相关论文: SwiftTron: An Efficient Hardware Accelerator for Q…

Neural network quantization is a promising compression technique to reduce memory footprint and save energy consumption, potentially leading to real-time inference. However, there is a performance gap between quantized and full-precision…

Spiking Neural Networks (SNNs) are amenable to deployment on edge devices and neuromorphic hardware due to their lower dissipation. Recently, SNN-based transformers have garnered significant interest, incorporating attention mechanisms akin…

Recently, Diffusion Transformers (DiTs) have emerged as a dominant architecture in video generation, surpassing U-Net-based models in terms of performance. However, the enhanced capabilities of DiTs come with significant drawbacks,…

Transformers have revolutionized deep learning and generative modeling to enable unprecedented advancements in natural language processing tasks and beyond. However, designing hardware accelerators for executing transformer models is…

FPGA accelerators for lightweight neural convolutional networks (LWCNNs) have recently attracted significant attention. Most existing LWCNN accelerators focus on single-Computing-Engine (CE) architecture with local optimization. However,…

Quantum computing is an emerging technology on the verge of reshaping industries, while simultaneously challenging existing cryptographic algorithms. FALCON, a recent standard quantum-resistant digital signature, presents a challenging…

With the surging popularity of edge computing, the need to efficiently perform neural network inference on battery-constrained IoT devices has greatly increased. While algorithmic developments enable neural networks to solve increasingly…

Token compression aims to speed up large-scale vision transformers (e.g. ViTs) by pruning (dropping) or merging tokens. It is an important but challenging task. Although recent advanced approaches achieved great success, they need to…

Transformer models gain popularity because of their superior inference accuracy and inference throughput. However, the transformer is computation-intensive, causing a long inference time. The existing works on transformer inference…

In natural language processing (NLP), the "Transformer" architecture was proposed as the first transduction model replying entirely on self-attention mechanisms without using sequence-aligned recurrent neural networks (RNNs) or convolution,…

Deep Learning has been one of the most disruptive technological advancements in recent times. The high performance of deep learning models comes at the expense of high computational, storage and power requirements. Sensing the immediate…

Recent advances in multimodal models highlight the pivotal role of image tokenization in high-resolution image generation. By compressing images into compact latent representations, tokenizers enable generative models to operate in…

When implementations of the Transformer's self-attention layer utilize SRAM instead of DRAM, they can achieve significant speedups. The Tenstorrent Grayskull architecture provides a large SRAM, distributed across a grid of cores. This work…

The deployment of intelligent reinforcement learning (RL) agents on resource-constrained edge devices remains a fundamental challenge due to the substantial memory, computational, and energy requirements of modern deep learning systems.…

The increasing amount of data processed on edge and the demand for reducing the energy consumption for large neural network architectures have initiated the transition from traditional von Neumann architectures towards in-memory computing…

We tackle the problem of producing compact models, maximizing their accuracy for a given model size. A standard solution is to train networks with Quantization Aware Training, where the weights are quantized during training and the…

Transformers have achieved state-of-the-art performance across various tasks, but suffer from a notable quadratic complexity in sequence length due to the attention mechanism. In this work, we propose MonarchAttention -- a novel approach to…

While significant improvements have been made in recent years in terms of end-to-end automatic speech recognition (ASR) performance, such improvements were obtained through the use of very large neural networks, unfit for embedded use on…

While vision transformers (ViTs) have continuously achieved new milestones in the field of computer vision, their sophisticated network architectures with high computation and memory costs have impeded their deployment on resource-limited…

This paper presents a novel network compression framework Kernel Quantization (KQ), targeting to efficiently convert any pre-trained full-precision convolutional neural network (CNN) model into a low-precision version without significant…