The rapid growth of web content has spurred the widespread adoption of on-device AI assistants powered by large language models (LLMs). However, deploying and personalizing these assistants in real-world environments remains challenging due to limited annotation budgets and scarce on-device fine-tuning resources. Existing edge–cloud collaboration frameworks typically rely on costly cloud-based supervision or perform full-layer finetuning, leading to inefficiencies in both computation and adaptation. To address these limitations, we propose EcoTune, a budget-constrained framework for efficient edge–cloud collaborative adaptation. EcoTune jointly optimizes representative data selection for cloud annotation and selective on-device model adaptation within a unified closed-loop process. Specifically, it employs a multi-armed bandit–based strategy to identify highvalue user interactions for cloud supervision and a layer importance–driven adaptation mechanism to update only critical components of the small language model (SLM). This coordinated optimization enables dynamic, resource-efficient personalization under stringent annotation and tuning budgets. Experiments on real-world testbeds demonstrate that EcoTune achieves up to 20%-60% reduction in annotation costs and significantly lowers fine-tuning memory consumption compared to state-of-the-art baselines, providing a practical and scalable solution for personalized on-device LLMs.

Time series analysis is a fundamental data mining task that has made encouraging progress in many real-world scenarios. Supervised training methods based on empirical risk minimization have proven their effectiveness on specific tasks and datasets. However, the acquisition of well-annotated data is costly, and a large amount of unlabeled series data is under-utilized. Due to distributional shifts across various domains and different patterns of interest across multiple tasks. The problem of cross-domain multi-task migration remains a significant challenge. To address these problems, this article proposes a novel cross-domain approach based on Wave Quantization for Time Series (termed as WQ4TS), which can be combined with any advanced time series model and applied to diverse downstream tasks. Specifically, we transfer the data from different domains into a common spectral latent space and enable the model to learn the temporal pattern knowledge of different domains directly from the common space and utilize it for the inference of downstream tasks, thereby mitigating the challenge of heterogeneous migration. The establishment of spectral latent space brings at least three benefits, cross-domain migration capability thus adapting to zero- and few-shot scenarios without relying on priori knowledge, general compatible cross-domain framework without changing the existing model structure, and robust modeling capability thus achieving SOTA results in multiple downstream tasks. To demonstrate the effectiveness of the proposed approach, we conduct extensive experiments including three important tasks: forecasting, imputation, and classification. And three common real-world scenarios are simulated: full-data, few-shot, and zero-shot. The proposed WQ4TS achieves the best performance on 87.5% of all tasks. Concretely, WQ4TS achieved 25.8% and 44.1% improvements in MSE metric for few-shot and zero-shot forecasting tasks, respectively, and demonstrated excellent 24.9% increase in average accuracy on few-shot classification tasks. The source codes of WQ4TS are publicly available on https://github.com/Xiang-Kai/WQ4TS .

Time series analysis has a wide range of applications in the fields of weather forecasting, traffic management, fault detection, intelligent operation, etc. In the real world, time series typically consist of dynamic fluctuations and mixtures of periodicities, which bring challenges on modeling and analyzing their patterns. To overcome the complexities, a common approach is to decompose long-term time series into sub-components for easier analysis. Unlike conventional time series decomposition that decouples a series into the trend and seasonal parts, we proposed a novel triple decomposition method to decouple a long-term series into three components: trend-part, regular-part, and fluctuant-part. Notably, the third part is a particular component that represents the dynamic spectral fluctuation in time series with the formulation of spectrum gradient. Based on triple decomposition, we propose a novel task-general deep learning model called TS3Net for long-term series analysis. It introduces a temporal-frequency block (TF-Block) with a multi-branch structure to expand the time series into a 2D temporal-frequency distribution. Subsequently, deep representation can be learned by a vision architecture that captures the dynamic variations from the complex multi-periodic series. The decomposed components are processed by TS3Net individually, and their results are integrated to form the final result for time series analysis. We conduct extensive experiment based on six open datasets to evaluate the proposed method in comparison with 10 baselines. Numerical results show that TS3Net significantly outperforms the state-of-the-art methods on both time series forecasting and imputation tasks. The source codes of TS3Net are publicly available on https://github.com/Xiang-Kai/TS3Net.

Anomaly detection on multivariate time series (MTS) is an important research topic in data mining, which has a wide range of applications in information technology, financial management, manufacturing system, and so on. However, the state-of-the-art unsupervised deep learning models for MTS anomaly detection are vulnerable to noise and have poor performance on the training data containing anomalies. In this article, we propose a novel Self-Training based Anomaly Detection with Generative Adversarial Network (GAN) model called STAD-GAN to address the practical challenge. The STAD-GAN model consists of a generator-discriminator structure for adversarial learning and a neural network classifier for anomaly classification. The generator is learned to capture the normal data distribution, and the discriminator is learned to amplify the reconstruction error of abnormal data for better recognition. The proposed model is optimized with a self-training teacher-student framework, where a teacher model generates reliable high-quality pseudo-labels to train a student model iteratively with a refined dataset so that the performance of the anomaly classifier can be gradually improved. Extensive experiments based on six open MTS datasets show that STAD-GAN is robust to noise and achieves significant performance improvement compared to the state-of-the-art.

Multivariate time series (MTS) clustering is an important technique for discovering co-evolving patterns and interpreting group characteristics in many areas including economics, bioinformatics, data science, etc. Although time series clustering has been widely studied in the past decades, no enough attention has been paid to capture time-varying correlation patterns in MTS. In this article, we propose a novel clustering approach for MTS data based on time-varying features. We introduce a time-varying Gaussian Markov Random Fields (T-GMRF) model to describe the correlation structure between MTS variables, and formulate the time-varying feature extraction problem as a convex optimization problem, which can be solved by a T-GMRF learning algorithm based on random block coordinate descent. We further apply a principal component analysis (PCA) based method on GMRF sequences to obtain low-dimensional feature vectors, and adopt a multi-density based clustering approach to form the cluster assignments. We conduct extensive experiments to compare the proposed T-GMRF method with 11 clustering algorithms based on 33 open MTS datasets, which show that T-GMRF significantly outperforms the state-of-the-arts with performance improvement up to 16%-64.5% on a variety of clustering performance metrics. The source codes of T-GMRF are publicly available at GitHub.

Graph Convolutional Networks (GCNs) are popular for learning representation of graph data and have a wide range of applications in social networks, recommendation systems, etc. However, training GCN models for large networks is resource intensive and time consuming, which hinders them from real deployment. The existing GCN training methods intended to optimize the sampling of mini-batches for stochastic gradient descent to accelerate training process, which did not reduce the problem size and had limited reduction in computation complexity. In this paper, we argue that a GCN can be trained with a sampled subgraph to produce approximate node representations, which inspires us a novel perspective to accelerate GCN training via network sampling. To this end, we propose a label-centric cumulative sampling (LCS) framework for training GCNs for large graphs. The proposed method constructs a subgraph cumulatively based on probabilistic sampling, and trains the GCN model iteratively to generate approximate node representations. The optimality of LCS is theoretically guaranteed to minimize the bias during node aggregation procedure in GCN training. Extensive experiments based on four real-world network datasets show that the LCS framework accelerates the training for the state-of-the-art GCN models up to 17x without causing noteworthy model accuracy drop.

Katz centrality is a fundamental concept to measure the influence of a vertex in a social network. However, existing approaches to calculating Katz centrality in a large-scale network are unpractical and computationally expensive. In this article, we propose a novel method to estimate Katz centrality based on graph sampling techniques, which object to achieve comparable estimation accuracy of the state-of-the-arts with much lower computational complexity. Specifically, we develop a Horvitz–Thompson estimate for Katz centrality by using a multi-round sampling approach and deriving an unbiased mean value estimator. We further propose SAKE , a S ampling-based A lgorithm for fast K atz centrality E stimation. We prove that the estimator calculated by SAKE is probabilistically guaranteed to be within an additive error from the exact value. Extensive evaluation experiments based on four real-world networks show that the proposed algorithm can estimate Katz centralities for partial vertices with low sampling rate, low computation time, and it works well in identifying high influence vertices in social networks.

Influence maximization in social networks is the problem of finding a set of seed nodes in the network that maximizes the spread of influence under certain information prorogation model, which has become an important topic in social network analysis. In this paper, we show that conventional influence maximization algorithms cause uneven spread of influence among different attribute groups in social networks, which could lead to severer bias in public opinion dissemination and viral marketing. We formulate the balanced influence maximization problem to address the trade-off between influence maximization and attribute balance, and propose a sampling based solution to solve the problem efficiently. To avoid full network exploration, we first propose an attribute-based (AB) sampling method to sample attributed social networks with respect to preserving network structural properties and attribute proportion among user groups. Then we propose an attributed-based reverse influence sampling (AB-RIS) algorithm to select seed nodes from the sampled graph. The proposed AB-RIS algorithm runs efficiently with guaranteed accuracy, and achieves the trade-off between influence maximization and attribute balance. Extensive experiments based on four real-world social network datasets show that AB-RIS significantly outperforms the state-of-the-art approaches in balanced influence maximization.