Sparse-attention decoders rely on exact Top-K selection to choose the most important key-value entries for each query token. In long-context LLM serving, this Top-K stage runs once per decode query and becomes a meaningful latency bottleneck even when the indexer and attention kernels are already highly optimized. We present \textbf{Guess-Verify-Refine (GVR)}, a data-aware exact Top-K algorithm for sparse-attention decoding on NVIDIA Blackwell. GVR exploits temporal correlation across consecutive decode steps: it uses the previous step's Top-K as a prediction signal, computes pre-indexed statistics, narrows to a valid threshold by secant-style counting in 1-2 global passes, verifies candidates with a ballot-free collector, and finishes exact selection in shared memory. We connect this behavior to the Toeplitz / RoPE structure of DeepSeek Sparse Attention (DSA) indexer scores and validate the design on real DeepSeek-V3.2 workloads integrated into TensorRT-LLM. GVR achieves an average \textbf{1.88x} single-operator speedup over the production radix-select kernel, with up to \textbf{2.42x} per layer per step, while preserving bit-exact Top-K outputs. In controlled TEP8 min-latency deployment, it improves end-to-end TPOT by up to \textbf{7.52%} at 100K context, with larger gains at longer contexts and smaller but still positive gains under speculative decoding. While implemented and validated in the current TensorRT-LLM DSA stack on Blackwell, the same principle may extend to sparse-attention decoders whose decode-phase Top-K exhibits temporal stability.

翻译：稀疏注意力解码器依赖精确的Top-K选择机制，为每个查询词元选取最重要的键值条目。在长上下文大语言模型服务中，该Top-K阶段在每个解码查询时执行一次，即便索引器与注意力核已高度优化，仍构成显著的延迟瓶颈。我们提出**猜证精炼（GVR）**，一种面向NVIDIA Blackwell稀疏注意力解码的数据感知精确Top-K算法。GVR利用连续解码步骤间的时间相关性：将前一步的Top-K结果作为预测信号，计算预索引统计量，通过1-2次全局遍历中的割线法计数缩小至有效阈值，使用无投票冲突收集器验证候选集，并在共享内存中完成精确选择。我们将此行为关联至DeepSeek稀疏注意力索引器评分的Toeplitz/旋转位置编码结构，并在集成于TensorRT-LLM的实际DeepSeek-V3.2工作负载上验证设计。相比生产级基数选择核，GVR实现平均**1.88倍**单算子加速比（每层每步最高**2.42倍**），同时保持比特级精确的Top-K输出。在受控TEP8最小延迟部署场景下，GVR在100K上下文长度下将端到端TPOT提升最高**7.52%**，更长上下文下增益更大，在推测解码场景中仍保持较小但正向的增益。该方案虽在Blackwell平台现有TensorRT-LLM DSA框架中实现验证，其核心原理可拓展至解码阶段Top-K具有时间稳定性的任意稀疏注意力解码器。