>\n<\/ul>\n<p>\u5b9e\u9a8c\u7ed3\u679c\u8fdb\u4e00\u6b65\u9a8c\u8bc1\u4e86\uff1a<strong>\u5728\u9884\u8bad\u7ec3\u9636\u6bb5\u5f15\u5165\u8de8\u672c\u4f53\u6570\u636e\u4e0d\u4ec5\u80fd\u63d0\u5347\u6cdb\u5316\u80fd\u529b\uff0c\u8fd8\u80fd\u8ba9\u6a21\u578b\u5728\u5c11\u6837\u672c\u548c\u9ad8\u590d\u6742\u4efb\u52a1\u4e0b\u8868\u73b0\u66f4\u4f73<\/strong>\u3002<\/p>\n<p><img decoding=\"async\" title=\"\u56fe\u7247\" src=\"https:\/\/s5.51cto.com\/oss\/202501\/02\/88a531781bc3832d3aa354a3d237f6e69952f6.webp\" alt=\"\u56fe\u7247\" data-type=\"inline\" \/><\/p>\n<h4>\u5c55\u671b\u672a\u6765\uff1aVLA \u7684\u8fdb\u9636\u4e4b\u8def<\/h4>\n<p>\u867d\u7136 RoboVLMs \u5df2\u7ecf\u5f88\u80fd\u6253\u4e86\uff0c\u4f46\u63a5\u4e0b\u6765\u7684\u53d1\u5c55\u7a7a\u95f4\u66f4\u8ba9\u4eba\u671f\u5f85\uff01\u672a\u6765\u53ef\u4ee5\u63a2\u7d22\uff1a<\/p>\n<ol class=\"list-paddingleft-1\" data-id=\"odd3d8fc-8AQEc2Hd\">\n<li data-id=\"ld70c578-Us6X2siF\">\u66f4\u7ec6\u5316\u7684\u8bbe\u8ba1\u4f18\u5316\uff1a\u6bd4\u5982\u518d\u6253\u78e8 VLM \u5185\u90e8\u7ed3\u6784\u3001\u4fe1\u606f\u878d\u5408\u6a21\u5757\u548c\u8bad\u7ec3\u76ee\u6807\uff0c\u8ba9\u5b83\u66f4\u9ad8\u6548\u3002<\/li>\n<li data-id=\"ld70c578-2bVowPdv\">\u6311\u6218\u590d\u6742\u4efb\u52a1\uff1a\u50cf \u201c\u505a\u65e9\u9910\u201d \u8fd9\u79cd\u957f\u94fe\u6761\u4efb\u52a1\uff0c\u4e5f\u8bb8\u662f\u4e0b\u4e00\u4e2a\u7a81\u7834\u70b9\uff01<\/li>\n<li data-id=\"ld70c578-HI9q1WS7\">\u591a\u6a21\u6001\u534f\u4f5c\u80fd\u529b\uff1a\u8fdb\u4e00\u6b65\u8ba9\u673a\u5668\u4eba \u201c\u770b\u61c2\u201d\u3001\u201c\u542c\u6e05\u201d\u3001\u201c\u52a8\u5f97\u66f4\u806a\u660e\u201d\u3002<\/li>\n<\/ol>\n<p>RoboVLMs \u7684\u51fa\u73b0\uff0c\u9a8c\u8bc1\u4e86\u89c6\u89c9\u8bed\u8a00\u52a8\u4f5c\u6a21\u578b\u7684\u53ef\u80fd\u6027\uff0c\u4e5f\u8ba9\u673a\u5668\u4eba\u66f4\u63a5\u8fd1\u6210\u4e3a\u6211\u4eec\u7684\u5168\u80fd\u52a9\u624b\u3002\u672a\u6765\uff0c\u5b83\u4eec\u6216\u8bb8\u4e0d\u4ec5\u80fd\u7406\u89e3\u8bed\u8a00\u548c\u89c6\u89c9\uff0c\u8fd8\u80fd\u771f\u6b63\u5e2e\u6211\u4eec\u5b8c\u6210\u90a3\u4e9b\u7e41\u7410\u53c8\u590d\u6742\u7684\u4efb\u52a1\u3002\u63a5\u4e0b\u6765\u4f1a\u6709\u66f4\u591a\u60ca\u559c\u7b49\u7740\u6211\u4eec\uff01<\/p>\n<p>\u6587\u7ae0\u6765\u81ea\uff1a51CTO<\/p>\n<\/div>\n<\/div>\n<div class=\"pvc_clear\"><\/div>\n<p id=\"pvc_stats_23901\" class=\"pvc_stats total_only  \" data-element-id=\"23901\" style=\"\"><i class=\"pvc-stats-icon medium\" aria-hidden=\"true\"><svg xmlns=\"http:\/\/www.w3.org\/2000\/svg\" version=\"1.0\" viewBox=\"0 0 502 315\" preserveAspectRatio=\"xMidYMid meet\"><g transform=\"translate(0,332) scale(0.1,-0.1)\" fill=\"\" stroke=\"none\"><path d=\"M2394 3279 l-29 -30 -3 -207 c-2 -182 0 -211 15 -242 39 -76 157 -76 196 0 15 31 17 60 15 243 l-3 209 -33 29 c-26 23 -41 29 -80 29 -41 0 -53 -5 -78 -31z\"\/><path d=\"M3085 3251 c-45 -19 -58 -50 -96 -229 -47 -217 -49 -260 -13 -295 52 -53 146 -42 177 20 16 31 87 366 87 410 0 70 -86 122 -155 94z\"\/><path d=\"M1751 3234 c-13 -9 -29 -31 -37 -50 -12 -29 -10 -49 21 -204 19 -94 39 -189 45 -210 14 -50 54 -80 110 -80 34 0 48 6 76 34 21 21 34 44 34 59 0 14 -18 113 -40 219 -37 178 -43 195 -70 221 -36 32 -101 37 -139 11z\"\/><path d=\"M1163 3073 c-36 -7 -73 -59 -73 -102 0 -56 133 -378 171 -413 34 -32 83 -37 129 -13 70 36 67 87 -16 290 -86 209 -89 214 -129 231 -35 14 -42 15 -82 7z\"\/><path d=\"M3689 3066 c-15 -9 -33 -30 -42 -48 -48 -103 -147 -355 -147 -375 0 -98 131 -148 192 -74 13 15 57 108 97 206 80 196 84 226 37 273 -30 30 -99 39 -137 18z\"\/><path d=\"M583 2784 c-38 -19 -67 -74 -58 -113 9 -42 211 -354 242 -373 16 -10 45 -18 66 -18 51 0 107 52 107 100 0 39 -1 41 -124 234 -80 126 -108 162 -133 173 -41 17 -61 16 -100 -3z\"\/><path d=\"M4250 2784 c-14 -9 -74 -91 -133 -183 -95 -150 -107 -173 -107 -213 0 -55 33 -94 87 -104 67 -13 90 8 211 198 130 202 137 225 78 284 -27 27 -42 34 -72 34 -22 0 -50 -8 -64 -16z\"\/><path d=\"M2275 2693 c-553 -48 -1095 -270 -1585 -649 -135 -104 -459 -423 -483 -476 -23 -49 -22 -139 2 -186 73 -142 361 -457 571 -626 285 -228 642 -407 990 -497 242 -63 336 -73 660 -74 310 0 370 5 595 52 535 111 1045 392 1455 803 122 121 250 273 275 326 19 41 19 137 0 174 -41 79 -309 363 -465 492 -447 370 -946 591 -1479 653 -113 14 -422 18 -536 8z m395 -428 c171 -34 330 -124 456 -258 112 -119 167 -219 211 -378 27 -96 24 -300 -5 -401 -72 -255 -236 -447 -474 -557 -132 -62 -201 -76 -368 -76 -167 0 -236 14 -368 76 -213 98 -373 271 -451 485 -162 444 86 934 547 1084 153 49 292 57 452 25z m909 -232 c222 -123 408 -262 593 -441 76 -74 138 -139 138 -144 0 -16 -233 -242 -330 -319 -155 -123 -309 -223 -461 -299 l-81 -41 32 46 c18 26 49 83 70 128 143 306 141 649 -6 957 -25 52 -61 116 -79 142 l-34 47 45 -20 c26 -10 76 -36 113 -56z m-2057 25 c-40 -58 -105 -190 -130 -263 -110 -324 -59 -707 132 -981 25 -35 42 -64 37 -64 -19 0 -241 119 -326 174 -188 122 -406 314 -532 468 l-58 71 108 103 c185 178 428 349 672 473 66 33 121 60 123 61 2 0 -10 -19 -26 -42z\"\/><path d=\"M2375 1950 c-198 -44 -350 -190 -395 -379 -18 -76 -8 -221 19 -290 114 -284 457 -406 731 -260 98 52 188 154 231 260 27 69 37 214 19 290 -38 163 -166 304 -326 360 -67 23 -215 33 -279 19z\"\/><\/g><\/svg><\/i> <img loading=\"lazy\" decoding=\"async\" width=\"16\" height=\"16\" alt=\"Loading\" src=\"https:\/\/aif.amtbbs.org\/wp-content\/plugins\/page-views-count\/ajax-loader-2x.gif\" border=0 \/><\/p>\n<div class=\"pvc_clear\"><\/div>\n","protected":false},"excerpt":{"rendered":"<p>\u867d\u7136 VLAs \u5728\u5404\u79cd\u4efb\u52a1\u548c\u573a\u666f\u4e2d\u8868\u73b0\u62a2\u773c\uff0c\u4f46\u5927\u5bb6\u5728\u6a21\u578b\u8bbe\u8ba1\u4e0a\u5374\u8d70\u4e86\u5f88\u591a\u4e0d\u540c\u7684\u8def\uff0c\u6bd4\u5982\u7528\u4ec0\u4e48\u67b6\u6784\u3001\u600e\u4e48\u9009\u6570\u636e\u3001 [&hellip;]<\/p>\n<div class=\"pvc_clear\"><\/div>\n<p id=\"pvc_stats_23901\" class=\"pvc_stats total_only  \" data-element-id=\"23901\" style=\"\"><i class=\"pvc-stats-icon medium\" aria-hidden=\"true\"><svg xmlns=\"http:\/\/www.w3.org\/2000\/svg\" version=\"1.0\" viewBox=\"0 0 502 315\" preserveAspectRatio=\"xMidYMid meet\"><g transform=\"translate(0,332) scale(0.1,-0.1)\" fill=\"\" stroke=\"none\"><path d=\"M2394 3279 l-29 -30 -3 -207 c-2 -182 0 -211 15 -242 39 -76 157 -76 196 0 15 31 17 60 15 243 l-3 209 -33 29 c-26 23 -41 29 -80 29 -41 0 -53 -5 -78 -31z\"\/><path d=\"M3085 3251 c-45 -19 -58 -50 -96 -229 -47 -217 -49 -260 -13 -295 52 -53 146 -42 177 20 16 31 87 366 87 410 0 70 -86 122 -155 94z\"\/><path d=\"M1751 3234 c-13 -9 -29 -31 -37 -50 -12 -29 -10 -49 21 -204 19 -94 39 -189 45 -210 14 -50 54 -80 110 -80 34 0 48 6 76 34 21 21 34 44 34 59 0 14 -18 113 -40 219 -37 178 -43 195 -70 221 -36 32 -101 37 -139 11z\"\/><path d=\"M1163 3073 c-36 -7 -73 -59 -73 -102 0 -56 133 -378 171 -413 34 -32 83 -37 129 -13 70 36 67 87 -16 290 -86 209 -89 214 -129 231 -35 14 -42 15 -82 7z\"\/><path d=\"M3689 3066 c-15 -9 -33 -30 -42 -48 -48 -103 -147 -355 -147 -375 0 -98 131 -148 192 -74 13 15 57 108 97 206 80 196 84 226 37 273 -30 30 -99 39 -137 18z\"\/><path d=\"M583 2784 c-38 -19 -67 -74 -58 -113 9 -42 211 -354 242 -373 16 -10 45 -18 66 -18 51 0 107 52 107 100 0 39 -1 41 -124 234 -80 126 -108 162 -133 173 -41 17 -61 16 -100 -3z\"\/><path d=\"M4250 2784 c-14 -9 -74 -91 -133 -183 -95 -150 -107 -173 -107 -213 0 -55 33 -94 87 -104 67 -13 90 8 211 198 130 202 137 225 78 284 -27 27 -42 34 -72 34 -22 0 -50 -8 -64 -16z\"\/><path d=\"M2275 2693 c-553 -48 -1095 -270 -1585 -649 -135 -104 -459 -423 -483 -476 -23 -49 -22 -139 2 -186 73 -142 361 -457 571 -626 285 -228 642 -407 990 -497 242 -63 336 -73 660 -74 310 0 370 5 595 52 535 111 1045 392 1455 803 122 121 250 273 275 326 19 41 19 137 0 174 -41 79 -309 363 -465 492 -447 370 -946 591 -1479 653 -113 14 -422 18 -536 8z m395 -428 c171 -34 330 -124 456 -258 112 -119 167 -219 211 -378 27 -96 24 -300 -5 -401 -72 -255 -236 -447 -474 -557 -132 -62 -201 -76 -368 -76 -167 0 -236 14 -368 76 -213 98 -373 271 -451 485 -162 444 86 934 547 1084 153 49 292 57 452 25z m909 -232 c222 -123 408 -262 593 -441 76 -74 138 -139 138 -144 0 -16 -233 -242 -330 -319 -155 -123 -309 -223 -461 -299 l-81 -41 32 46 c18 26 49 83 70 128 143 306 141 649 -6 957 -25 52 -61 116 -79 142 l-34 47 45 -20 c26 -10 76 -36 113 -56z m-2057 25 c-40 -58 -105 -190 -130 -263 -110 -324 -59 -707 132 -981 25 -35 42 -64 37 -64 -19 0 -241 119 -326 174 -188 122 -406 314 -532 468 l-58 71 108 103 c185 178 428 349 672 473 66 33 121 60 123 61 2 0 -10 -19 -26 -42z\"\/><path d=\"M2375 1950 c-198 -44 -350 -190 -395 -379 -18 -76 -8 -221 19 -290 114 -284 457 -406 731 -260 98 52 188 154 231 260 27 69 37 214 19 290 -38 163 -166 304 -326 360 -67 23 -215 33 -279 19z\"\/><\/g><\/svg><\/i> <img loading=\"lazy\" decoding=\"async\" width=\"16\" height=\"16\" alt=\"Loading\" src=\"https:\/\/aif.amtbbs.org\/wp-content\/plugins\/page-views-count\/ajax-loader-2x.gif\" border=0 \/><\/p>\n<div class=\"pvc_clear\"><\/div>\n","protected":false},"author":56,"featured_media":23903,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[23,20,80],"tags":[1522],"class_list":["post-23901","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-23","category-20","category-80","tag-ai--"],"_links":{"self":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts\/23901","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/users\/56"}],"replies":[{"embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/comments?post=23901"}],"version-history":[{"count":1,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts\/23901\/revisions"}],"predecessor-version":[{"id":23908,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts\/23901\/revisions\/23908"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/media\/23903"}],"wp:attachment":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/media?parent=23901"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/categories?post=23901"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/tags?post=23901"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}