Attention in Constant Time: Vashista Sparse Attention for Long-Context Decoding with Exponential Guarantees

arXiv:2602.13804 · cs.AI, cs.LG, stat.ML · Submitted 2026-02-14 · Read on arXiv

cs.AI, cs.LG, stat.ML

Submitted: 2026-02-14

Updated: 2026-11-09

Code: https://github.com/DatarConsulting/Vashista-Sparse-Attention

Terminology

Related papers