{"version":2,"relationshipURLBase":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html","relationshipResolution":"Resolve relative relationship URLs against relationshipURLBase; complete records and absolute URLs are in blocks.","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html","markdown":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.md","pages":[{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","unit":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","unitId":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","chapter":"course","number":1}],"aliases":[{"oldId":"b-94d6faa242b71ed9af1aa482","targetId":"b-94d6faa242b71ed9af1aa482","page":"reading-opening","operation":"move","beforeSHA256":"343fb71807f077069b41c6495a8824ea4d353c740175b5775ccc5145b3bf19ca","afterSHA256":"343fb71807f077069b41c6495a8824ea4d353c740175b5775ccc5145b3bf19ca","reason":"Same source body in authored semantic page"},{"oldId":"b-9a5249e7095e571ff7b8ccf8","targetId":"b-9a5249e7095e571ff7b8ccf8","page":"reading-opening","operation":"move","beforeSHA256":"a1db129cad717190bf997513fd0d7dfab66b07a7164a01c4a459bf9e17b83ece","afterSHA256":"a1db129cad717190bf997513fd0d7dfab66b07a7164a01c4a459bf9e17b83ece","reason":"Same source body in authored semantic page"},{"oldId":"b-d5d7ce9db086cda40c9e8cfa","targetId":"b-d5d7ce9db086cda40c9e8cfa","page":"reading-opening","operation":"move","beforeSHA256":"42cba31deccba2a350b9b23bd0ca1dcaa4c4c5cb9d35c773e1fcb89124788ccd","afterSHA256":"42cba31deccba2a350b9b23bd0ca1dcaa4c4c5cb9d35c773e1fcb89124788ccd","reason":"Same source body in authored semantic page"},{"oldId":"b-b93da83841d3027c7028f5fb","targetId":"b-b93da83841d3027c7028f5fb","page":"reading-opening","operation":"move","beforeSHA256":"41c70c2c07b82678ff0a60f2d009b6afa750c5c691d109e754a69abf23868b50","afterSHA256":"41c70c2c07b82678ff0a60f2d009b6afa750c5c691d109e754a69abf23868b50","reason":"Same source body in authored semantic page"},{"oldId":"b-b9cebaf8f9b4945585d9563a","targetId":"b-b9cebaf8f9b4945585d9563a","page":"reading-opening","operation":"move","beforeSHA256":"45ba6b3d2c09727e6a1c315bca9819ed3dc9a528bcd3967fd024e4e5c6a5647f","afterSHA256":"45ba6b3d2c09727e6a1c315bca9819ed3dc9a528bcd3967fd024e4e5c6a5647f","reason":"Same source body in authored semantic page"},{"oldId":"b-c0690f1d28bcf9be56703a5c","targetId":"b-c0690f1d28bcf9be56703a5c","page":"reading-opening","operation":"move","beforeSHA256":"c6febdec2ae248a7832a8dd6f8171f5558f284b515bf60f2b725a94807eec1e8","afterSHA256":"c6febdec2ae248a7832a8dd6f8171f5558f284b515bf60f2b725a94807eec1e8","reason":"Same source body in authored semantic page"},{"oldId":"b-33db88057d2d0964562baec0","targetId":"b-33db88057d2d0964562baec0","page":"reading-opening","operation":"move","beforeSHA256":"c6febdec2ae248a7832a8dd6f8171f5558f284b515bf60f2b725a94807eec1e8","afterSHA256":"c6febdec2ae248a7832a8dd6f8171f5558f284b515bf60f2b725a94807eec1e8","reason":"Same source body in authored semantic page"},{"oldId":"b-f2dfcb14688206ed1559ade3","targetId":"b-f2dfcb14688206ed1559ade3","page":"reading-opening","operation":"move","beforeSHA256":"a5f88743244fce1f0c7154025389a242e51ec43f3098e14c0e1d51ad7930b72d","afterSHA256":"a5f88743244fce1f0c7154025389a242e51ec43f3098e14c0e1d51ad7930b72d","reason":"Same source body in authored semantic page"},{"oldId":"b-339bd4a849b999c751cc8950","targetId":"b-339bd4a849b999c751cc8950","page":"reading-opening","operation":"move","beforeSHA256":"2074f6f25833791334e2622f481240a866c683e7f8b3b0ad0536db73dd43fd64","afterSHA256":"2074f6f25833791334e2622f481240a866c683e7f8b3b0ad0536db73dd43fd64","reason":"Same source body in authored semantic page"}],"blocks":[{"id":"b-94d6faa242b71ed9af1aa482","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-94d6faa242b71ed9af1aa482","content":"[CLS 03 · 视觉与语言：任务、训练与接口](/course/notebooks/680da245145447dd88983aabd136fd24.html)","sha256":"343fb71807f077069b41c6495a8824ea4d353c740175b5775ccc5145b3bf19ca","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":null,"nextBlock":"b-9a5249e7095e571ff7b8ccf8","prerequisites":[],"relatedBlocks":[]},{"id":"b-9a5249e7095e571ff7b8ccf8","type":"h1","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-9a5249e7095e571ff7b8ccf8","content":"# 我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sha256":"a1db129cad717190bf997513fd0d7dfab66b07a7164a01c4a459bf9e17b83ece","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-94d6faa242b71ed9af1aa482","nextBlock":"b-d5d7ce9db086cda40c9e8cfa","prerequisites":[{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-d5d7ce9db086cda40c9e8cfa","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-d5d7ce9db086cda40c9e8cfa","content":"主线阅读与优先实验","sha256":"42cba31deccba2a350b9b23bd0ca1dcaa4c4c5cb9d35c773e1fcb89124788ccd","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-9a5249e7095e571ff7b8ccf8","nextBlock":"b-b93da83841d3027c7028f5fb","prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-b93da83841d3027c7028f5fb","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-b93da83841d3027c7028f5fb","content":"caption历史、VQA、CLIP候选、关系控制、GLIP/LLaVA与生成桥梁","sha256":"41c70c2c07b82678ff0a60f2d009b6afa750c5c691d109e754a69abf23868b50","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-d5d7ce9db086cda40c9e8cfa","nextBlock":"b-b9cebaf8f9b4945585d9563a","prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-b9cebaf8f9b4945585d9563a","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-b9cebaf8f9b4945585d9563a","content":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题。现在拿走图像，只留下：","sha256":"45ba6b3d2c09727e6a1c315bca9819ed3dc9a528bcd3967fd024e4e5c6a5647f","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-b93da83841d3027c7028f5fb","nextBlock":"b-c0690f1d28bcf9be56703a5c","prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-c0690f1d28bcf9be56703a5c","type":"blockquote","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-c0690f1d28bcf9be56703a5c","content":"红色杯子放在木桌上。","sha256":"c6febdec2ae248a7832a8dd6f8171f5558f284b515bf60f2b725a94807eec1e8","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-b9cebaf8f9b4945585d9563a","nextBlock":"b-33db88057d2d0964562baec0","prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-33db88057d2d0964562baec0","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-33db88057d2d0964562baec0","content":"红色杯子放在木桌上。","sha256":"c6febdec2ae248a7832a8dd6f8171f5558f284b515bf60f2b725a94807eec1e8","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-c0690f1d28bcf9be56703a5c","nextBlock":"b-f2dfcb14688206ed1559ade3","prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-f2dfcb14688206ed1559ade3","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-f2dfcb14688206ed1559ade3","content":"使用者不再递交一张图供分析，而是要求系统产生新的视觉内容。**给候选打分，还没有给出生成这些像素的方法。** 我们需要说明输出的表示、学习目标，以及从条件到一个具体结果的生成过程。","sha256":"a5f88743244fce1f0c7154025389a242e51ec43f3098e14c0e1d51ad7930b72d","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-33db88057d2d0964562baec0","nextBlock":"b-339bd4a849b999c751cc8950","prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-339bd4a849b999c751cc8950","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/045f75954e664345ab8c0a5c766a2f16.html#b-339bd4a849b999c751cc8950","content":"第三讲的 captioning 与看图回答已经可能通过生成文本实现，MAE 也涉及重构。本讲的焦点是表示与任务接口；第四讲将把**生成分布、训练和采样本身**作为研究对象。","sha256":"2074f6f25833791334e2622f481240a866c683e7f8b3b0ad0536db73dd43fd64","chapter":{"id":"course","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"unit":{"id":"notebook-blocks/045f75954e664345ab8c0a5c766a2f16","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题"},"page":{"id":"reading-opening","title":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","number":1},"sectionTitle":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","currentQuestion":"我们已经用图像与句子做匹配，用词语指定区域，也能根据图像回答问题","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-f2dfcb14688206ed1559ade3","nextBlock":null,"prerequisites":[{"id":"b-9a5249e7095e571ff7b8ccf8","url":"#b-9a5249e7095e571ff7b8ccf8","role":"section-title"},{"id":"b-94d6faa242b71ed9af1aa482","url":"#b-94d6faa242b71ed9af1aa482","role":"problem-context"}],"relatedBlocks":[]}]}