Session 8A - Top-k-Convolution and the Quest for Near-Linear Output-Sensitive Subset Sum