Lingcn: Structural linearized graph convolutional network for homomorphically encrypted inference
Abstract The growth of Graph Convolution Network (GCN) model sizes has revolutionized
numerous applications, surpassing human performance in areas such as personal …
numerous applications, surpassing human performance in areas such as personal …
Understanding the potential of fpga-based spatial acceleration for large language model inference
Recent advancements in large language models (LLMs) boasting billions of parameters
have generated a significant demand for efficient deployment in inference workloads. While …
have generated a significant demand for efficient deployment in inference workloads. While …
A Review on the emerging technology of TinyML
Tiny Machine Learning (TinyML) is an emerging technology proposed by the scientific
community for developing autonomous and secure devices that can gather, process, and …
community for developing autonomous and secure devices that can gather, process, and …
A survey of FPGA and ASIC designs for transformer inference acceleration and optimization
BJ Kang, HI Lee, SK Yoon, YC Kim, SB Jeong… - Journal of Systems …, 2024 - Elsevier
Recently, transformer-based models have achieved remarkable success in various fields,
such as computer vision, speech recognition, and natural language processing. However …
such as computer vision, speech recognition, and natural language processing. However …
Edgellm: A highly efficient cpu-fpga heterogeneous edge accelerator for large language models
The rapid advancements in artificial intelligence (AI), particularly the Large Language
Models (LLMs), have profoundly affected our daily work and communication forms …
Models (LLMs), have profoundly affected our daily work and communication forms …
A survey on hardware accelerators for large language models
C Kachris - arXiv preprint arXiv:2401.09890, 2024 - arxiv.org
Large Language Models (LLMs) have emerged as powerful tools for natural language
processing tasks, revolutionizing the field with their ability to understand and generate …
processing tasks, revolutionizing the field with their ability to understand and generate …
Resource Efficient Deep Learning Hardware Watermarks with Signature Alignment
J Clements, Y Lao - Proceedings of the AAAI Conference on Artificial …, 2024 - ojs.aaai.org
Deep learning intellectual properties (IPs) are high-value assets that are frequently
susceptible to theft. This vulnerability has led to significant interest in defending the field's …
susceptible to theft. This vulnerability has led to significant interest in defending the field's …
A Cost-Efficient FPGA Implementation of Tiny Transformer Model using Neural ODE
I Okubo, K Sugiura, H Matsutani - arXiv preprint arXiv:2401.02721, 2024 - arxiv.org
Transformer is an emerging neural network model with attention mechanism. It has been
adopted to various tasks and achieved a favorable accuracy compared to CNNs and RNNs …
adopted to various tasks and achieved a favorable accuracy compared to CNNs and RNNs …
A theoretical and empirical exploration of TileTrans for effective tile pruning
Y Li, F Ino - Knowledge-Based Systems, 2024 - Elsevier
In this paper, we propose a reparameterization method that is capable of transforming the
attention layer of deep neural networks (DNNs) for reducing the loss of tile pruning. The …
attention layer of deep neural networks (DNNs) for reducing the loss of tile pruning. The …
Applications of Pruning Methods in Natural Language Processing
M Touheed, U Zubair, D Sabir, A Hassan… - IEEE …, 2024 - ieeexplore.ieee.org
Deep neural networks (DNN) are in high demand because of their widespread applications
in natural language processing, image processing, and a lot of other domains. However …
in natural language processing, image processing, and a lot of other domains. However …