arXiv:2608. 04057v1 Announce Type: cross Abstract: Top-$k$ selection determines which components of a sparse model remain active.
Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.
arXiv:2608. 06912v1 Announce Type: new Abstract: The top-$k$ operation is a fundamental building block of modern sparse computation, enabling token routing, expert activation, memory selection, and attention pruning.
arXiv:2606. 01155v1 Announce Type: cross Abstract: Scaling laws for dense LLMs under infinite data are well explored, but how sparsity interacts with limited data is not.
arXiv:2606. 14648v1 Announce Type: new Abstract: Robust machine learning and optimization rely on the uncertainty model choice.
arXiv:2512. 13996v3 Announce Type: replace Abstract: Sparse Mixture-of-Experts architectures are essential for scaling model capacity efficiently, yet the standard Top-$k$ routing imposes a rigid sparsity pattern that ignores the intrinsic variance in token difficulty and layer-specific computational needs.
arXiv:2606. 11347v1 Announce Type: cross Abstract: We propose Annealed Entropic Allocation, an annealed weighted soft-min framework for sequential budget allocation in ranking and selection.
arXiv:2607. 27000v1 Announce Type: cross Abstract: Optimization in non-convex neural network models is strongly influenced by the geometry of the solution space: sparse, isolated, point-like clusters are typically algorithmically inaccessible, whereas wide and flat regions can be found efficiently despite being relatively rare.