[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-vision-language-models":3},{"tag":4,"articles":10},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":9},"8f9c3d57-f8b5-4e57-a26f-cc84962e6114","vision-language models","vision-language-models",0,null,[11,20,27],{"id":12,"slug":13,"title":14,"summary":15,"category":16,"image_url":17,"cover_image":17,"language":18,"created_at":19},"1d84a671-4772-43ea-af56-3d447893a94c","memdreamer-long-video-understanding-memory-retrieval-en","MemDreamer tackles long-video overload","MemDreamer splits perception from reasoning to make hours-long video understanding fit in a tiny context window.","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780902190707-ajbq.png","en","2026-06-08T07:02:32.833899+00:00",{"id":21,"slug":22,"title":23,"summary":24,"category":16,"image_url":25,"cover_image":25,"language":18,"created_at":26},"a65ad2e8-de08-4108-82cb-c3737a17ac6f","ipt-vlms-hidden-space-reasoning-en","IPT helps VLMs reason about hidden space","Imaginative Perception Tokens improve multimodal models’ ability to reason about unseen spatial structure.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780468449119-aqbt.png","2026-06-03T06:32:47.048757+00:00",{"id":28,"slug":29,"title":30,"summary":31,"category":16,"image_url":32,"cover_image":32,"language":18,"created_at":33},"348358ba-3a10-4057-9694-235127ebd848","protoada-multimodal-continual-instruction-tuning-en","ProtoAda tackles multimodal continual tuning drift","ProtoAda adds format-aware prototypes and geometry-aware updates to reduce interference in multimodal continual instruction tuning.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780381094836-d2yl.png","2026-06-02T06:17:35.36763+00:00"]