6 papers
Approaching I/O-optimality for Approximate Attention
Pál András Papp, Aleksandros Sobczyk, Anastasios Zouzias
We revisit the I/O complexity of attention in large language models. Given query-key-value matrices , and a machine with fast memory size , the g…
Fast and Stable Triangular Inversion for Delta-Rule Linear Transformers
Aleksandros Sobczyk, Gioele Gottardo, Christos K. Matzoros +4
Linear attention has emerged as a cornerstone for efficient long-context architectures, as evidenced by its integration into state-of-the-art open-source models including Qwen3.5/3…
Parallel Scan on Ascend AI Accelerators
BartÅomiej Wróblewski, Gioele Gottardo, Anastasios Zouzias
We design and implement parallel prefix sum (scan) algorithms using Ascend AI accelerators. Ascend accelerators feature specialized computing units: the cube units for efficient ma…
Prefix Sums via Kronecker Products
Aleksandros Sobczyk, Anastasios Zouzias
In this work, we revisit prefix sums through the lens of linear algebra. We describe an identity that decomposes triangular all-ones matrices as a sum of two Kronecker products, an…
Segmented Operations using Matrix Multiplications
Aleksandros Sobczyk, Giuseppe Sorrentino, Anastasios Zouzias
Specialized computational units that perform small matrix multiplications as primitive operations are typically present in modern AI accelerators. However, these Matrix Multiplicat…
A Parallel Scan Algorithm in the Tensor Core Unit Model
Anastasios Zouzias, William F. McColl
We present a parallel scan (prefix sum) algorithm in the Tensor Core Unit (TCU) model of computation. The TCU model assumes that multiplication between two square matrices of const…