[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-067a3f68-9bc8-4486-9715-5a391e537909":3},{"id":4,"title":5,"summary":6,"original_url":7,"source_id":8,"tags":9,"published_at":23,"created_at":24,"modified_at":25,"is_published":26,"publish_type":27,"image_url":13,"view_count":28},"067a3f68-9bc8-4486-9715-5a391e537909","ETH Zurich 把四种循环线性注意力摆到统一坐标系:Kimi Delta Attention 拿下最低损失,新跨层路由 CLVR 让 DeltaNet 再涨一分","让 LLM 处理长上下文最大的瓶颈是 softmax 注意力的平方复杂度。过去一年,基于 delta-rule 的循环线性注意力在 Mamba \u002F RWKV 之后爆发出 DeltaNet、Gated DeltaNet、Kimi Delta Attention、Gated DeltaNet-2 等多个变体——但它们各自的读写控制记忆衰减差异几乎不透明,工程团队无法判断该选哪种。\n\n7 月 8 日,ETH Zurich 的 Cerruti 等人在 arXiv:2607.07953 把四种循环线性注意力用统一的循环记忆符号重新表达,并在 350M 参数、15B tokens 的匹配设定下做了横向对比。三条关键观察:**一、Kimi Delta Attention + Muon 优化器在最终验证损失上最低**,反映 Moonshot 在长上下文架构上的领先;**二、纯 Gated DeltaNet + AdamW 训练吞吐最高**,而 hybrid(混合 softmax 与循环层)虽然在损失上略胜,但吞吐显著下降;**三、Muon 优化器在所有匹配架构下都稳定优于 AdamW**,提示 community 应重新审视优化器选择。\n\n团队还提出一个轻量跨层路由机制 CLVR (Cross-Layer Value Routing):把上一层的写值路由到下一层的 hidden stream,而不是更直觉的写误差。在 DeltaNet 与 Gated DeltaNet 上,CLVR 都让最终验证损失下降。\n\n这篇论文的真正价值不是某项突破,而是把过去一年碎片化的 linear attention 实验拉到同一张控制变量表上——但它没有测推理延迟,而真正上线最在意的恰恰是 1k–100k 上下文下的每 token 推理速度。配套代码已开源,值得复现到下游长上下文模型。","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.07953","7437aeb9-930c-4866-a2e9-48003c1a792b",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"5e628969-6d2a-437f-998a-104e4b16cfb1","ai-progress",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"0ef8513a-0a26-42f0-b6f9-5b6dadded45c","efficiency",{"id":18,"name":19,"slug":19,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":21,"name":22,"slug":22,"description":13,"color":13},"4f214978-cac1-4f39-aa4b-f92a0d0934b7","transformer","2026-07-12T04:00:00Z","2026-07-12T04:06:03.934019Z","2026-07-12T04:06:03.934031Z",true,"agent",2]