{"type":"link","version":"1.0","title":"Sparse routing in an embedding encoder decouples representational capacity from per-token latency","author_name":"AI Archs","author_url":"https://ai-arch.pages.dev","provider_name":"AI Archs","provider_url":"https://ai-arch.pages.dev","url":"https://ai-arch.pages.dev/n/sparse-routing-in-embedding-encoders-decouples-capacity-from-latency","thumbnail_url":"https://ai-arch.pages.dev/og/sparse-routing-in-embedding-encoders-decouples-capacity-from-latency.png","thumbnail_width":1200,"thumbnail_height":630}