[{"data":1,"prerenderedAt":40},["ShallowReactive",2],{"post-document:\u002Fdeep_learning\u002Ftransformer\u002F2026\u002F08\u002F06\u002Ftransformer-vision-transformer\u002F":3},{"id":4,"title":5,"body":6,"categories":20,"date":23,"description":24,"extension":25,"image":26,"key_concepts":26,"last_modified_at":26,"legacyPath":27,"meta":28,"navigation":30,"part":26,"path":31,"published":30,"robots":26,"seo":32,"series":26,"stem":33,"strengths":26,"summary":34,"tags":35,"tradeoffs":26,"__hash__":39},"posts\u002Fposts\u002FDeep_Learning\u002FTransformer\u002F2026-08-06-transformer-vision-transformer.md","Transformer와 Vision Transformer",{"type":7,"value":8,"toc":16},"minimark",[9,13],[10,11,12],"p",{},"Transformer 기본\nTransformer가 등장한 이유\nToken과 Embedding\nSelf-Attention의 직관\nMulti-Head Attention\nPositional Encoding\nEncoder와 Decoder\nMasked Attention\nKV Cache\nPre-training과 Fine-tuning\nContext Length",[10,14,15],{},"Vision Transformer\n이미지를 Patch로 나누는 이유\nCNN과 ViT 비교\nViT의 전체 데이터 흐름\nSwin Transformer의 Window Attention\nHierarchical Vision Transformer\nHybrid CNN–Transformer\nDETR\nMask Transformer\nVision Foundation Model\nSelf-supervised Vision Model",{"title":17,"searchDepth":18,"depth":18,"links":19},"",2,[],[21,22],"Deep_Learning","Transformer","2026-08-06 00:00:00 +0900","Transformer의 기본 구조와 Self-Attention, Vision Transformer, Swin Transformer, DETR의 흐름을 정리한다.","md",null,"\u002Fdeep_learning\u002Ftransformer\u002F2026\u002F08\u002F06\u002Ftransformer-vision-transformer\u002F",{"layout":29},"post",true,"\u002Fposts\u002Fdeep_learning\u002Ftransformer\u002F2026-08-06-transformer-vision-transformer",{"title":5,"description":24},"posts\u002FDeep_Learning\u002FTransformer\u002F2026-08-06-transformer-vision-transformer","Transformer와 Vision Transformer 계열의 핵심 개념과 발전 흐름을 정리한다.",[36,22,37,38],"Deep Learning","Vision Transformer","ViT","xxymtgPRY5Es6iln6PiRmAEjtEepiaQpDmc3qdRrrRA",1788744789312]