日志异常检测是人工智能 IT 运维（AIOps）领域的关键组件。考虑到不同领域的日志数据，在真实工业场景中为未知领域重新训练整个网络是低效的。然而，以往的深度模型仅关注提取同一领域内日志序列的语义，导致在多领域日志上泛化能力差。为缓解此问题，我们提出一个统一的基于 Transformer 的日志异常检测框架（LogFormer），以提升跨不同领域的泛化能力，其中我们建立了一个包含预训练和基于适配器微调阶段的两阶段流程。具体而言，我们的模型首先在源域上预训练，以获取日志数据的共享语义知识。然后，我们通过共享参数将此类知识迁移到目标域。此外，提出了 Log-Attention 模块，以补充日志解析所忽略的信息。所提方法在三个公开数据集和一个真实世界数据集上进行了评估。在多个基准上的实验结果证明了我们 LogFormer 的有效性，且具有更少的可训练参数和更低的训练成本。
英文摘要：
Log anomaly detection is a key component in the field of artificial intelligence for IT operations (AIOps). Considering log data of variant domains, retraining the whole network for unknown domains is inefficient in real industrial scenarios. However, previous deep models merely focused on extracting the semantics of log sequences in the same domain, leading to poor generalization on multi-domain logs. To alleviate this issue, we propose a unified Transformer-based framework for Log anomaly detection (LogFormer) to improve the generalization ability across different domains, where we establish a two-stage process including the pre-training and adapter-based tuning stage. Specifically, our model is first pre-trained on the source domain to obtain shared semantic knowledge of log data. Then, we transfer such knowledge to the target domain via shared parameters. Besides, the Log-Attention module is proposed to supplement the information ignored by the log-paring. The proposed method is evaluated on three public and one real-world datasets. Experimental results on multiple benchmarks demonstrate the effectiveness of our LogFormer with fewer trainable parameters and lower training costs.
MINES: Explainable Anomaly Detection through Web API Invariant Inference
National University of Singapore（国立新加坡大学）；Shanghai Jiao Tong University（上海交通大学）；Singapore Management University（新加坡管理学院）；GovTech Singapore（新加坡政府科技局）；Chongqing University（重庆大学）
NSF-MAP: Neurosymbolic Multimodal Fusion for Robust and Interpretable Anomaly Prediction in Assembly Pipelines
Artificial Intelligence Institute, University of South Carolina（南卡罗来纳大学人工智能研究所）；Clemson Composites Center, Clemson University（克莱姆森大学复合材料中心）；Intelligent Circuits, Architectures and Systems Lab, University of South Carolina（南卡罗来纳大学智能电路、架构和系统实验室）
Semi-Supervised Anomaly Detection Pipeline for SOZ Localization Using Ictal-Related Chirp
Krembil Research Institute – University Health Network (UHN)（克伦比尔研究所 – 大学健康网络）；Institute of Biomaterials & Biomedical Engineering (IBBME), University of Toronto（多伦多大学生物材料与生物医学工程研究所）；Department of Physiology, University of Toronto（多伦多大学生理学系）；KITE Research Institute, Toronto Rehabilitation Institute - University Health Network (UHN)（KITE研究所，多伦多康复研究所 – 大学健康网络）
MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data
Marietta Blau Institute for Particle Physics, Austrian Academy of Sciences（奥地利科学院玛丽埃塔·布劳粒子物理研究所）；Università degli Studi di Urbino ‘Carlo Bo’（乌尔比诺“卡洛·博”大学）；INFN（意大利国家核物理研究所）
A Classifier-Based Approach to Multi-Class Anomaly Detection Applied to Astronomical Time-Series
Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of Technology（麻省理工学院卡弗里天体物理与空间研究所）；Irvington High School（欧文顿高中）；Department of Physics and Astronomy, University of the Western Cape（西开普大学物理与天文学系）；South African Radio Astronomy Observatory（南非射电天文观测台）
Revisiting Non-separable Binary Classification and its Applications in Anomaly Detection
Georgia Institute of Technology（佐治亚理工学院）；School of Cybersecurity and Privacy（网络安全与隐私学院）；Lengo AI（Lengo AI公司）；School of Electrical and Computer Engineering（电气与计算机工程学院）；Apple（苹果公司）
European Space Agency Benchmark for Anomaly Detection in Satellite Telemetry
KP Labs；Airbus Defence and Space GmbH（空客防务与航天有限公司）；Opole University of Technology（奥波莱理工大学）；Silesian University of Technology（西里西亚理工大学）；CGI Deutschland B.V. & Co. KG（CGI 德国有限两合公司）；European Space Operations Centre, European Space Agency（欧洲空间局欧洲空间操作中心）；Telespazio Germany GmbH（意大利空间公司德国有限公司）；LSE Space GmbH（LSE Space 有限公司）；Solenix GmbH（Solenix 有限公司）
What ZTF Saw Where Rubin Looked: Anomaly Hunting in DR23
Sternberg astronomical institute, Lomonosov Moscow State University, Universitetsky pr. 13, Moscow, 119234, Russia；Lomonosov Moscow State University, Faculty of Physics, Leninskie Gory, 1-2, Moscow, 119991, Russia；Space Research Institute of the Russian Academy of Sciences (IKI), 84/32 Profsoyuznaya Street, Moscow, 117997, Russia；Physics Department, University of Surrey, Stag Hill campus, Guildford, GU2 7XH, UK.；Laboratory of Astrochemical Research, Ural Federal University, Ekaterinburg, ul. Mira d. 19, Yekaterinburg, 620002, Russia；Universit\'e Clermont Auvergne, CNRS/IN2P3, LPC, F-63000 Clermont-Ferrand, France；National Research University Higher School of Economics, 21/4 Staraya Basmannaya Ulitsa, Moscow, 105066, Russia；The McWilliams Center for Cosmology \& Astrophysics, Department of Physics, Carnegie Mellon University, Pittsburgh, PA 15213, USA
An Explainable LLM Agent Layer for Open-World Anomaly Detection in Oil Wells
Laboratory of Scientific Computing and Visualization, Federal University of Alagoas（阿拉戈斯联邦大学科学计算与可视化实验室）；Computing Institute, Federal University of Alagoas（阿拉戈斯联邦大学计算学院）
Context-Aware Graph Attention for Unsupervised Telco Anomaly Detection
Telenor Research and Innovation（Telenor研究与创新）；Norwegian Research Council（挪威研究理事会）；Telenor Denmark OSS & Tech Analytics Denmark（Telenor丹麦OSS与技术分析丹麦）；Norwegian University of Science and Technology (NTNU)（挪威科学与技术大学（NTNU））；SINTEF Digital（SINTEF数字技术）；Hance
Anomaly Detection and Radio-frequency Interference Classification with Unsupervised Learning in Narrowband Radio Technosignature Searches
Breakthrough Listen（突破聆听）；University of California, Berkeley（加州大学伯克利分校）；Cornell University（康奈尔大学）；Cornell Center for Astrophysics（康奈尔天体物理中心）；Planetary Science, Cornell University（康奈尔行星科学）；University of Oxford（牛津大学）；SETI Institute（SETI研究所）；Harvard University（哈佛大学）；University of Malta（马耳他大学）；University of Manchester（曼彻斯特大学）
MSAD: A Deep Dive into Model Selection for Time series Anomaly Detection
DI ENS, ENS, PSL University, CNRS, Inria, Paris, France（巴黎大学）；The Ohio State University（俄亥俄州立大学）；Aristotle University Thessaloniki（塞萨洛尼基亚里士多德大学）；Université Paris Cité, IUF Paris, France（巴黎大学）
Anomaly-Driven Approach for Enhanced Prostate Cancer Segmentation
University of Amsterdam（阿姆斯特丹大学）；Department of Radiology, The Netherlands Cancer Institute（荷兰癌症研究所放射科）；GROW Research Institute for Oncology and Reproduction, Maastricht University（马斯特里赫特大学肿瘤学与生殖医学研究学院）
