arXiv Machine Learning By Chad A. Capps

Depth-Staggered Fibonacci Spacing for Sparse Attention: Static Schedules Beat Learned Dilation and Extrapolate Where Dense Attention Fails

Read the original on arXiv Machine Learning →

arXiv:2606. 28560v1 Announce Type: cross Abstract: We study sparse self-attention in which each query attends to a dense local window plus a set of Fibonacci-spaced offsets, with a per-layer scalar alpha that compresses or expands the spacing.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.