{"version":2,"relationshipURLBase":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html","relationshipResolution":"Resolve relative relationship URLs against relationshipURLBase; complete records and absolute URLs are in blocks.","title":"VQA：问题控制读什么，图像控制答什么","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html","markdown":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.md","pages":[{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","unit":"VQA：问题控制读什么，图像控制答什么","unitId":"notebook-blocks/1b53d32e9e684591be91385c30d46982","chapter":"course","number":1}],"aliases":[{"oldId":"b-d67c157ee8473d3215d53813","targetId":"b-d67c157ee8473d3215d53813","page":"reading-opening","operation":"move","beforeSHA256":"fa1d05e087449a3f39b93c1c4f3e6086732599f759cb40cfca565caa8afb8709","afterSHA256":"fa1d05e087449a3f39b93c1c4f3e6086732599f759cb40cfca565caa8afb8709","reason":"Same source body in authored semantic page"},{"oldId":"b-af208bfde05b5728c4d62414","targetId":"b-af208bfde05b5728c4d62414","page":"reading-opening","operation":"move","beforeSHA256":"21b128c28bd06de227bf551930692c46b35ded14c1e5519a2ead3a3b176fd798","afterSHA256":"21b128c28bd06de227bf551930692c46b35ded14c1e5519a2ead3a3b176fd798","reason":"Same source body in authored semantic page"},{"oldId":"b-f09ea1250916de0704b6d4c9","targetId":"b-f09ea1250916de0704b6d4c9","page":"reading-opening","operation":"move","beforeSHA256":"e91f24e4e4ca40e2099a0d81d30db6568f4bf2074b499eeee6e3959617bd97ac","afterSHA256":"e91f24e4e4ca40e2099a0d81d30db6568f4bf2074b499eeee6e3959617bd97ac","reason":"Same source body in authored semantic page"},{"oldId":"b-62a191556af9d2a132e274ff","targetId":"b-62a191556af9d2a132e274ff","page":"reading-opening","operation":"move","beforeSHA256":"a192bf58b03c510999643a276b8d2261f6c9389ea09aebd6ad3419fe70e2a7b9","afterSHA256":"a192bf58b03c510999643a276b8d2261f6c9389ea09aebd6ad3419fe70e2a7b9","reason":"Same source body in authored semantic page"},{"oldId":"b-46b7d7f448c5df9bd62c3b24","targetId":"b-46b7d7f448c5df9bd62c3b24","page":"reading-opening","operation":"move","beforeSHA256":"7cd1087c34e727cf26f3c9f759ef1904c52a96c60228e8922bcb000cfa7dc8d8","afterSHA256":"7cd1087c34e727cf26f3c9f759ef1904c52a96c60228e8922bcb000cfa7dc8d8","reason":"Same source body in authored semantic page"},{"oldId":"b-936cc27edbe593f6d2e67725","targetId":"b-936cc27edbe593f6d2e67725","page":"reading-opening","operation":"move","beforeSHA256":"efb06658a2c54105750282521a6f6c76787166ee696d375661365ac389ae0fa7","afterSHA256":"efb06658a2c54105750282521a6f6c76787166ee696d375661365ac389ae0fa7","reason":"Same source body in authored semantic page"},{"oldId":"b-4f46aa6524fa92be532356cd","targetId":"b-4f46aa6524fa92be532356cd","page":"reading-opening","operation":"move","beforeSHA256":"831626528161a3c074e9863fbc320f83c4427099d962c088e152ac90ac1b36db","afterSHA256":"831626528161a3c074e9863fbc320f83c4427099d962c088e152ac90ac1b36db","reason":"Same source body in authored semantic page"},{"oldId":"b-9facac346672ad769f5224d9","targetId":"b-9facac346672ad769f5224d9","page":"reading-opening","operation":"move","beforeSHA256":"7b74cc3e2f3205469e80f80e2972062d26a1c35de2e0f355d3b31ebd4564d1ad","afterSHA256":"7b74cc3e2f3205469e80f80e2972062d26a1c35de2e0f355d3b31ebd4564d1ad","reason":"Same source body in authored semantic page"},{"oldId":"b-a0063255f5172d1414c4601a","targetId":"b-a0063255f5172d1414c4601a","page":"reading-opening","operation":"rewrite","beforeSHA256":"b2f1ae3ddd998d2fd4d04045f73e8ac1a983afde542afa752a36e3866cda4682","afterSHA256":"c029a7f961a8f3f4fc6629058322f9734e26c4f0f26ae8f34ba8cb501a091ec6","reason":"Same source identity with figure semantics or reading context enriched"},{"oldId":"b-00cb6cfa44afa2527ed0fd53","targetId":"b-00cb6cfa44afa2527ed0fd53","page":"reading-opening","operation":"move","beforeSHA256":"dd655aba933eeca3efa133773efe24b93715ba055621e0986f10caa85cd381e9","afterSHA256":"08d0aa022e98bb75079211b073bb26e8245e159a9c4dcb17e3839b2d504f1d23","reason":"Same source body in authored semantic page"},{"oldId":"b-92cc925a344169467a8a2bd2","targetId":"b-92cc925a344169467a8a2bd2","page":"reading-opening","operation":"move","beforeSHA256":"707e7bc48c07443cf12149644f705a6c7c56301b0b49bf4524b81dbab3c92e29","afterSHA256":"707e7bc48c07443cf12149644f705a6c7c56301b0b49bf4524b81dbab3c92e29","reason":"Same source body in authored semantic page"},{"oldId":"b-dbaff8618fb05896b471fe85","targetId":"b-dbaff8618fb05896b471fe85","page":"reading-opening","operation":"move","beforeSHA256":"3db3903d93095f36602506f7c9a92536b7083ddc36b16b3535a3e4f55a1cdff6","afterSHA256":"3db3903d93095f36602506f7c9a92536b7083ddc36b16b3535a3e4f55a1cdff6","reason":"Same source body in authored semantic page"},{"oldId":"b-939e0675f9fe00804d825aca","targetId":"b-939e0675f9fe00804d825aca","page":"reading-opening","operation":"move","beforeSHA256":"3df86520780b10397874c31809ad84a890727b5848b6a2551cd77e3f4ffd45bd","afterSHA256":"3df86520780b10397874c31809ad84a890727b5848b6a2551cd77e3f4ffd45bd","reason":"Same source body in authored semantic page"},{"oldId":"block-1b53d32e9e684591be91385c30d46982--vqa问题控制读什么图像控制答什么","targetId":"b-46b7d7f448c5df9bd62c3b24","operation":"retain-source-identity"}],"blocks":[{"id":"b-d67c157ee8473d3215d53813","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-d67c157ee8473d3215d53813","content":"[CLS 09 · 图像条件输出：描述与问答](/course/notebooks/c9f328a00ef44ecc90804e47276a7811.html)","sha256":"fa1d05e087449a3f39b93c1c4f3e6086732599f759cb40cfca565caa8afb8709","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":null,"nextBlock":"b-af208bfde05b5728c4d62414","prerequisites":[],"relatedBlocks":[]},{"id":"b-af208bfde05b5728c4d62414","type":"h1","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-af208bfde05b5728c4d62414","content":"# VQA：问题控制读什么，图像控制答什么","sha256":"21b128c28bd06de227bf551930692c46b35ded14c1e5519a2ead3a3b176fd798","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-d67c157ee8473d3215d53813","nextBlock":"b-f09ea1250916de0704b6d4c9","prerequisites":[{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-f09ea1250916de0704b6d4c9","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-f09ea1250916de0704b6d4c9","content":"按问题继续","sha256":"e91f24e4e4ca40e2099a0d81d30db6568f4bf2074b499eeee6e3959617bd97ac","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-af208bfde05b5728c4d62414","nextBlock":"b-62a191556af9d2a132e274ff","prerequisites":[{"id":"b-af208bfde05b5728c4d62414","url":"#b-af208bfde05b5728c4d62414","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-62a191556af9d2a132e274ff","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-62a191556af9d2a132e274ff","content":"图像条件、teacher forcing、真实生成、greedy/beam与VQA控制","sha256":"a192bf58b03c510999643a276b8d2261f6c9389ea09aebd6ad3419fe70e2a7b9","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-f09ea1250916de0704b6d4c9","nextBlock":"b-46b7d7f448c5df9bd62c3b24","prerequisites":[{"id":"b-af208bfde05b5728c4d62414","url":"#b-af208bfde05b5728c4d62414","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-46b7d7f448c5df9bd62c3b24","type":"h2","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-46b7d7f448c5df9bd62c3b24","content":"## VQA：问题控制读什么，图像控制答什么","sha256":"7cd1087c34e727cf26f3c9f759ef1904c52a96c60228e8922bcb000cfa7dc8d8","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-62a191556af9d2a132e274ff","nextBlock":"b-936cc27edbe593f6d2e67725","prerequisites":[{"id":"b-af208bfde05b5728c4d62414","url":"#b-af208bfde05b5728c4d62414","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-936cc27edbe593f6d2e67725","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-936cc27edbe593f6d2e67725","content":"VQA建模 $p(a\\mid I,q)$。同图问数量、颜色、关系，需要不同信息；同一问题换相关图像，答案也应改变。输出可为有限答案分类，也可逐词生成，二者的答案范围和评价协议不同。","sha256":"efb06658a2c54105750282521a6f6c76787166ee696d375661365ac389ae0fa7","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[{"latex":"p(a\\mid I,q)","definitionContext":"VQA：问题控制读什么，图像控制答什么"}],"previousBlock":"b-46b7d7f448c5df9bd62c3b24","nextBlock":"b-4f46aa6524fa92be532356cd","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-4f46aa6524fa92be532356cd","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-4f46aa6524fa92be532356cd","content":"图A左红右蓝：问左色答红、问右色答蓝。换成图B左蓝右红，保持左色问题则应答蓝。若“视觉”分支只读取场景字典的 `left_color`，它是已知属性 oracle，不是从像素感知。[任务条件与VQA研究背景](/course/blocks/vqa-question-selects-information.html)。","sha256":"831626528161a3c074e9863fbc320f83c4427099d962c088e152ac90ac1b36db","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-936cc27edbe593f6d2e67725","nextBlock":"b-9facac346672ad769f5224d9","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-9facac346672ad769f5224d9","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-9facac346672ad769f5224d9","content":"[LAB 14](https://deepnote.com/workspace/ise3309jingao-5ba66b0a-4283-4f5a-ae06-3f68d3201e99/project/38755e4e-23f6-4f5f-8c68-8a2e4114b1c1/notebook/11d9f74871544a77b42024a9fe06b81a?utm_source=openai&utm_medium=mcp&utm_campaign=openaimcp&utm_content=11d9f74871544a77b42024a9fe06b81a&utm_term=get_notebook) 已有真正读取像素和问题ID的CNN问答，以及单独训练的question-only。2026-09-22本地保留180张图、每图三问共540图问对，先按图划分再扩展问题，避免同图跨切分。","sha256":"7b74cc3e2f3205469e80f80e2972062d26a1c35de2e0f355d3b31ebd4564d1ad","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-4f46aa6524fa92be532356cd","nextBlock":"b-a0063255f5172d1414c4601a","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-a0063255f5172d1414c4601a","type":"table","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-a0063255f5172d1414c4601a","content":"| 保存条件 | 数量 | 左色 | 右形 | 全部 |\n| --- | --- | --- | --- | --- |\n| 图像＋问题 | 100% | 100% | 100% | 100% |\n| question-only | 50% | 33.33% | 50% | 44.44% |\n| 同模型，像素置零 | 50% | 33.33% | 50% | 44.44% |\n| 同模型，循环错配图像 | 83.33% | 50% | 0% | 44.44% |","sha256":"c029a7f961a8f3f4fc6629058322f9734e26c4f0f26ae8f34ba8cb501a091ec6","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"table":{"headers":["保存条件","数量","左色","右形","全部"],"rows":[["保存条件","数量","左色","右形","全部"],["图像＋问题","100%","100%","100%","100%"],["question-only","50%","33.33%","50%","44.44%"],["同模型，像素置零","50%","33.33%","50%","44.44%"],["同模型，循环错配图像","83.33%","50%","0%","44.44%"]]},"previousBlock":"b-9facac346672ad769f5224d9","nextBlock":"b-00cb6cfa44afa2527ed0fd53","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-00cb6cfa44afa2527ed0fd53","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-00cb6cfa44afa2527ed0fd53","content":"![同图换问、同问换图](/assets/display/2f7b2168f5b21ee2095709fa.webp)","sha256":"08d0aa022e98bb75079211b073bb26e8245e159a9c4dcb17e3839b2d504f1d23","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-a0063255f5172d1414c4601a","nextBlock":"b-92cc925a344169467a8a2bd2","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-92cc925a344169467a8a2bd2","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-92cc925a344169467a8a2bd2","content":"循环错配不是独立重抽，排列常保留数量却改变形状；总分相同掩盖不同错误。正确图像下的成功和干预后的下降，支持这个有限任务使用了视觉信息。问题是三个固定模板ID，不主张开放语言泛化。[逐问题指标的完整解释](/course/blocks/vqa-pixel-evidence.html)。","sha256":"707e7bc48c07443cf12149644f705a6c7c56301b0b49bf4524b81dbab3c92e29","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-00cb6cfa44afa2527ed0fd53","nextBlock":"b-dbaff8618fb05896b471fe85","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-dbaff8618fb05896b471fe85","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-dbaff8618fb05896b471fe85","content":"**备课模拟对话。** 学生：“无图也有44%，图像是不是不重要？”教师：“二选一、三选一、二选一的常量率平均就是4/9；同任务视觉模型100%且错配后下降，才说明新增条件有作用。”","sha256":"3db3903d93095f36602506f7c9a92536b7083ddc36b16b3535a3e4f55a1cdff6","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-92cc925a344169467a8a2bd2","nextBlock":"b-939e0675f9fe00804d825aca","prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-939e0675f9fe00804d825aca","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/1b53d32e9e684591be91385c30d46982.html#b-939e0675f9fe00804d825aca","content":"本轮没有重跑这些实验。Caption产生新句子，匹配返回候选分数，VQA还要根据问题选择信息。理解区别后，按需要进入CLS 10 的匹配或 CLS 11 的区域指代，不必把所有册重新读一遍。","sha256":"3df86520780b10397874c31809ad84a890727b5848b6a2551cd77e3f4ffd45bd","chapter":{"id":"course","title":"VQA：问题控制读什么，图像控制答什么"},"unit":{"id":"notebook-blocks/1b53d32e9e684591be91385c30d46982","title":"VQA：问题控制读什么，图像控制答什么"},"page":{"id":"reading-opening","title":"VQA：问题控制读什么，图像控制答什么","number":1},"sectionTitle":"VQA：问题控制读什么，图像控制答什么","currentQuestion":"VQA：问题控制读什么，图像控制答什么","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-dbaff8618fb05896b471fe85","nextBlock":null,"prerequisites":[{"id":"b-46b7d7f448c5df9bd62c3b24","url":"#b-46b7d7f448c5df9bd62c3b24","role":"section-title"},{"id":"b-d67c157ee8473d3215d53813","url":"#b-d67c157ee8473d3215d53813","role":"problem-context"}],"relatedBlocks":[]}]}