The World Wide Web increasingly relies on intelligent services that require accurate time series forecasting, from urban mobility platforms to adaptive web-based decision systems. In practice, building effective forecasting models typically requires abundant high-quality data, which may not always be available in all cities due to sensing limitations or data sparsity. To address this challenge, transfer learning methods aim to transfer knowledge from data-rich source cities to data-scarce target cities. However, source and target data distributions are often not identical: while some patterns from source cities may be beneficial, others can be irrelevant or even misleading. Existing transfer learning methods generally train the target model using all available source data without explicitly distinguishing between useful and non-useful knowledge, which may hinder performance. In this work, we propose xRAG4TS, a novel framework that integrates Retrieval-Augmented Generation (RAG) with Large Language Models (LLMs) for cross-city time series forecasting. xRAG4TS introduces a Cross-City Selective Retriever Module that filters semantically relevant historical patterns and documents from data-rich source cities, and incorporates them as structured prompts in an LLM Inference Module to guide forecasting in data-scarce target cities. By enabling selective, interpretable, and context-aware knowledge transfer, our method enhances robustness and scalability in web-oriented spatio-temporal applications. Extensive experiments on real-world cross-city datasets demonstrate that xRAG4TS significantly outperforms state-of-the-art baselines, highlighting its potential for powering adaptive and trustworthy web services under severe data scarcity.

The widespread adoption of location-based services has led to the generation of vast amounts of mobility data, providing significant opportunities to model user movement dynamics within urban environments. Recent advancements have focused on adapting Large Language Models (LLMs) for mobility analytics. However, existing methods face two primary limitations: inadequate semantic representation of locations (i.e., discrete IDs) and insufficient modeling of mobility signals within LLMs (i.e., single templated instruction fine-tuning). To address these issues, we propose QT-Mob, a novel framework that significantly enhances LLMs for mobility analytics. QT-Mob introduces a location tokenization module that learns compact, semantically rich tokens to represent locations, preserving contextual information while ensuring compatibility with LLMs. Furthermore, QT-Mob incorporates a series of complementary fine-tuning objectives that align the learned tokens with the internal representations in LLMs, improving the model's comprehension of sequential movement patterns and location semantics. The proposed QT-Mob framework not only enhances LLMs' ability to interpret mobility data but also provides a more generalizable approach for various mobility analytics tasks. Experiments on three real-world dataset demonstrate the superior performance in both next-location prediction and mobility recovery tasks, outperforming existing deep learning and LLM-based methods.

Missing values pose a formidable obstacle in multivariate time series analysis. Existing imputation methods rely on entangled representations that struggle to simultaneously capture multiple orthogonal time-series patterns, leading to suboptimal performance and limited interpretability. Meanwhile, requiring the entire data span as input renders these models impractical for long time series. To address these issues, we propose${\sf TIDER}$and its enhanced version,${\sf AdaTIDER}$.${\sf TIDER}$employs low-rank matrix factorization and disentangled temporal representations to model intricate dynamics like trend, seasonality, and local bias. However,${\sf TIDER}$is limited to single-period modeling and does not explicitly capture dependencies between channels. To overcome these limitations,${\sf AdaTIDER}$incorporates adaptive cross-channel dependency modeling and multi-period seasonality representations. These advancements enable it to dynamically capture variable relationships and complex multi-period patterns, significantly enhancing imputation accuracy and interpretability, while maintaining${\sf TIDER}$'s scalability. Extensive experiments on real-world datasets validate the superiority of our models in imputation accuracy, scalability, interpretability, and robustness.

With the development of astronomical facilities, large-scale time series data observed by these facilities is being collected. Analyzing anomalies in these astronomical observations is crucial for uncovering potential celestial events and physical phenomena, thus advancing the scientific research process. However, existing time series anomaly detection methods fall short in tackling the unique characteristics of astronomical observations where each star is inherently independent but interfered by random concurrent noise, resulting in a high rate of false alarms. To overcome the challenges, we propose AERO, a novel two-stage framework tailored for unsupervised anomaly detection in astronomical observations. In the first stage, we employ a Transformer-based encoder-decoder architecture to learn the normal temporal patterns on each variate (i.e., star) in alignment with the characteristic of variate independence. In the second stage, we enhance the graph neural network with a window-wise graph structure learning to tackle the occurrence of concurrent noise characterized by spatial and temporal randomness. In this way, AERO is not only capable of distinguishing normal temporal patterns from potential anomalies but also effectively differentiating concurrent noise, thus decreasing the number of false alarms. We conducted extensive experiments on three synthetic datasets and three real-world datasets. The results demonstrate that AERO outperforms the compared baselines. Notably, compared to the state-of-the-art model, AERO improves the F1-score by up to 8.76% and 2.63% on synthetic and real-world datasets respectively.

Time series forecasting is essential for our daily activities and precise modeling of the complex correlations and shared patterns among multiple time series is essential for improving forecasting performance. Spatial-Temporal Graph Neural Networks (STGNNs) are widely used in multivariate time series forecasting tasks and have achieved promising performance on multiple real-world datasets for their ability to model the underlying complex spatial and temporal dependencies. However, existing studies have mainly focused on datasets comprising only a few hundred sensors due to the heavy computational cost and memory cost of spatial-temporal GNNs. When applied to larger datasets, these methods fail to capture the underlying complex spatial dependencies and exhibit limited scalability and performance. To this end, we present a Scalable Adaptive Graph Diffusion Forecasting Network (SAGDFN) to capture complex spatial-temporal correlation for large-scale multivariate time series and thereby, leading to exceptional performance in multivariate time series forecasting tasks. The proposed SAGDFN is scalable to datasets of thousands of nodes without the need of prior knowledge of spatial correlation. Extensive experiments demonstrate that SAGDFN achieves comparable performance with state-of-the-art baselines on one real-world dataset of 207 nodes and outperforms all state-of-the-art baselines by a significant margin on three real-world datasets of 2000 nodes.

Multi-agent trajectory prediction, as a critical task in modeling complex interactions of objects in dynamic systems, has attracted significant research attention in recent years. Despite the promising advances, existing studies all follow the assumption that data distribution observed during model learning matches that encountered in real-world deployments. However, this assumption often does not hold in practice, as inherent distribution shifts might exist in the mobility patterns for deploy-ment environments, thus leading to poor domain generalization and performance degradation. Consequently, it is appealing to leverage trajectories from multiple source domains to mitigate such discrepancies for multi-agent trajectory prediction task. However, the development of multi-source domain generalization in this task presents two notable issues: (1) negative transfer; (2) inadequate modeling for external factors. To address these issues, we propose a new causal formulation to explicitly model four types of features: domain-invariant and domain-specific features for both the focal agent and neighboring agents. Building upon the new formulation, we propose AdapTraj, a multi-source domain generalization framework specifically tailored for multi-agent trajectory prediction. AdapTraj serves as a plug-and-play module that is adaptable to a variety of models. Extensive experiments on four datasets with different domains demonstrate that AdapTraj consistently outperforms other baselines by a substantial margin.

With the expansion of modern database services, multi-user access has become a crucial feature in various practical application scenarios, including enterprise applications and e-commerce platforms. However, if multiple users submit queries within a short time frame, it can result in potential issues such as redundant computation and query concurrency. Unfortunately, most existing multi-query optimization methods, which aim to enhance query processing efficiency, have not adequately addressed these two problems, especially in the setting where multiple queries are being executed concurrently. To this end, we propose a novel method named Lemo for the multi-query optimization problem. Specifically, we propose a novel value network to predict latencies of concurrent queries as the foundation model for query plan generation. Furthermore, we introduce a shared buffer manager component to cache the intermediate results of sub-queries. The shared buffer manager applies a novel replacement policy to maintain the cached buffer with the objective of maximizing the opportunity for the reuse of the cached sub-queries. Based on the shared buffer, our proposed value network can incorporate the cached results into cost estimation to further guide Lemo in generating query plans, thus avoiding redundant computation. Lemo has been integrated into PostgreSQL and experiments conducted on real datasets with PostgreSQL show that it outperforms all the baselines in efficiency.

The proliferation of trajectory data has facilitated various applications in urban spaces, such as travel time estimation, traffic monitoring, and flow prediction. These applications require a substantial volume of high-quality trajectories as the prerequisite to achieve effective performance. Unfortunately, a large number of real-world trajectories are inevitably collected in unsatisfactory quality due to device constraints. To address this issue, previous studies have proposed numerous trajectory recovery methods to augment the quality of such trajectories, thereby ensuring the performance of related applications. However, these methods all assume the awareness of the recovery positions in advance, which is a condition not always available in practice. In this paper, we discard this strong assumption and focus on trajectory recovery with irregular time intervals as a more prevalent setting in downstream scenarios. We propose a novel framework, called TERI, to tackle trajectory recovery without prior information in a two-stage process, where recovery positions are first detected, followed by the imputation of the missing data points. In each stage, TERI framework deploys a model named RETE, which is based on Transformer encoder architecture enhanced by novel designs to boost the performance for the new problem setting. Specifically, RETE features a learnable Fourier encoding module to better model spatial and temporal correlations, and integrates collective transition pattern learning and trajectory contrastive learning to effectively capture sequential transition patterns. Extensive experiments on three real-world datasets demonstrate that TERI consistently outperforms all the baselines by a significant large margin.

There has been sustained interest from both academia and industry throughout the years due to the importance and practicability of recommendation systems. However, several recent papers have pointed out critical issues with the evaluation process in recommender systems. Likewise, this paper takes an in-depth look at a fundamental but often neglected aspect of the evaluation procedure, i.e. the datasets themselves. To do so, we adopt a systematic and comprehensive approach to understand the datasets used for implicit feedback based top-K recommendation. We start by examining recent papers from top-tier conferences to find out how different datasets have been utilised thus far. Next, we look at the characteristics of these datasets to understand their similarities and differences. Finally, we conduct an empirical study to determine whether the choice of datasets used for evaluation can influence the observations and/or conclusions obtained. Our findings suggest that greater attention needs to be paid to the selection process of datasets used for evaluating recommender systems in order to improve the robustness of the obtained results.

In this work, we propose a robust road network representation learning framework called Toast, which comes to be a cornerstone to boost the performance of numerous demanding transport planning tasks. Specifically, we first propose a traffic context aware skip-gram module to incorporate auxiliary tasks of predicting the traffic context of a target road segment. Furthermore, we propose a trajectory-enhanced Transformer module that utilizes trajectory data to extract traveling semantics on road networks. Apart from obtaining effective road segment representations, this module also enables us to obtain the route representations. With these two modules, we can learn representations which can capture multi-faceted characteristics of road networks to be applied in both road segment based applications and trajectory based applications. Last, we design a benchmark containing four typical transport planning tasks to evaluate the usefulness of Toast and comprehensive experiments verify that Toast consistently outperforms the state-of-the-art baselines across all tasks.

Recommender systems play a vital role in modern web services. In a typical recommender system, we are given a set of observed user-item interaction records and seek to uncover the hidden behavioral patterns of users from these historical interactions. By exploiting these hidden patterns, we aim to discover users’ personalized tastes and recommend them new items. Among various types of recommendation methods, the latent factor collaborative filtering models have dominated the field. In this paper, we develop a unified view for the existing latent factor models from a probabilistic perspective. The unified framework enables us to discern the underlying connections of different latent factor models and deepen our understandings of their advantages and limitations. In particular, we observe that the loss functions adopted by the existing models are oblivious to the geometry induced by the item-similarity. To address this, we propose a novel model—SinkhornCF—based on Sinkhorn divergence. To address the challenge of the expensive computational cost of Sinkhorn divergence, we also propose new techniques to enable the resulting model to be able to scale to large datasets. Its effectiveness is verified on two real-world recommendation datasets.

As a fundamental component in location-based services, inferring the relationship between points-of-interests (POIs) is very critical for service providers to offer good user experience to business owners and customers. Most of the existing methods for relationship inference are not targeted at POI, thus failing to capture unique spatial characteristics that have huge effects on POI relationships. In this work we propose PRIM to tackle POI relationship inference for multiple relation types. PRIM features four novel components, including a weighted relational graph neural network, category taxonomy integration, a self-attentive spatial context extractor, and a distance-specific scoring function. Extensive experiments on two real-world datasets show that PRIM achieves the best results compared to state-of-the-art baselines and it is robust against data sparsity and is applicable to unseen cases in practice.

Mobility prediction, which is to predict where a user will arrive based on the user's historical mobility records, has attracted much attention. We argue that it is more useful to know not only where but also when a user will arrive next in many scenarios such as targeted advertising and taxi service. In this paper, we propose a novel context-aware deep model called DeepJMT for jointly performing mobility prediction (to know where) and time prediction (to know when). The DeepJMT model consists of (1) a hierarchical recurrent neural network (RNN) based sequential dependency encoder, which is more capable of capturing a user's mobility regularities and temporal patterns compared to vanilla RNN based models; (2) a spatial context extractor and a periodicity context extractor to extract location semantics and the user's periodicity, respectively; and (3) a co-attention based social & temporal context extractor which could extract the mobility and temporal evidence from social relationships. Experiments conducted on three real-world datasets show that DeepJMT outperforms the state-of-the-art mobility prediction and time prediction methods.