DoSA：点稀疏注意力，无损加速 Top-K 选择
原标题：Paper page - DoSA: Dots Sparse Attention
AI 摘要
该论文提出 DoSA（Dots Sparse Attention），一种无损稀疏注意力方法，通过分数上下界证书提前拒绝候选 token，仅对仍符合条件的候选进行完整打分，从而保持精确的 Top-K 选择。其 GPU 实现使完整 Indexer 相比 SGLang 提速最高 2.74 倍、相比 LiteTopK 提速 1.69 倍。在 GLM-5.2 和 DeepSeek-V3.2 上的端到端评估中，首 token 延迟分别相比两个基线最高提速 1.94 倍和 1.50 倍。
正文节选
DoSA: Dots Sparse Attention Abstract Sparse attention visits few keys, yet finding them can still require scoring every key with every indexer head. We introduce DoSA (Dots Sparse Attention), a lossless method for making this computation conditional while preserving exact token selection. Its central observation is that Top-K selection requires sufficient evidence to reject a candidate, not necessarily its full score. A seed set supplies a lower bound on the winning threshold. Nonnegative head p