{"author_name":"Paper Radio","author_url":"https://montaan.com/","height":480,"html":"<iframe src=\"https://montaan.com/embed/episode/6317058b-393c-4a8f-a56f-2f6494a3d92d\" width=\"1280\" height=\"480\" title=\"AVSG: Accelerated Vectorized Sparse Gather for Efficient KV Cache Offload in Sparse-Attention LLM Serving — Paper Radio\" frameborder=\"0\" allow=\"autoplay; fullscreen; picture-in-picture\" allowfullscreen loading=\"lazy\" style=\"border:0;display:block;max-width:100%;\"></iframe>","provider_name":"Paper Radio","provider_url":"https://montaan.com/","thumbnail_url":"https://montaan.com/media/episodes/6317058b-393c-4a8f-a56f-2f6494a3d92d/episode.poster.webp","title":"AVSG: Accelerated Vectorized Sparse Gather for Efficient KV Cache Offload in Sparse-Attention LLM Serving","type":"rich","version":"1.0","width":1280}