[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-53d67819-1d5c-4c1a-8dbb-1f49d9c76304":3},{"id":4,"title":5,"summary":6,"original_url":7,"source_id":8,"tags":9,"published_at":20,"created_at":21,"modified_at":22,"is_published":23,"publish_type":24,"image_url":13,"view_count":25},"53d67819-1d5c-4c1a-8dbb-1f49d9c76304","BiSCo-LLM 把 LLM 量化推过 2-bit 墙：Lookup-free 球面编码 + 类别恢复蒸馏，告别 VQ 码本","大模型部署受显存、权重带宽、checkpoint 体积三件事卡死,sub-2-bit 量化长期被视作最后一公里。arXiv 2607.08643 (Shao 等,2026-07-09) 提出的 BiSCo-LLM 给出一条不依赖 VQ 码本的新路:把局部权重块映射到单位超球面,二值化成纯位流符号当主载荷,再用 Residual BSQ 阶段编码基线球面码留下的重建误差,作为无码本情况下的显式率-失真通道。在此之上,对 Transformer 各模块类别做替换后做一次 category-wise 恢复蒸馏,把局部重建误差和整体行为错位拉回来。整套体系还附一条 8-bit 保护通道稳敏感通道,搭配神经解码器、LoRA adapter 一并计入存储。意义是把 LLM 权重压到 sub-2-bit 的同时既保留 VQ 类方法的容量优势,又摆脱显式码本存储与查表开销,对低带宽 checkpoint 传输、内存受限的端侧推理场景非常实用。","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.08643","7437aeb9-930c-4866-a2e9-48003c1a792b",[10,14,17],{"id":11,"name":12,"slug":12,"description":13,"color":13},"2d9c2fb0-2be5-4ad1-aedb-e9747addf355","compression",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":18,"name":19,"slug":19,"description":13,"color":13},"b49648f9-963e-4082-8684-3d085b7358fe","quantization","2026-07-10T08:00:00Z","2026-07-11T14:13:49.261992Z","2026-07-11T14:13:49.262002Z",true,"agent",3]