A Simple Transformer Pipeline for Full-Key Side-Channel Attacks on Uncropped Datasets

2026-08-31 · Jimmy Gammell, Kaushik Roy arxiv

Deep learning-based side-channel analysis has historically focused on single-byte targets and manually cropped traces, which risks discarding exploitable leakage. While recent work has proposed specialized architectures and resampling techniques to address this gap, the literature lacks a simple transformer baseline for simultaneous full-key attacks on uncropped traces. We present an open-source transformer implementation for uncropped full-key attacks which uses the standard transformer encoder backbone, adapting only the input and output layers to the side-channel setting. We release our implementation, training recipes, and pretrained weights for uncropped ASCADv1f, ASCADv1r, and CHES-CTF-2018 which achieve performance competitive with previously-reported results, while using less than 10GB of VRAM and requiring at most 3.34 hours of training on a single NVIDIA A6000.

📄 PDF Abstract BibTeX arXiv:2608.30105

Code (0)

등록된 구현이 없습니다.

Deep Learning-Based Channel Estimation

2018-10-13 · Mehran Soltani, Vahid Pourahmadi, Ali Mirzaei, Hamid Sheikhzadeh

In this paper, we present a deep learning (DL) algorithm for channel estimation in communication systems. We consider the time-frequency response of a fast fading communication channel as a two-dimensional image. The aim…

Deep Learning Denoising Image Restoration Image Super-Resolution+1

Group Generalized Mean Pooling for Vision Transformer

2022-12-08 · Byungsoo Ko, Han-Gyu Kim, Byeongho Heo, Sangdoo Yun 외

Vision Transformer (ViT) extracts the final representation from either class token or an average of all patch tokens, following the architecture of Transformer in Natural Language Processing (NLP) or Convolutional Neural…

Image Retrieval Representation Learning Retrieval

SCT: A Simple Baseline for Parameter-Efficient Fine-Tuning via Salient Channels

2023-09-15 · Henry Hengyuan Zhao, Pichao Wang, Yuyang Zhao, Hao Luo 외

Pre-trained vision transformers have strong representation benefits to various downstream tasks. Recently, many parameter-efficient fine-tuning (PEFT) methods have been proposed, and their experiments demonstrate that tu…

Domain Generalization Few-Shot Learning parameter-efficient fine-tuning Transfer Learning

CCTrans: Simplifying and Improving Crowd Counting with Transformer

2021-09-29 · Ye Tian, Xiangxiang Chu, Hongpeng Wang

Most recent methods used for crowd counting are based on the convolutional neural network (CNN), which has a strong ability to extract local features. But CNN inherently fails in modeling the global context due to the li…

Crowd Counting

PETformer: Long-term Time Series Forecasting via Placeholder-enhanced Transformer

2023-08-09 · Shengsheng Lin, Weiwei Lin, Wentai Wu, SongBo Wang 외

Recently, the superiority of Transformer for long-term time series forecasting (LTSF) tasks has been challenged, particularly since recent work has shown that simple models can outperform numerous Transformer-based appro…

Computational Efficiency Multivariate Time Series Forecasting Time Series Time Series Forecasting