정기석

전체 206건 중 1번부터 10번까지의 결과를 표시합니다.

2026

Article

TiMM-ViT: Token-Importance-Aware Token Merging with Multi-Exit Training and an FPGA-based Vision Transformer Accelerator for Image Classification

Article

Hardware-Efficient Activation Approximation based on Error-Sensitivity Analysis for Deep Neural Networks

Article

LOP+SAMM: DNN Inference Accelerator with Hardware Loop Offloading and Segment-Wise On-Chip Memory Data Synchronization

Article

SpinOut: Enhanced Rotation-based Quantization for LLM by Outlier Injection

Article

An Area-Efficient Mixed-Precision Accelerator with Output-Error-Based Quantization for ViT

2025

Article

HPN-SpGEMM: Hybrid PIM-NMP for SpGEMM

Article

Hope: An Efficient Accelerator With Head-Wise Overlap Processing for Sparse Attention in Vision Transformer

Article

FloatMax: An Efficient Accelerator for Transformer-Based Models Exploiting Tensor-Wise Adaptive Floating-Point Quantization

2024

Article

A Low Power Attention and Softmax Accelerator for Large Language Models Inference

Article

CAMPuS: Concurrent Acceleration of Memory Access and Parallel Processing in Near-Memory SpMV Architecture

상세 보기

정기석

연구분야

자료유형

발행연도

키워드

언어

TiMM-ViT: Token-Importance-Aware Token Merging with Multi-Exit Training and an FPGA-based Vision Transformer Accelerator for Image Classification

Hardware-Efficient Activation Approximation based on Error-Sensitivity Analysis for Deep Neural Networks

LOP+SAMM: DNN Inference Accelerator with Hardware Loop Offloading and Segment-Wise On-Chip Memory Data Synchronization

SpinOut: Enhanced Rotation-based Quantization for LLM by Outlier Injection

An Area-Efficient Mixed-Precision Accelerator with Output-Error-Based Quantization for ViT

HPN-SpGEMM: Hybrid PIM-NMP for SpGEMM

Hope: An Efficient Accelerator With Head-Wise Overlap Processing for Sparse Attention in Vision Transformer

FloatMax: An Efficient Accelerator for Transformer-Based Models Exploiting Tensor-Wise Adaptive Floating-Point Quantization

A Low Power Attention and Softmax Accelerator for Large Language Models Inference

CAMPuS: Concurrent Acceleration of Memory Access and Parallel Processing in Near-Memory SpMV Architecture