IEEE

EMBARK: Memory Bounded Architectural Improvement in CSR-CSC Sparse Matrix Multiplication

Abstract: Sparse Matrix Multiplication (SpMM) is a crucial algorithm in modern platforms such as Artificial Intelligence (AI), Graph Neural Network (GNN), Graph Convolutional Network (GCN), and neural ...
marktechpost

Google AI Introduces STATIC: A Sparse Matrix Framework Delivering 948x Faster Constrained Decoding for LLM Based Generative Retrieval

In industrial recommendation systems, the shift toward Generative Retrieval (GR) is replacing traditional embedding-based nearest neighbor search with Large Language Models (LLMs). These models ...
GitHub

COO to CSR Sparse Matrix Converter on FPGA

python main.py # modify benchmarklist.csv if necessary. COO and CSR data are stored in coo_data/ and csr_data/, respectively. make sim_start # Run simulation (default: ./sim_coo2csr) Output waveforms ...
GitHub

03-matrix-multiplication.py

* Program re-ordering for improved L2 cache hit rate. * Automatic performance tuning. # Motivations # Matrix multiplications are a key building block of most modern high-performance computing systems.
IEEE

Vectorization of Narrow Matrix Multiplication for Ascend AI Inference Acceleration

Abstract: This research proposes and evaluates a novel approach to optimizing matrix multiplication (MatMul) on Huawei Ascend NPUs, motivated by a key insight: during matrix-vector multiplication ...