[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-speculative-decoding":3},{"tag":4,"articles":10},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":9},"64d6308f-76b6-4e70-8b2f-fba3e5fb760a","speculative decoding","speculative-decoding",0,null,[11,20],{"id":12,"slug":13,"title":14,"summary":15,"category":16,"image_url":17,"cover_image":17,"language":18,"created_at":19},"6dcd6852-b95a-4f62-853a-cc7eb32fff1a","gemma-4-assistant-models-faster-draft-tokens-en","Gemma 4 assistant models get faster draft tokens","Gemma 4 E2B and E4B assistant models use centroid masking to cut lm_head work about 45x with little quality loss.","tools","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778278254841-r19z.png","en","2026-05-08T22:10:34.02358+00:00",{"id":21,"slug":22,"title":23,"summary":24,"category":25,"image_url":26,"cover_image":26,"language":18,"created_at":27},"3d747e63-24a0-4e20-9e83-e2263d06a779","speckv-adaptive-speculative-decoding-gamma-en","SpecKV tunes speculative decoding on the fly","SpecKV adapts speculative decoding’s token budget per step, using draft-model signals to beat fixed gamma across compression settings.","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1777961487463-lssf.png","2026-05-05T06:10:40.207648+00:00"]