{"version":2,"relationshipURLBase":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html","relationshipResolution":"Resolve relative relationship URLs against relationshipURLBase; complete records and absolute URLs are in blocks.","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html","markdown":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.md","pages":[{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","unit":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","unitId":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","chapter":"course","number":1}],"aliases":[{"oldId":"b-e524db241733e2e7f4d4ea9c","targetId":"b-e524db241733e2e7f4d4ea9c","page":"reading-opening","operation":"move","beforeSHA256":"343fb71807f077069b41c6495a8824ea4d353c740175b5775ccc5145b3bf19ca","afterSHA256":"343fb71807f077069b41c6495a8824ea4d353c740175b5775ccc5145b3bf19ca","reason":"Same source body in authored semantic page"},{"oldId":"b-89dae56ce7967ec4b0aa070d","targetId":"b-89dae56ce7967ec4b0aa070d","page":"reading-opening","operation":"move","beforeSHA256":"f9f28a02293b2301a5f43844f5da568dcfc393eb6d2195ae56ede05bd4c8a069","afterSHA256":"f9f28a02293b2301a5f43844f5da568dcfc393eb6d2195ae56ede05bd4c8a069","reason":"Same source body in authored semantic page"},{"oldId":"b-2be843134d9fb37f3725e677","targetId":"b-2be843134d9fb37f3725e677","page":"reading-opening","operation":"move","beforeSHA256":"42cba31deccba2a350b9b23bd0ca1dcaa4c4c5cb9d35c773e1fcb89124788ccd","afterSHA256":"42cba31deccba2a350b9b23bd0ca1dcaa4c4c5cb9d35c773e1fcb89124788ccd","reason":"Same source body in authored semantic page"},{"oldId":"b-78b74fd4ab07866a12251f90","targetId":"b-78b74fd4ab07866a12251f90","page":"reading-opening","operation":"move","beforeSHA256":"41c70c2c07b82678ff0a60f2d009b6afa750c5c691d109e754a69abf23868b50","afterSHA256":"41c70c2c07b82678ff0a60f2d009b6afa750c5c691d109e754a69abf23868b50","reason":"Same source body in authored semantic page"},{"oldId":"b-006d991dee1aee631db3c1ed","targetId":"b-006d991dee1aee631db3c1ed","page":"reading-opening","operation":"move","beforeSHA256":"826c37982b1fccb48a900e7b147f828e26b6399914cb55a9725a7e159f48cebb","afterSHA256":"826c37982b1fccb48a900e7b147f828e26b6399914cb55a9725a7e159f48cebb","reason":"Same source body in authored semantic page"},{"oldId":"b-e9636eaacb5f66792e8f1ebf","targetId":"b-e9636eaacb5f66792e8f1ebf","page":"reading-opening","operation":"move","beforeSHA256":"3d5c7bd3c37b0b27a8ff2291c779f7cb4f476415ede88c31e8859e72e9e04035","afterSHA256":"3d5c7bd3c37b0b27a8ff2291c779f7cb4f476415ede88c31e8859e72e9e04035","reason":"Same source body in authored semantic page"},{"oldId":"b-11eecf58cbd78ed8bf755a22","targetId":"b-11eecf58cbd78ed8bf755a22","page":"reading-opening","operation":"move","beforeSHA256":"c08ee8b374daa51edb2a12d7498572a81bca062005f11e4fa535ab6c6cdec853","afterSHA256":"c08ee8b374daa51edb2a12d7498572a81bca062005f11e4fa535ab6c6cdec853","reason":"Same source body in authored semantic page"}],"blocks":[{"id":"b-e524db241733e2e7f4d4ea9c","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-e524db241733e2e7f4d4ea9c","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":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":null,"nextBlock":"b-89dae56ce7967ec4b0aa070d","prerequisites":[],"relatedBlocks":[]},{"id":"b-89dae56ce7967ec4b0aa070d","type":"h1","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-89dae56ce7967ec4b0aa070d","content":"# CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sha256":"f9f28a02293b2301a5f43844f5da568dcfc393eb6d2195ae56ede05bd4c8a069","chapter":{"id":"course","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-e524db241733e2e7f4d4ea9c","nextBlock":"b-2be843134d9fb37f3725e677","prerequisites":[{"id":"b-e524db241733e2e7f4d4ea9c","url":"#b-e524db241733e2e7f4d4ea9c","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-2be843134d9fb37f3725e677","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-2be843134d9fb37f3725e677","content":"主线阅读与优先实验","sha256":"42cba31deccba2a350b9b23bd0ca1dcaa4c4c5cb9d35c773e1fcb89124788ccd","chapter":{"id":"course","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-89dae56ce7967ec4b0aa070d","nextBlock":"b-78b74fd4ab07866a12251f90","prerequisites":[{"id":"b-89dae56ce7967ec4b0aa070d","url":"#b-89dae56ce7967ec4b0aa070d","role":"section-title"},{"id":"b-e524db241733e2e7f4d4ea9c","url":"#b-e524db241733e2e7f4d4ea9c","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-78b74fd4ab07866a12251f90","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-78b74fd4ab07866a12251f90","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":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-2be843134d9fb37f3725e677","nextBlock":"b-006d991dee1aee631db3c1ed","prerequisites":[{"id":"b-89dae56ce7967ec4b0aa070d","url":"#b-89dae56ce7967ec4b0aa070d","role":"section-title"},{"id":"b-e524db241733e2e7f4d4ea9c","url":"#b-e524db241733e2e7f4d4ea9c","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-006d991dee1aee631db3c1ed","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-006d991dee1aee631db3c1ed","content":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域。这里需要分开目标、参照物与关系，不能看到 blue 就返回蓝杯。","sha256":"826c37982b1fccb48a900e7b147f828e26b6399914cb55a9725a7e159f48cebb","chapter":{"id":"course","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-78b74fd4ab07866a12251f90","nextBlock":"b-e9636eaacb5f66792e8f1ebf","prerequisites":[{"id":"b-89dae56ce7967ec4b0aa070d","url":"#b-89dae56ce7967ec4b0aa070d","role":"section-title"},{"id":"b-e524db241733e2e7f4d4ea9c","url":"#b-e524db241733e2e7f4d4ea9c","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-e9636eaacb5f66792e8f1ebf","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-e9636eaacb5f66792e8f1ebf","content":"[P63](/course/pages/l3-r2-p63.html) 的 GLIP 说明对象级语言定位的学习路线；[CLS 11 · 语言定位与视觉模型接口](https://deepnote.com/workspace/ise3309jingao-5ba66b0a-4283-4f5a-ae06-3f68d3201e99/project/38755e4e-23f6-4f5f-8c68-8a2e4114b1c1/notebook/95169c3e298c4c819373bb4534f3dff9?utm_source=openai&utm_medium=mcp&utm_campaign=openaimcp&utm_content=95169c3e298c4c819373bb4534f3dff9&utm_term=get_notebook) 展开监督、空间输出和模型接口；[LAB 15 · 语言指代小练习：目标、参照与关系](https://deepnote.com/workspace/ise3309jingao-5ba66b0a-4283-4f5a-ae06-3f68d3201e99/project/38755e4e-23f6-4f5f-8c68-8a2e4114b1c1/notebook/7dc83786d067465aa76264475db2c62d?utm_source=openai&utm_medium=mcp&utm_campaign=openaimcp&utm_content=7dc83786d067465aa76264475db2c62d&utm_term=get_notebook) 用像素提框与关系规则做完整对照。后者不是 GLIP 推理，但能准确解释为什么旧关键词规则选错。","sha256":"3d5c7bd3c37b0b27a8ff2291c779f7cb4f476415ede88c31e8859e72e9e04035","chapter":{"id":"course","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-006d991dee1aee631db3c1ed","nextBlock":"b-11eecf58cbd78ed8bf755a22","prerequisites":[{"id":"b-89dae56ce7967ec4b0aa070d","url":"#b-89dae56ce7967ec4b0aa070d","role":"section-title"},{"id":"b-e524db241733e2e7f4d4ea9c","url":"#b-e524db241733e2e7f4d4ea9c","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-11eecf58cbd78ed8bf755a22","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/443e8c7bfc094e03bed653667baf21a7.html#b-11eecf58cbd78ed8bf755a22","content":"[带坐标的完整算例](/course/blocks/grounding-target-reference-relation.html) · [GLIP 的方法和正面迁移证据](/course/blocks/glip-language-objective.html)。","sha256":"c08ee8b374daa51edb2a12d7498572a81bca062005f11e4fa535ab6c6cdec853","chapter":{"id":"course","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"unit":{"id":"notebook-blocks/443e8c7bfc094e03bed653667baf21a7","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域"},"page":{"id":"reading-opening","title":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","number":1},"sectionTitle":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","currentQuestion":"CLIP 整图相似度给配对程度；“蓝杯左边的杯子”还要求输出目标区域","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-e9636eaacb5f66792e8f1ebf","nextBlock":null,"prerequisites":[{"id":"b-89dae56ce7967ec4b0aa070d","url":"#b-89dae56ce7967ec4b0aa070d","role":"section-title"},{"id":"b-e524db241733e2e7f4d4ea9c","url":"#b-e524db241733e2e7f4d4ea9c","role":"problem-context"}],"relatedBlocks":[]}]}
