top-k-selection

Prof-K is a probabilistic one-pass filtering algorithm for fast, scalable top-k selection with correctness guarantees, achieving 1.5x–10x speedups over PyTorch topk and RadiK, especially in large-scale small-k regimes.

LaPrune 引入了一个可微的稀疏选择层，在百万规模下独立控制预算和掩码硬度，利用 LapSum 屏障和归一化二阶矩约束，在保留所选质量的同时近似硬 top-k 选择。