paper-with-me

Papers

Accelerating Sparse Ternary GEMM for Quantized ML on Apple Silicon

2025-10-08 · Baraq Lipshitz, Alessio Melone, Charalampos Maraziaris, Muhammed Bilal arxiv

Sparse Ternary General Matrix-Matrix Multiplication (GEMM) remains under-optimized in existing libraries for Apple Silicon CPUs. We present a Sparse Ternary GEMM kernel optimized specifically for Apple's M-series processors. We propose a set of architecture-aware optimizations, including a novel blocked and interleaved sparse data format to improve memory locality, strategies to increase Instruction-Level Parallelism (ILP), and NEON-based Single Instruction Multiple Data (SIMD) vectorization to exploit data-level parallelism. Our scalar implementation achieves up to a 5.98x performance increase over a traditional Ternary Compressed Sparse Column (TCSC) baseline for large matrices with 50% ternary nonzero values (sparsity), reaching up to a 50.2% of the processor's theoretical peak performance, and remains stable across varying sparsity levels. Our vectorized implementation delivers up to a 5.59x performance increase for large matrices with 25% sparsity, and remains stable across varying sparsity levels.

📄 PDF Abstract BibTeX arXiv:2510.06957

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

NGEMM: Optimizing GEMM for Deep Learning via Compiler-based Techniques

2019-10-01 · Wenlei Bao, Li-Wen Chang, Yang Chen, Ke Deng 외

Quantization has emerged to be an effective way to significantly boost the performance of deep neural networks (DNNs) by utilizing low-bit computations. Despite having lower numerical precision, quantized DNNs are able t…

Deep LearningQuantization

Agent Memory Below the Prompt: Persistent Q4 KV Cache for Multi-Agent LLM Inference on Edge Devices

2026-02-17 · Yakov Pyotr Shkolnikov arxiv

Multi-agent LLM systems on edge devices face a memory management problem: device RAM is too small to hold every agent's KV cache simultaneously. On Apple M4 Pro with 10.2 GB of cache budget, only 3 agents fit at 8K conte…

LoTA-QAF: Lossless Ternary Adaptation for Quantization-Aware Fine-Tuning

2025-05-24 · Junyu Chen, Junzhuo Li, Zhen Peng, Wenjie Wang 외

Quantization and fine-tuning are crucial for deploying large language models (LLMs) on resource-constrained edge devices. However, fine-tuning quantized models presents significant challenges, primarily stemming from: Fi…

Computational EfficiencyMMLUQuantization

LiquidGEMM: Hardware-Efficient W4A8 GEMM Kernel for High-Performance LLM Serving

2025-09-01 · Huanqi Hu, Bowen Xiao, Shixuan Sun, Jianian Yin 외 arxiv

Quantization is a critical technique for accelerating LLM inference by reducing memory footprint and improving computational efficiency. Among various schemes, 4-bit weight and 8-bit activation quantization (W4A8) offers…

Computational Efficiency

Ternary and Binary Quantization for Improved Classification

2022-03-31 · Weizhi Lu, Mingrui Chen, Kai Guo, Weiyu Li

Dimension reduction and data quantization are two important methods for reducing data complexity. In the paper, we study the methodology of first reducing data dimension by random projection and then quantizing the proje…

ClassificationDimensionality ReductionQuantization