论文检索

Mingyang Yu 0003, Xiahui Guo, Peng Chen 0038, Zhenkai Li, Yang Shu 0001

Time Series forecasting is critical in diverse domains such as weather forecasting, financial investment, and traffic management. While traditional numerical metrics like mean squared error (MSE) can quantify point-wise accuracy, they fail to evaluate the geometric structure of time series data, which is essential to understand temporal dynamics. To address this issue, we propose the time series Geometric Structure Index (TGSI), a novel evaluation metric that transforms time series into images to leverage their inherent two-dimensional geometric representations. However, since the image transformation process is non-differentiable, TGSI cannot be directly integrated as a training loss. We further introduce the Shape-Aware Temporal Loss (SATL), a multi-component loss function operating in the time series modality to bridge this gap and enhance structure modeling during training. SATL combines three components: a first-order difference loss that measures structural consistency through the MSE between first-order differences, a frequency domain loss that captures essential periodic patterns using the Fast Fourier Transform while minimizing noise, and a perceptual feature loss that measures geometric structure difference in time-series by aligning temporal features with geometric structure features through a pre-trained temporal feature extractor and time-series image autoencoder. Experiments across multiple datasets demonstrate that models trained with SATL achieve superior performance in both MSE and the proposed TGSI metrics compared to baseline methods, without additional computational cost during inference.

Yao Zhang, Ping Huang, Rui Zhang

The integration of Evolutionary Algorithms (EAs) and Reinforcement Learning (RL) offers a new paradigm for complex decision-making tasks, especially through methods that optimize actor populations and critics collaboratively (e.g., DBCEM-TD3, QD-RL), collectively referred to as the Actor-Population Critic (APC) framework. However, existing methods still face two major challenges: traditional feature inputs fail to fully capture the spatial relationships, and the design of the critic struggles to balance both the quality and diversity of the policies. To address these issues, we propose Multimodal Dual Population Evolutionary Reinforcement Learning (M-DPERL), which achieves breakthroughs through cross-modal feature augmentation and dual-population coevolution. On one hand, a feature-image bimodal input enhancement mechanism is proposed, which dynamically encodes environmental features into spatial heatmaps. On the other hand, this method introduces a critic population into APC and, for the first time, proposes a population-guided fitness metric to optimize the critic's ability to guide the actor population in balancing quality and diversity. Additionally, we design the Flow-Fix Dynamics (FFD) mechanism to regulate the update rhythm of the dual populations and alleviate the evolutionary chaos in their coevolution. The results across a series of MUJOCO tasks demonstrate that M-DPERL significantly outperforms the baselines, with a 19.2% improvement in sample efficiency and a 17.1% increase in final performance.

Kuiye Ding, Fanda Fan, Yao Wang, Ruijie Jian, Xiaorui Wang, Luqi Gong, Yishan Jiang, Chunjie Luo, Jianfeng Zhan

Multivariate Time Series Forecasting plays a key role in many applications. Recent works have explored using Large Language Models for MTSF to take advantage of their reasoning abilities. However, many methods treat LLMs as end-to-end forecasters, which often leads to a loss of numerical precision and forces LLMs to handle patterns beyond their intended design. Alternatively, methods that attempt to align textual and time series modalities within latent space frequently encounter alignment difficulty. In this paper, we propose to treat LLMs not as standalone forecasters, but as semantic guidance modules within a dual-stream framework. We propose DualSG, a dual -stream framework that provides explicit semantic guidance, where LLMs act as Semantic Guides to refine rather than replace traditional predictions. As part of DualSG, we introduce Time Series Caption, an explicit prompt format that summarizes trend patterns in natural language and provides interpretable context for LLMs, rather than relying on implicit alignment between text and time series in the latent space. We also design a caption-guided fusion module that explicitly models inter-variable relationships while reducing noise and computation. Experiments on real-world datasets from diverse domains show that DualSG consistently outperforms 15 state-of-the-art baselines, demonstrating the value of explicitly combining numerical forecasting with semantic guidance. The code is made available at https://github.com/BenchCouncil/DualSG

Wei Fan 0010, Jingru Fei, Dingyu Guo, Kun Yi 0001, Xiaozhuang Song, Haolong Xiang, Hangting Ye, Min Li 0007

Medical time series has been playing a vital role in real-world healthcare systems as valuable information in monitoring health conditions of patients. Traditional methods towards medical time series classification rely on handcrafted feature extraction and statistical methods; with the recent advancement of artificial intelligence, the machine learning and deep learning methods have become more popular. However, existing methods often fail to fully model the complex spatial dynamics under different scales, which ignore the dynamic multi-resolution spatial and temporal joint inter-dependencies. Moreover, they are less likely to consider the special baseline wander problem as well as the multi-view characteristics of medical time series, which largely hinders their prediction performance. To address these limitations, we propose a Multi-resolution Spatiotemporal Graph Learning framework, MedGNN, for medical time series classification. Specifically, we first propose to construct multi-resolution adaptive graph structures to learn dynamic multi-scale embeddings. Then, to address the baseline wander problem, we propose Difference Attention Networks to operate self-attention mechanisms on the finite difference for temporal modeling. Moreover, to learn the multi-view characteristics, we utilize the Frequency Convolution Networks to capture complementary information of medical time series from the frequency domain. In addition, we introduce the Multi-resolution Graph Transformer architecture to model the dynamic dependencies and fuse the information from different resolutions. Finally, we have conducted extensive experiments on multiple medical real-world datasets that demonstrate the superior performance of our method. Our Code is available at this repository: https://github.com/aikunyi/MedGNN.

Qihe Huang, Zhengyang Zhou, Kuo Yang 0002, Yang Wang 0015

The World Wide Web thrives on intelligent services that depend heavily on accurate time series forecasting to navigate dynamic and evolving environments. Due to the partially-observed nature of real world, exclusively focusing on the target of interest, so-called endogenous variables, is insufficient for accurate forecasting, especially in web systems that are susceptible to external influences. Thus, utilizing exogenous variables to harness external information, i.e., forecasting with exogenous variable (FEV), is imperative. Nevertheless, as the external environment is complex and ever-evolving, inadequately capturing external influences can even lead to learning spurious correlations and invalid prediction. Fortunately, recent studies have demonstrated that large language models (LLMs) exhibit exceptional recognition capabilities across open real-world systems, including a deep understanding of exogenous environments. However, it is difficult to directly apply LLMs for FEV due to challenges of task activation, exogenous knowledge extraction, and feature space alignment. In this work, we devise ExoLLM, an LLM-driven method to sufficiently utilize Exogenous variables for time series forecasting. We begin by Meta-task Instruction to activate the knowledge transfer of LLM from natural language processing to FEV. To comprehensively understand the intricate and hierarchical influences of exogenous variables, we propose Multi-grained Prompts, encompassing diverse external influences, including natural attributes, trend correlations, and period relationships between two types of variables. Additionally, a Dual TS-Text Attention is devised to bridge the feature gap between text and numeric data in LLM. Evaluation on real-world datasets demonstrates ExoLLM's superiority in exploiting exogenous information for forecasting with open-world language knowledge.

Yixiao Teng, Jiamei Lv, Ziping Wang, Yi Gao 0001, Wei Dong 0001

Blockchain-based distributed storage systems offer enhanced security, transparency, and lower costs compared to traditional centralized storage, making them ideal for peer-to-peer collaboration. However, with the trend towards the Web of Things (WoT), lower transaction speeds and higher computational requirements limit their access to high-density data such as IoT. To address this, we propose TimeChain, an efficient off-chain blockchain storage system for IoT time series data. TimeChain batches discrete time series data, storing only the hash value of each batch on-chain while keeping the complete data off-chain. This significantly reduces storage overhead on the blockchain and storage latency by 37.4 times. TimeChain adopts an adaptive packaging mechanism to reduce the additional latency in range queries by converting the batch processing problem into a graph partitioning problem. To reduce the overhead of node selection, TimeChain integrates a node selection mechanism based on consensus protocol, combining node selection and consensus processes together. TimeChain also proposes a Locality-Sensitive Hashing tree-based data integrity verification mechanism to reduce transmission size. Our evaluation shows a reduction in query latency by 64.6% and storage latency by 35.3% compared to existing systems.

Ruichu Cai, Zhifan Jiang, Kaitao Zheng, Zijian Li 0001, Weilin Chen 0001, Xuexin Chen, Yifan Shen 0004, Guangyi Chen 0002, Zhifeng Hao 0004, Kun Zhang 0001

Multi-modal time series data is common in web technologies like the Internet of Things (IoT). Existing methods for multi-modal time series representation learning aim to disentangle the modality-shared and modality-specific latent variables. Although achieving notable performances on downstream tasks, they usually assume an orthogonal latent space. However, the modality-specific and modality-shared latent variables might be dependent on real-world scenarios. Therefore, we propose a general generation process, where the modality-shared and modality-specific latent variables are dependent, and further develop a Multi-modAl TEmporal Disentanglement (MATE) model. Specifically, our MATE model is built on a temporally variational inference architecture with the modality-shared and modality-specific prior networks for the disentanglement of latent variables. Furthermore, we establish identifiability results to show that the extracted representation is disentangled. More specifically, we first achieve the subspace identifiability for modality-shared and modality-specific latent variables by leveraging the pairing of multi-modal data. Then we establish the component-wise identifiability of modality-specific latent variables by employing sufficient changes of historical latent variables. Extensive experimental studies on 12 datasets show a general improvement in different downstream tasks, highlighting the effectiveness of our method in real-world scenarios.

Tomoyoshi Kimura, Xinlin Li, Osama A. Hanna, Yatong Chen 0001, Yizhuo Chen, Denizhan Kara, Tianshi Wang 0002, Jinyang Li 0004, Xiaomin Ouyang, Shengzhong Liu 等

Standard multimodal self-supervised learning (SSL) algorithms regard cross-modal synchronization as implicit supervisory labels during pretraining, thus posing high requirements on the scale and quality of multimodal samples. These constraints significantly limit the performance of sensing intelligence in IoT applications, as the heterogeneity and the non-interpretability of time-series signals result in abundant unimodal data but scarce high-quality multimodal pairs. This paper proposes InfoMAE, a cross-modal alignment framework that tackles the challenge of multimodal pair efficiency under the SSL setting by facilitating efficient cross-modal alignment of pretrained unimodal representations. InfoMAE achieves efficient cross-modal alignment with limited data pairs through a novel information theory-inspired formulation that simultaneously addresses distribution-level and instance-level alignment. Extensive experiments on two real-world IoT applications are performed to evaluate InfoMAE's pairing efficiency to bridge pretrained unimodal models into a cohesive joint multimodal model. InfoMAE enhances downstream multimodal tasks by over 60% with significantly improved multimodal pairing efficiency. It also improves unimodal task accuracy by an average of 22%.

Xiao Tan 0005, Yangyang Shen, Yan Zhang 0100, Jingwen Shao, Dian Shen, Meng Wang 0009, Beilun Wang

Dynamic Dependence Network (DDN) inference is crucial for understanding evolving relationships in multimodal time series web data, with broad applications in fields like medical and financial network analysis. The inherent dynamic nature, temporal continuity, and heterogeneous data sources in multimodal time series data pose three fundamental challenges: computational efficiency, prediction stability and robustness, and modality quality disparity. Previous methods, generally lacking utilization of multiple modalities, either struggle with computational efficiency due to the time-intensive manual hyperparameter tuning, or compromise prediction stability and robustness by neglecting temporal coherence. To address these challenges, we propose a Normalized mutual information-driven Tuning-free Dynamic Dependence Network inference method for multimodal data, namely NoTeNet. NoTeNet provides a promising paradigm that can integrate two different data modalities to enhance prediction accuracy. It uses normalized mutual information transforms noisy auxiliary data into relationship matrices and employs a kernel function for smooth temporal estimation. Additionally, NoTeNet significantly reduces the need for manual hyperparameter adjustments, offering a tuning-free approach with theoretical guarantees. On various synthetic datasets and real-world data, NoTeNet demonstrates superior prediction accuracy and efficiency without the need for hyperparameter tuning, making it potential for a wide range of web data applications.

Alexandra Khirianova, Ekaterina Solodneva, Andrey Pudovikov, Sergey Osokin, Egor Samosvat, Yuriy Dorn, Alexander Ledovsky, Yana Zenkova

The optimization of bidding strategies for online advertising slot auctions presents a critical challenge across numerous digital marketplaces. A significant obstacle to the development, evaluation, and refinement of real-time autobidding algorithms is the scarcity of comprehensive datasets and standardized benchmarks. To address this deficiency, we present an auction benchmark encompassing the two most prevalent auction formats. We implement a series of robust baselines on a novel dataset, addressing the most salient Real-Time Bidding (RTB) problem domains: budget pacing uniformity and Cost Per Click (CPC) constraint optimization. This benchmark provides a user-friendly and intuitive framework for researchers and practitioners to develop and refine innovative autobidding algorithms, thereby facilitating advancements in the field of programmatic advertising. The implementation and additional resources can be accessed at the following repository https://github.com/avito-tech/bat-autobidding-benchmark, https://doi.org/10.5281/zenodo.14794182.

Yuanmin Huang 0001, Mi Zhang 0001, Zhaoxiang Wang, Wenxuan Li, Min Yang 0002

Time series classification (TSC) is a cornerstone of modern web applications, powering tasks such as financial data analysis, network traffic monitoring, and user behavior analysis. In recent years, deep neural networks (DNNs) have greatly enhanced the performance of TSC models in these critical domains. However, DNNs are vulnerable to backdoor attacks, where attackers can covertly implant triggers into models to induce malicious outcomes. Existing backdoor attacks targeting DNN-based TSC models remain elementary. In particular, early methods borrow trigger designs from computer vision, which are ineffective for time series data. More recent approaches utilize generative models for trigger generation, but at the cost of significant computational complexity. In this work, we analyze the limitations of existing attacks and introduce an enhanced method, FreqBack. Drawing inspiration from the fact that DNN models inherently capture frequency domain features in time series data, we identify that improper perturbations in the frequency domain are the root cause of ineffective attacks. To address this, we propose to generate triggers both effectively and efficiently, guided by frequency analysis. FreqBack exhibits substantial performance across five models and eight datasets, achieving an impressive attack success rate of over 90%, while maintaining less than a 3% drop in model accuracy on clean data.

Yongzheng Xie, Hongyu Zhang 0002, Muhammad Ali Babar 0001

Multivariate time series anomaly detection is essential for failure management in web application operations, as it directly influences the effectiveness and timeliness of implementing remedial or preventive measures. This task is often framed as a semi-supervised learning problem, where only normal data are available for model training, primarily due to the labor-intensive nature of data labeling and the scarcity of anomalous data. Existing semi-supervised methods often detect anomalies by capturing intra-variate temporal dependencies and/or inter-variate relationships to learn normal patterns, flagging timestamps that deviate from these patterns as anomalies. However, these approaches often fail to capture salient intra-variate temporal and inter-variate dependencies in time series due to their focus on excessively fine granularity, leading to suboptimal performance. In this study, we introduce MtsCID, a novel semi-supervised multivariate time series anomaly detection method. MtsCID employs a dual network architecture: one network operates on the attention maps of multi-scale intra-variate patches for coarse-grained temporal dependency learning, while the other works on variates to capture coarse-grained inter-variate relationships through convolution and interaction with sinusoidal prototypes. This design enhances the ability to capture the patterns from both intra-variate temporal dependencies and inter-variate relationships, resulting in improved performance. Extensive experiments across seven widely used datasets demonstrate that MtsCID achieves performance comparable or superior to state-of-the-art benchmark methods.

Yan Liu 0002

Recent advancements in deep learning and artificial intelligence have driven significant progress in time series modeling and analysis. On one hand, researchers seek breakthroughs in performance on classical tasks such as forecasting, anomaly detection, classification, etc. On the other hand, it is intriguing to explore the potential for answering more complex inference and reasoning tasks from time series. In this keynote, I will examine the pathways toward foundation models for time series and discuss future research directions in this rapidly evolving field. The remarkable success of foundation models in natural language processing - exemplified by Generative Pre-trained Transformers (GPT) - suggests their potential to revolutionize time series analysis. I will introduce our recent efforts along this direction, including TEMPO, a novel framework designed to learn effective time series representations by leveraging two key inductive biases: one is explicit decomposition of trend, seasonal, and residual components, and the second is prompt-based distribution adaptation for diverse time series types. Beyond representation learning, practical applications demands advanced reasoning capabilities with multi-step time series inference task, requiring both compositional reasoning and computational precision. To tackle this challenge, I will discuss TS-reasoner, a program-aided inference agent that integrates large language models (LLMs) with structured execution pipelines, in-context learning, and self-correction mechanisms. I will discuss a new benchmark dataset and evaluation framework to systematically assess multi-step time series reasoning. By bridging deep learning advances with structured reasoning, I will highlight the next frontier in time series research, i.e., developing foundation models that enhance forecasting performance, generative models, and reasoning capabilities from time series across diverse applications.

Sanjay Purushotham, Dongjin Song, Qingsong Wen, Jun Huan, Yuxuan Liang 0002, Cong Shen 0001, Stefan Zohren, Yuriy Nevmyvaka

Time series data is now pervasive across domains such as healthcare, finance, entertainment, and transportation, driven by advances in sensing technologies that enable continuous data collection. The resulting increase in data volume and complexity poses significant challenges to traditional analysis methods, calling for the development of advanced, interdisciplinary approaches to temporal data mining. This workshop aims to: (1) identify key challenges in learning from time series data, including irregular sampling, spatiotemporal dependencies, and uncertainty quantification; (2) explore recent advances in algorithmic, statistical, theoretical, and systems-based solutions-ranging from classical methods to emerging techniques involving large language models (LLMs); and (3) foster collaboration by highlighting open problems and novel research directions in time series analysis. Bridging theory and practice, the workshop provides a platform for researchers and practitioners from academia, industry, and government to exchange ideas, discuss technical challenges, and showcase practical applications. Contributions from related areas such as AI, machine learning, data science, and statistics are strongly encouraged.

Abraham Bagherjeiran, Nemanja Djuric, Kuang-chih Lee, Linsey Pang, Vladan Radosavljevic, Suju Rajan

The digital advertising field has always had challenging ML problems, learning from petabytes of data that is highly imbalanced, reactivity times in the milliseconds, and more recently compounded with the complex user's path to purchase across devices, across platforms, and even online/real-world behavior. The AdKDD workshop continues to be a forum for researchers in advertising, during and after KDD. Our website, which hosts slides and abstracts, continues to receive a large number of monthly visits and active users. In surveys during AdKDD 2019 and 2020, over 60% agreed that AdKDD is the reason they attended KDD, and over 90% indicated they would attend next year. The 2025 edition is particularly timely because of the increasing application of Graph-based NN and Generative AI models in advertising. Coupled with privacy-preserving initiatives, such as those enforced by GDPR or CCPA, the future of computational advertising is at an interesting crossroads. For this edition, we plan to solicit papers that span the spectrum of deep user understanding while remaining privacy-preserving. In addition, we will seek papers that discuss fairness in the context of advertising, to what extent does hyper-personalization work, and whether the ad industry as a whole needs to think through more effective business models such as incrementality. We have hosted several academic and industry luminaries as keynote speakers and have found our invited speaker series hosting expert practitioners to be an audience favorite. We will continue fielding a diverse set of keynote speakers and invited talks for this edition as well. As with past editions, we hope to motivate researchers in this space to think not only about the ML aspects but also to spark conversations about the societal impact of online advertising.

Aidong Zhang 0001, Vipin Kumar 0001, Yan Liu 0002

The past decade has been an inspiring time for artificial intelligence (AI) research. AI systems have transformed norms and practices across industries and have permeated the fabric of human society. Moreover, AI is ushering in a transformative technological age by making remarkable breakthroughs in a number of scientific fields such as protein structure prediction and medical imaging. There is increasing consensus in the wider scientific community that AI is poised to disrupt science by unlocking entirely new approaches, driving new scientific inquiry, and enabling greater scientific leaps with far-reaching social consequences. However, there are substantial barriers preventing science from realizing that potential, and addressing these barriers will require support for advances in AI methods and the adoption of these methods in routine scientific research. In this special day at KDD 2025, we host a series of talks by distinguished researchers on AI for science.

Quentin Nater, Mourad Khayati, Philippe Cudré-Mauroux

Although missing gaps are common in time series data, most existing imputation libraries have a narrow focus. They typically rely on a limited set of techniques and make overly simplistic assumptions about the nature of missing data. Consequently, they fail to model the true intricate complexity of real-world time series. To overcome these challenges, we developed ImputeGAP, a versatile and comprehensive library for time series imputation. ImputeGAP supports a wide range of imputation algorithms and modular missing data simulation, catering to datasets with varying characteristics. It also streamlines imputation analysis with features such as automated hyperparameter tuning, benchmarking, explainability, and downstream evaluation. In this tutorial, we will provide an engaging hands-on tutorial where you will learn time series imputation using the powerful Python library, ImputeGAP. The session is divided into two parts. In the first part, we will dive into building an end-to-end imputation workflow with the library, where you will explore real-world missingness patterns simulation, leverage automated tuning for optimal imputation, and benchmark imputation techniques-all with extensive customization options. In the second part, we will unlock advanced functionalities, including assessing the impact of imputation on downstream analytics and understanding how time series features influence imputation outcomes. Whether you are a researcher or practitioner, this tutorial will provide you with the expertise to handle missing data in time series. The ImputeGAP library is accessible at: https://imputegap.readthedocs.io.

Kun Yi 0001, Qi Zhang 0020, Wei Fan 0010, Longbing Cao, Shoujin Wang, Hui He, Guodong Long, Liang Hu 0004, Qingsong Wen, Hui Xiong 0001

Recently, frequency transformation (FT) has been increasingly incorporated into deep learning models to significantly enhance state-of-the-art accuracy and efficiency in time series analysis. The advantages of FT, such as high efficiency and a global view, have been rapidly explored and exploited in various time series tasks and applications, demonstrating the promising potential of FT as a new deep learning paradigm for time series analysis. Despite the growing attention and the proliferation of research in this emerging field, there is currently a lack of a systematic review and in-depth analysis of deep learning-based time series models with FT. It is also unclear why FT can enhance time series analysis and what its limitations are in the field. To address these gaps, we present a comprehensive review that systematically investigates and summarizes the recent research advancements in deep learning-based time series analysis with FT. Specifically, we explore the primary approaches used in current models that incorporate FT, the types of neural networks that leverage FT, and the representative FT-equipped models in deep time series analysis. We propose a novel taxonomy to categorize the existing methods in this field, providing a structured overview of the diverse approaches employed in incorporating FT into deep learning models for time series analysis. Finally, we highlight the advantages and limitations of FT for time series modeling and identify potential future research directions that can further contribute to the community of time series analysis.

Ran Xu 0002, Patrick Jiang, Linhao Luo, Cao Xiao, Adam Cross, Shirui Pan, Jimeng Sun 0001, Carl Yang 0001

In recent years, the landscape of digital biomedicine and healthcare has been reshaped due to the disruptive breakthroughs in AI-facilitated by tremendous data and high-performance computers, large language models (LLMs) have transformed information technology from accessing data to performing analytical tasks. While demonstrating unprecedented capabilities, LLMs have been found unreliable in tasks requiring factual knowledge and rigorous reasoning. Biomedicine and healthcare, as an important vertical domain rapidly benefitting from progress in AI, necessitates strict requirements on the accuracy, controllability, and interpretability of analytical models, posing critical challenges for LLMs. Despite recent studies addressing the hallucination problem of LLMs, research on empowering LLMs with the ability to plan, reason, and ground with explicit knowledge has also started to prosper, especially in the biomedicine and healthcare domain. On the other hand, biomedical data are enormous and notoriously complex, coming from various sources (e.g., biomedical knowledge bases, online literature, and hospitals) and bearing various modalities (e.g., tables, texts, images and time-series). Healthcare professionals have spent decades collecting, cleaning, and curating various types of data. The processes are extremely costly, producing various datasets with different data schemas, coding systems, and quality standards, many privately owned by the creators, making their integrative analysis and utilization through unified AI techniques still rather challenging. The generalizability of LLMs across different types of data endow them strong promises in automating the processing of large-scale complex healthcare data such as into unified knowledge graphs (KGs). Our goal in this survey is to systematically investigate and summarize recent studies on the unification of LLMs and KGs, towards fully utilizing the value of complex data, unleashing the power of generative AI, and expediting next-generation AI for biomedicine and healthcare applications.

John Paparrizos, Paul Boniol, Qinghua Liu, Themis Palpanas

Recent advances in data collection technology, accompanied by the ever-rising volume and velocity of streaming data, underscore the vital need for time series analytics. In this regard, time-series anomaly detection has been an important activity, entailing various applications in fields such as cyber security, financial markets, law enforcement, and health care. While traditional literature on anomaly detection is centered on statistical measures, the increasing number of machine learning algorithms in recent years call for a structured, general characterization of the research methods for time-series anomaly detection. In this paper, we present a process-centric taxonomy for time-series anomaly detection methods, systematically categorizing traditional statistical approaches and contemporary machine learning techniques. Beyond this taxonomy, we conduct a meta-analysis of the existing literature to identify broad research trends. Given the absence of a one-size-fits-all anomaly detector, we also introduce emerging trends for time-series anomaly detection. Furthermore, we review commonly used evaluation measures and benchmarks, followed by an analysis of benchmark results to provide insights into the impact of different design choices on model performance. Through these contributions, we aim to provide a holistic perspective on time-series anomaly detection and highlight promising avenues for future investigation.