面向异常检测的联邦轻量头部时序Transformer CAN编码器

arXiv cs.LG 论文
anomaly-detection federated-learning transformer automotive-security can-bus edge-ai

摘要

提出一种基于轻量级时序Transformer CAN编码器与联邦学习轻量头部的隐私保护框架，用于检测车内网络中的细微时序与上下文异常，无需交换原始CAN数据即可提升共享模型。

arXiv:2610.10613v1 Announce Type: new Abstract: Modern vehicles rely on large numbers of Electronic Control Units (ECUs) that constantly exchange information over the Controller Area Network (CAN) bus. Due to the rapidity, structure, and repetition of this communication, even slight variations in timing, payload values, or message patterns can point to unusual activity. Whether due to errors, malfunctions, or deliberate interference, these anomalies are frequently subtle and challenging to identify with conventional methods that handle messages separately or rely on manually created rules. Motivated by this gap, we present a privacy-preserving framework for anomaly detection in in-vehicle networks, based on a Temporal Transformer CAN Encoder with Federated Lightweight Heads, to better capture these irregularities. The detection of subtle temporal and contextual anomalies is made possible by a lightweight Transformer encoder that learns how these signals evolve over time, while a federated learning mechanism enables several vehicles or ECUs to work together to improve a shared model without exchanging raw CAN data. This combination of federated learning and temporal sequence modeling provides robust anomaly detection performance while maintaining efficiency and privacy, according to experiments conducted on open-source datasets.
查看原文
查看缓存全文

缓存时间: 2026/10/09 06:39 

# Temporal transformer CAN encoder with federated lightweight heads for anomaly detection
Source: [https://arxiv.org/abs/2610.10613](https://arxiv.org/abs/2610.10613)
[View PDF](https://arxiv.org/pdf/2610.10613)

> Abstract:Modern vehicles rely on large numbers of Electronic Control Units \(ECUs\) that constantly exchange information over the Controller Area Network \(CAN\) bus\. Due to the rapidity, structure, and repetition of this communication, even slight variations in timing, payload values, or message patterns can point to unusual activity\. Whether due to errors, malfunctions, or deliberate interference, these anomalies are frequently subtle and challenging to identify with conventional methods that handle messages separately or rely on manually created rules\. Motivated by this gap, we present a privacy\-preserving framework for anomaly detection in in\-vehicle networks, based on a Temporal Transformer CAN Encoder with Federated Lightweight Heads, to better capture these irregularities\. The detection of subtle temporal and contextual anomalies is made possible by a lightweight Transformer encoder that learns how these signals evolve over time, while a federated learning mechanism enables several vehicles or ECUs to work together to improve a shared model without exchanging raw CAN data\. This combination of federated learning and temporal sequence modeling provides robust anomaly detection performance while maintaining efficiency and privacy, according to experiments conducted on open\-source datasets\.

## Submission history

From: Eirini Liotou Dr\. \[[view email](https://arxiv.org/show-email/5b09350c/2610.10613)\] **\[v1\]**Wed, 7 Oct 2026 07:13:20 UTC \(474 KB\)

相似文章

面向金融欺诈检测的多流时序融合

arXiv cs.LG

提出用于金融欺诈检测的多流欺诈Transformer（MSFT），该模型使用Transformer独立编码交易、登录和风险事件流，并通过时间感知位置编码和门控融合进行融合，在大型数据集上实现了0.9961的AUROC。

面向 VANETs 的编码器共享分层联邦多任务学习

arXiv cs.LG

本文提出了 EN-HMTFL，一种面向车载自组织网络（VANETs）的编码器共享分层多任务联邦学习框架。该框架允许车辆通过共享编码器协作训练异构感知任务，同时将任务特定的解码器保留在本地，从而将准确率提升高达 24%，并将通信轮次减少多达 28.8%。

空间感应头：多维细胞自动机的上下文学习

arXiv cs.LG

本文提出"空间感应头"（spatial induction heads），一种两层 gather-and-match 电路，解释 Transformer 如何在多维细胞自动机的序列化数据中进行上下文学习。研究表明，该电路可逼近贝叶斯后验并在确定性规则上实现精确预测，训练后的两层 Transformer 能泛化到未见过的规则。