Wei-Chieh Huang, Henry Peng Zou, Yaozu Wu, Dongyuan Li, Yankai Chen, Weizhi Zhang, Yangning Li, Angelo Zangari, Jizhou Guo, Chunyu Miao, Liancheng Fang, Langzhou He, Yinghui Li, Renhe Jiang, Philip S. Yu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Ding Xia, Xinyue Gui, Mark Colley, Fan Gao, Zhongyi Zhou, Dongyuan Li, Renhe Jiang, Takeo Igarashi. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Traffic prediction remains a key challenge in spatio-temporal data mining, despite progress in deep learning. Accurate forecasting is hindered by the complex influence of external factors such as traffic accidents and regulations, often overlooked by existing models due to limited data integration. To address these limitations, we present two enriched traffic datasets from Tokyo and California, incorporating traffic accident and regulation data. Leveraging these datasets, we propose ConFormer (Conditional Transformer), a novel framework that integrates graph propagation with guided normalization layer. This design dynamically adjusts spatial and temporal node relationships based on historical patterns, enhancing predictive accuracy. Our model surpasses the state-of-the-art STAEFormer in both predictive performance and efficiency, achieving lower computational costs and reduced parameter demands. Extensive evaluations demonstrate that ConFormer consistently outperforms mainstream spatio-temporal baselines across multiple metrics, underscoring its potential to advance traffic prediction research. The code is released in https://github.com/Dreamzz5/ConFormer.

Dynamic graph clustering aims to detect and track time-varying clusters in dynamic graphs, revealing how complex real-world systems evolve over time. However, existing methods are predominantly black-box models. They lack interpretability in their clustering decisions and fail to provide semantic explanations of why clusters form or how they evolve, severely limiting their use in safety-critical domains such as healthcare or transportation. To address these limitations, we propose an end-to-end interpretable framework that maps continuous graph embeddings into discrete semantic concepts through learnable prototypes. Specifically, we first decompose node representations into orthogonal role and clustering subspaces, so that nodes with similar roles (e.g., hubs, bridges) but different cluster affiliations can be properly distinguished. We then introduce five node role prototypes (Leader, Contributor, Wanderer, Connector, Newcomer) in the role subspace as semantic anchors, transforming continuous embeddings into discrete concepts to facilitate LLM understanding of node roles within communities. Finally, we design a hierarchical LLM reasoning mechanism to generate both clustering results and natural language explanations, while providing consistency feedback as weak supervision to refine node representations. Experimental results on four synthetic and six real-world benchmarks demonstrate the effectiveness, interpretability, and robustness of DyG-RoLLM. Code is available at https: //github.com/Clearloveyuan/DyG-RoLLM.

The development of the Internet of Things (IoT) has enhanced smart city services for traffic monitoring, leading to numerous schemes for accurate flow prediction based on traffic sensors. However, existing approaches primarily capture spatio-temporal (ST) dependencies from traffic graphs and train their models using randomly ordered data. This overlooks the fact that the modeling difficulty of each sensor/node in the ST traffic graph can vary significantly due to its spatial dependencies and temporal trends, resulting in unreliable and unstable predictions in IoT scenarios. In this context, we argue that a well-designed curriculum with an easy-to-difficult order can improve the training of ST models. Therefore, this paper introduces an ST difficulty measurer to score the node-level difficulty of traffic graph from both spatial and temporal aspects, and then implements a curriculum in the ST model training process. More specifically, based on the tiered ST difficulty score, the ST model training begins with a subgraph consisting of “easy” nodes characterized by relatively consistent spatial relationships and regular temporal patterns. Gradually, more difficult nodes are incorporated into the subgraph and participate in subsequent training stages. Comprehensive experiments and analysis on two real-world traffic flow datasets confirm the effectiveness of our proposed approach.

Scaling laws offer valuable insights into the design of time series foundation models (TSFMs). However, previous research has largely focused on the scaling laws of TSFMs for in-distribution (ID) data, leaving their out-of-distribution (OOD) scaling behavior and the influence of model architectures less explored. In this work, we examine two common TSFM architectures—encoder-only and decoder-only Transformers—and investigate their scaling behavior on both ID and OOD data. These models are trained and evaluated across varying parameter counts, compute budgets, and dataset sizes. Our experiments reveal that the log-likelihood loss of TSFMs exhibits similar scaling behavior in both OOD and ID settings. We further compare the scaling properties across different architectures, incorporating two state-of-the-art TSFMs as case studies, showing that model architecture plays a significant role in scaling. The encoder-only Transformers demonstrate better scalability than the decoder-only Transformers, while the architectural enhancements in the two advanced TSFMs primarily improve ID performance but reduce OOD scalability. While scaling up TSFMs is expected to drive performance breakthroughs, the lack of a comprehensive understanding of TSFM scaling laws has hindered the development of a robust framework to guide model scaling. We fill this gap in this work by synthesizing our findings and providing practical guidelines for designing and scaling larger TSFMs with enhanced model capabilities.

Time-series Generation (TSG) is a prominent research area with broad applications in simulations, data augmentation, and counterfactual analysis. While existing methods have shown promise in unconditional single-domain TSG, real-world applications demand for cross-domain approaches capable of controlled generation tailored to domain-specific constraints and instance-level requirements. In this paper, we argue that text can provide semantic insights, domain information and instance-specific temporal patterns, to guide and improve TSG. We introduce “Text-Controlled TSG”, a task focused on generating realistic time series by incorporating textual descriptions. To address data scarcity in this setting, we propose a novel LLM-based Multi-Agent framework that synthesizes diverse, realistic text-to-TS datasets. Furthermore, we introduce Bridge, a hybrid text-controlled TSG framework that integrates semantic prototypes with text description for supporting domain-level guidance. This approach achieves state-of-the-art generation fidelity on 11 of 12 datasets, and improves controllability by up to 12% on MSE and 6% MAE compared to no text input generation, highlighting its potential for generating tailored time-series data.

Next POInt-of-Interest (POI) recommendation predicts a user's next move and facilitates location-based services such as navigation and travel planning. SOTA methods fuse each POI and its contexts (e.g., time, category, and region) into a single representation to model sequential user movement. This hinders the effective utilization of context information, and diverse user preferences are also neglected. To tackle these limitations, we propose Disentangled and Personalized Representation Learning (DPRL) as a novel method for next POI recommendation. DPRL decouples POIs and contexts during representation learning, capturing their sequential regularities independently using separate recurrent neural networks (RNNs). To model the preference of each user, DPRL adopts an aggregation mechanism that integrates dynamic user preferences and spatial-temporal factors into the learned representations. We compare DPRL with 16 state-of-the-art baselines. The results show that DPRL outperforms all baselines and achieves an average accuracy improvement of 10.53% over the best-performing baseline.

Popularity bias occurs when popular items are recommended far more frequently than they should be, negatively impacting both user experience and recommendation accuracy. Existing debiasing methods mitigate popularity bias often uniformly across all users and only partially consider the time evolution of users or items. However, users have different levels of preference for item popularity, and this preference is evolving over time. To address these issues, we propose a novel method called CausalEPP (Causal Intervention on Evolving Personal Popularity) for taming recommendation bias, which accounts for the evolving personal popularity of users. Specifically, we first introduce a metric called Evolving Personal Popularity to quantify each user's preference for popular items. Then, we design a causal graph that integrates evolving personal popularity into the conformity effect, and apply deconfounded training to mitigate the popularity bias of the causal graph. During inference, we consider the evolution consistency between users and items to achieve a better recommendation. Empirical studies demonstrate that CausalEPP outperforms baseline methods in reducing popularity bias while improving recommendation accuracy.

Large-scale human mobility exhibits spatial and temporal patterns that can assist policymakers in decision making. Although traditional prediction models attempt to capture these patterns, they are often affected by nonperiodic public events, such as disasters and occasional celebrations. Since regular human mobility patterns are affected by these events, estimating their causal effects is critical to accurate mobility predictions. News articles provide unique perspectives on these events, though processing them is a challenge. In this study, we propose a causality based prediction model, CausalMob, to analyze the causal effects of public events. We first utilize large language models (LLMs) to extract human intentions from news and transform them into features that act as causal treatments. Next, the model learns representations of spatio-temporal regional covariates from multiple data sources to serve as confounders for causal inference. Finally, we present a causal effect estimation framework to ensure that event features remain independent of confounders during prediction. Based on large-scale real-world data, the experimental results show that the proposed model excels in human mobility prediction, outperforming state-of-the-art models.

Graphs are ubiquitous for modeling complex relationships between objects across various fields. Graph neural networks (GNNs) have become a mainstream technique for graph-based applications, but their performance heavily relies on abundant labeled data. To reduce labeling requirement, pre-training and prompt learning has become a popular alternative. However, most existing prompt methods do not distinguish between homophilic and heterophilic characteristics in graphs. In particular, many real-world graphs are non-homophilic-neither strictly nor uniformly homophilic-as they exhibit varying homophilic and heterophilic patterns across graphs and nodes. In this paper, we propose ProNoG, a novel pre-training and prompt learning framework for such non-homophilic graphs. First, we examineexisting graph pre-training methods, providing insights into the choice of pre-training tasks. Second, recognizing that each node exhibits unique non-homophilic characteristics, we propose a conditional network to characterize node-specific patterns in downstream tasks. Finally, we thoroughly evaluate and analyze ProNoG through extensive experiments on ten public datasets.

Transparent object segmentation from a single image has been investigated for several years. However, detecting transparent areas from video has not been well explored, especially for different kinds of transparent categories besides glass, due to the scarcity of such a dataset. Therefore, in this paper, we propose the video-based transparent object segmentation task and introduce the first-of-its-kind corresponding dataset named TransVid, which contains nearly 400 videos with a total of 18,523 frames. Based on TranVid, we further propose a new method called TranSeg, in which we innovatively introduce Graph Neural Networks into the temporal segmentation task and combined with a novel Diffusion Model to make the model's segmentation results more accurate. Experimental results show that TranSeg achieves higher accuracy with fewer parameters than previous state-of-the-art models, demonstrating the effectiveness of our method. Moreover, comprehensive ablation analysis reveal several fascinating insights and suggest viable paths for further research.

Real-world data distributions often shift continuously across multiple latent factors such as time, geography, and socioeconomic contexts. However, existing domain generalization approaches typically treat domains as discrete or as evolving along a single axis (e.g., time). This oversimplification fails to capture the complex, multidimensional nature of real-world variation. This paper introduces the task of Continuous Domain Generalization (CDG), which aims to generalize predictive models to unseen domains defined by arbitrary combinations of continuous variations. We present a principled framework grounded in geometric and algebraic theories, showing that optimal model parameters across domains lie on a low-dimensional manifold. To model this structure, we propose a Neural Lie Transport Operator (NeuralLio), which enables structure-preserving parameter transitions by enforcing geometric continuity and algebraic consistency. To handle noisy or incomplete domain variation descriptors, we introduce a gating mechanism to suppress irrelevant dimensions and a local chart-based strategy for robust generalization. Extensive experiments on synthetic and real-world datasets, including remote sensing, scientific documents, and traffic forecasting, demonstrate that our method significantly outperforms existing baselines in both generalization accuracy and robustness.

Spatio-temporal forecasting is essential for real-world applications such as traffic management and urban computing. Although recent methods have shown improved accuracy, they often fail to account for dynamic deviations between current inputs and historical patterns. These deviations contain critical signals that can significantly affect model performance. To fill this gap, we propose $\textbf{ST-SSDL}$, a $\underline{S}$patio-$\underline{T}$emporal time series forecasting framework that incorporates a $\underline{S}$elf-$\underline{S}$upervised $\underline{D}$eviation $\underline{L}$earning scheme to capture and utilize such deviations. ST-SSDL anchors each input to its historical average and discretizes the latent space using learnable prototypes that represent typical spatio-temporal patterns. Two auxiliary objectives are proposed to refine this structure: a contrastive loss that enhances inter-prototype discriminability and a deviation loss that regularizes the distance consistency between input representations and corresponding prototypes to quantify deviation. Optimized jointly with the forecasting objective, these components guide the model to organize its hidden space and improve generalization across diverse input conditions. Experiments on six benchmark datasets show that ST-SSDL consistently outperforms state-of-the-art baselines across multiple metrics. Visualizations further demonstrate its ability to adaptively respond to varying levels of deviation in complex spatio-temporal scenarios. Our code and datasets are available at https://github.com/Jimmy-7664/ST-SSDL.

Dynamic graph modeling aims to uncover evolutionary patterns in real-world systems, enabling accurate social recommendation and early detection of cancer cells. Inspired by the success of recent state space models in efficiently capturing long-term dependencies, we propose DyG-Mamba by translating dynamic graph modeling into a long-term sequence modeling problem. Specifically, inspired by Ebbinghaus' forgetting curve, we treat the irregular timespans between events as control signals, allowing DyG-Mamba to dynamically adjust the forgetting of historical information. This mechanism ensures effective usage of irregular timespans, thereby improving both model effectiveness and inductive capability. In addition, inspired by Ebbinghaus' review cycle, we redefine core parameters to ensure that DyG-Mamba selectively reviews historical information and filters out noisy inputs, further enhancing the model’s robustness. Through exhaustive experiments on 12 datasets covering dynamic link prediction and node classification tasks, we show that DyG-Mamba achieves state-of-the-art performance on most datasets, while demonstrating significantly improved computational and memory efficiency. Our code is available at https://github.com/Clearloveyuan/DyG-Mamba.

In the field of multi-document summarization (MDS), transformerbased models have demonstrated remarkable success, yet they suffer an input length limitation.Current methods apply truncation after the retrieval process to fit the context length; however, they heavily depend on manually well-crafted queries, which are impractical to create for each document set for MDS.Additionally, these methods retrieve information at a coarse granularity, leading to the inclusion of irrelevant content.To address these issues, we propose a novel retrieval-based framework that integrates query selection and document ranking and shortening into a unified process.Our approach identifies the most salient elementary discourse units (EDUs) from input documents and utilizes them as latent queries.These queries guide the document ranking by calculating relevance scores.Instead of traditional truncation, our approach filters out irrelevant EDUs to fit the context length, ensuring that only critical information is preserved for summarization.We evaluate our framework on multiple MDS datasets, demonstrating consistent improvements in ROUGE metrics while confirming its scalability and flexibility across diverse model architectures.Additionally, we validate its effectiveness through an in-depth analysis, emphasizing its ability to dynamically select appropriate queries and accurately rank documents based on their relevance scores.These results demonstrate that our framework effectively addresses context-length constraints, establishing it as a robust and reliable solution for MDS. 1 * Both authors contributed equally to this research.

Dynamic graph clustering aims to detect and track time-varying clusters in dynamic graphs, revealing the evolutionary mechanisms of complex real-world dynamic systems. Matrix factorization-based methods are promising approaches for this task; however, these methods often struggle with scalability and can be time-consuming when applied to large-scale dynamic graphs. Moreover, they tend to lack robustness and are vulnerable to real-world noisy data. To address these issues, we make three key contributions. First, to improve scalability, we propose temporal separated matrix factorization, where a single matrix is divided into multiple smaller matrices for independent factorization, resulting in faster computation. Second, to improve robustness, we introduce bi-clustering regularization, which jointly optimizes graph embedding and clustering, thereby filtering out noisy features from the graph embeddings. Third, to further enhance effectiveness and efficiency, we propose selective embedding updating, where we update only the embeddings of dynamic nodes while the embeddings of static nodes are fixed among different timestamps. Experimental results on six synthetic and five real-world benchmarks demonstrate the scalability, robustness and effectiveness of our proposed method. Source code is available at https://github.com/Clearloveyuan/DyG-MF.

Road trajectory generation creates synthetic yet realistic trajectories to tackle data collection costs and privacy concerns. Existing methods generate a trajectory either segment-by-segment using sequence models or holistically in one step using diffusion models. Sequence-based models have good regularity and consistency (i.e., resemble the input trajectories) but lack diversity, while diffusion-based models enhance diversity but sacrifice regularity and consistency. To combine the merits of existing methods, we propose Seed, by bridging sequence and diffusion models for trajectory generation. In particular, Seed adopts a conditional diffusion structure, where a Transformer models the movement of each trajectory along the road segments, and conditioned on the Transformer's output, a diffusion model recovers the next road segment from random noise. The rationale is that the Transformer captures sequential patterns for regularity and consistency, while the diffusion model introduces diversity by recovering from noise. We use a trajectory reconstruction task to train Seed, and design a curriculum learning strategy to accelerate convergence. We compare Seed with 8 state-of-the-art trajectory generation methods on 3 datasets, and the results show that Seed improves the best-performing baseline by over 50%.

Abstract Trajectories serve as a cornerstone of intelligent transportation systems, playing an important role in many applications such as traffic flow prediction, route planning, and urban management. However, the availability of such data is limited due to privacy issues, ethical concerns, and the high cost associated with infrastructure deployment. In recent years, rapidly developing generative models such as Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and Diffusion Models (DMs) have demonstrated strong capabilities in learning complex data distributions and generating synthetic data, thereby alleviating the data accessibility issue. In this survey, we systematically review the existing literature on deep generative models that address the problem of trajectory generation. First, we classify the existing literature into two categories: unconditional and conditional trajectory generation. In unconditional generation, trajectories are generated without contextual constraints, whereas conditional generation incorporates several important factors such as road network topology, time of day, and user preferences to guide the trajectory generation process. Then, for each category, we further classify the literature into three methodological types, including VAEs, GANs, and DMs, and analyze how these models address key challenges under different settings. Finally, we discuss promising directions for future research and hope to inspire further advances in trajectory generation.

Accurately and jointly predicting multimodal transportation demand is crucial for pre-allocating transport resources, enhancing the resilience of traffic systems. However, current approaches insufficiently explore inter- and intra-mode heterogeneity, resulting in undifferentiated dependency extraction. Moreover, existing research struggles to model cross-mode interactions among three or more transportation modes and adapt to dynamic relations in multimodal demand. To address these limitations, we propose a novel multimodal demand prediction model based on a meta-parameter learning network (MMDNet), centered on characterizing multimodal traffic spatial-temporal heterogeneity and unifying the modeling of cross-mode interactions. Our model features: 1) a spatial-temporal heterogeneity meta-parameter learning method, capturing both inter- and intra-mode heterogeneity to steer more targeted dependency extraction than previous studies; 2) a spatial-temporal evolving unified graph generator, transcending prior studies’ limitations in unifying dynamic interactions across three or more modes by creating dynamic unified graphs. Extensive experiments on three real-world datasets (New York, Beijing and Chicago) covering four different traffic modes are carried out to evaluate the MMDNet. The model achieves a 6.65% performance gain over advanced baselines and demonstrates strong cross-city adaptability. Abundant interpretability analyses show our model can semantically encode explainable cross-mode interactions and differences between modes. Source codes are available athttps://github.com/zhjiang1/MMDNet

Next point-of-interest (POI) recommendationpredicts user’s next movement and facilitates location-based applications such as destination suggestion and travel planning. State-of-the-art (SOTA) methods learn an adaptive graph from user trajectories and compute POI representations using graph neural networks (GNNs). However, a single graph cannot capture thediverse dependenciesamong the POIs (e.g., geographical proximity and transition frequency). To tackle this limitation, we propose theAdaptiveGraphContrastiveLearning(AGCL) framework. AGCL constructs multiple adaptive graphs, each modeling a kind of POI dependency and producing one POI representation; and the POI representations from different graphs are merged into amulti-facet representationthat encodes comprehensive information. To train the POI representations, we tailor agraph-based contrastive learning, which encourages the representations of similar POIs to align and dissimilar POIs to differentiate. Moreover, to learn the sequential regularities of user trajectories, we design an attention mechanism to integrate spatial-temporal information into the POI representations. An explicitspatial-temporal biasis also employed to adjust the predictions for enhanced accuracy. We compare AGCL with 10 state-of-the-art baselines on 3 datasets. The results show that AGCL outperforms all baselines and achieves an improvement of 10.14% over the best performing baseline in average accuracy.

Spatiotemporal neural networks have shown great promise in urban scenarios by effectively capturing temporal and spatial correlations. However, urban environments are constantly evolving, and current model evaluations are often limited to traffic scenarios and use data mainly collected only a few weeks after training period to evaluate model performance. The generalization ability of these models remains largely unexplored. To address this, we propose a Spatiotemporal Out-of-Distribution (ST-OOD) benchmark, which comprises six urban scenario: bike-sharing, 311 services, pedestrian counts, traffic speed, traffic flow, ride-hailing demand, and bike-sharing, each with in-distribution (same year) and out-of-distribution (next years) settings. We extensively evaluate state-of-the-art spatiotemporal models and find that their performance degrades significantly in out-of-distribution settings, with most models performing even worse than a simple Multi-Layer Perceptron (MLP). Our findings suggest that current leading methods tend to over-rely on parameters to overfit training data, which may lead to good performance on in-distribution data but often results in poor generalization. We also investigated whether dropout could mitigate the negative effects of overfitting. Our results showed that a slight dropout rate could significantly improve generalization performance on most datasets, with minimal impact on in-distribution performance. However, balancing in-distribution and out-of-distribution performance remains a challenging problem. We hope that the proposed benchmark will encourage further research on this critical issue.

Active learning seeks to achieve strong performance with fewer training samples. It does this by iteratively asking an oracle to label newly selected samples in a human-in-the-loop manner. This technique has gained increasing popularity due to its broad applicability, yet its survey papers, especially for deep active learning (DAL), remain scarce. Therefore, we conduct an advanced and comprehensive survey on DAL. We first introduce reviewed paper collection and filtering. Second, we formally define the DAL task and summarize the most influential baselines and widely used datasets. Third, we systematically provide a taxonomy of DAL methods from five perspectives, including annotation types, query strategies, deep model architectures, learning paradigms, and training processes, and objectively analyze their strengths and weaknesses. Then, we comprehensively summarize the main applications of DAL in natural language processing (NLP), computer vision (CV), data mining (DM), and so on. Finally, we discuss challenges and perspectives after a detailed analysis of current studies. This work aims to serve as a useful and quick guide for researchers in overcoming difficulties in DAL. We hope that this survey will spur further progress in this burgeoning field.

Training models on spatio-temporal (ST) data poses an open problem due to the complicated and diverse nature of the data itself, and it is challenging to ensure the model's performance directly trained on the original ST data. While limiting the variety of training data can make training easier, it can also lead to a lack of knowledge and information for the model, resulting in a decrease in performance. To address this challenge, we presented an innovative paradigm that incorporates three separate forms of curriculum learning specifically targeting from spatial, temporal, and quantile perspectives. Furthermore, our framework incorporates a stacking fusion module to combine diverse information from three types of curriculum learning, resulting in a strong and thorough learning process. We demonstrated the effectiveness of this framework with extensive empirical evaluations, highlighting its better performance in addressing complex ST challenges. We provided thorough ablation studies to investigate the effectiveness of our curriculum and to explain how it contributes to the improvement of learning efficiency on ST data.

Anomaly detection significantly enhances the robustness of cloud systems. While neural network-based methods have recently demonstrated strong advantages, they encounter practical challenges in cloud environments: the contradiction between the impracticality of maintaining a unique model for each service and the limited ability to deal with diverse normal patterns by a unified model, as well as issues with handling heavy traffic in real time and short-term anomaly detection sensitivity. Thus, we propose MACE, a multi-normal-pattern accommodated and efficient anomaly detection method in the frequency domain for time series anomaly detection. There are three novel characteristics of it: (i) a pattern extraction mechanism excelling at handling diverse normal patterns with a unified model, which enables the model to identify anomalies by examining the correlation between the data sample and its service normal pattern, instead of solely focusing on the data sample itself; (ii) a dualistic convolution mechanism that amplifies short-term anomalies in the time domain and hinders the reconstruction of anomalies in the frequency domain, which enlarges the reconstruction error disparity between anomaly and normality and facilitates anomaly detection; (iii) leveraging the sparsity and parallelism of frequency domain to enhance model efficiency. We theoretically and experimentally prove that using a strategically selected subset of Fourier bases can not only reduce computational overhead but is also profitable to distinguish anomalies, compared to using the complete spectrum. Moreover, extensive experiments demonstrate MACE's effectiveness in handling diverse normal patterns with a unified model and it achieves state-of-the-art performance with high efficiency.

Graph augmentation has received great attention in recent years for graph contrastive learning (GCL) to learn well-generalized node/graph representations. However, mainstream GCL methods often favor randomly disrupting graphs for augmentation, which shows limited generalization and inevitably leads to the corruption of high-level graph information, i.e., the graph community. Moreover, current knowledge-based graph augmentation methods can only focus on either topology or node features, causing the model to lack robustness against various types of noise. To address these limitations, this research investigated the role of the graph community in graph augmentation and figured out its crucial advantage for learnable graph augmentation. Based on our observations, we propose a community-invariant GCL framework to maintain graph community structure during learnable graph augmentation. By maximizing the spectral changes, this framework unifies the constraints of both topology and feature augmentation, enhancing the model’s robustness. Empirical evidence on 21 benchmark datasets demonstrates the exclusive merits of our framework. Code is released on Github (https://github.com/ShiyinTan/CI-GCL.git).

Spatiotemporal time series forecasting plays a key role in a wide range of real-world applications. While significant progress has been made in this area, fully capturing and leveraging spatiotemporal heterogeneity remains a fundamental challenge. Therefore, we propose a novel Heterogeneity-Informed Meta-Parameter Learning scheme. Specifically, our approach implicitly captures spatiotemporal heterogeneity through learning spatial and temporal embeddings, which can be viewed as a clustering process. Then, a novel spatiotemporal meta-parameter learning paradigm is proposed to learn spatiotemporal-specific parameters from meta-parameter pools, which is informed by the captured heterogeneity. Based on these ideas, we develop a Heterogeneity-Informed Spatiotemporal Meta-Network (HimNet) for spatiotemporal time series forecasting. Extensive experiments on five widely-used benchmarks demonstrate our method achieves state-of-the-art performance while exhibiting superior interpretability. Our code is available at https://github.com/XDZhelheim/HimNet.

Low-light image enhancement has been researched several years. However, current image restoration methods predominantly focus on recovering images from RGB images, overlooking the potential of incorporating more modalities. With the advancements in personal handheld devices, we can now easily capture images with depth information using devices such as mobile phones. The integration of depth information into image restoration is a research question worthy of exploration. Therefore, in this paper, we propose a multimodal low-light image enhancement task based on depth information and establish a dataset named LED (Low-light Image Enhanced with Depth Map), consisting of 1,365 samples. Each sample in our dataset includes a low-light image, a normal-light image, and the corresponding depth map. Moreover, for the LED dataset, we design a corresponding multimodal method, which can processes the input images and depth map information simultaneously to generate the predicted normal-light images. Experimental results and detailed ablation studies proves the efficiency of our method which exceeds previous single-modal state-of-the arts methods from relevant field.

With the popularity of location-based services, human mobility prediction plays a key role in enhancing personalized navigation, optimizing recommendation systems, and facilitating urban mobility and planning. This involves predicting a user's next POI (point-of-interest) visit using their past visit history. However, the uneven distribution of visitations over time and space, namely the long-tail problem in spatial distribution, makes it difficult for AI models to predict those POIs that are less visited by humans. In light of this issue, we propose the $\underline{\bf{Lo}}$ng-$\underline{\bf{T}}$ail Adjusted $\underline{\bf{Next}}$ POI Prediction (LoTNext) framework for mobility prediction, combining a Long-Tailed Graph Adjustment module to reduce the impact of the long-tailed nodes in the user-POI interaction graph and a novel Long-Tailed Loss Adjustment module to adjust loss by logit score and sample weight adjustment strategy. Also, we employ the auxiliary prediction task to enhance generalization and accuracy. Our experiments with two real-world trajectory datasets demonstrate that LoTNext significantly surpasses existing state-of-the-art works.

Temporal Domain Generalization (TDG) addresses the challenge of training predictive models under temporally varying data distributions. Traditional TDG approaches typically focus on domain data collected at fixed, discrete time intervals, which limits their capability to capture the inherent dynamics within continuous-evolving and irregularly-observed temporal domains. To overcome this, this work formalizes the concept of Continuous Temporal Domain Generalization (CTDG), where domain data are derived from continuous times and are collected at arbitrary times. CTDG tackles critical challenges including: 1) Characterizing the continuous dynamics of both data and models, 2) Learning complex high-dimensional nonlinear dynamics, and 3) Optimizing and controlling the generalization across continuous temporal domains. To address them, we propose a Koopman operator-driven continuous temporal domain generalization (Koodos) framework. We formulate the problem within a continuous dynamic system and leverage the Koopman theory to learn the underlying dynamics; the framework is further enhanced with a comprehensive optimization strategy equipped with analysis and control driven by prior knowledge of the dynamics patterns. Extensive experiments demonstrate the effectiveness and efficiency of our approach. The code can be found at: https://github.com/Zekun-Cai/Koodos.

This paper introduces a novel approach using Large Language Models (LLMs) integrated into an agent framework for flexible and effective personal mobility generation. LLMs overcome the limitations of previous models by effectively processing semantic data and offering versatility in modeling various tasks. Our approach addresses three research questions: aligning LLMs with real-world urban mobility data, developing reliable activity generation strategies, and exploring LLM applications in urban mobility. The key technical contribution is a novel LLM agent framework that accounts for individual activity patterns and motivations, including a self-consistency approach to align LLMs with real-world activity data and a retrieval-augmented strategy for interpretable activity generation. We evaluate our LLM agent framework and compare it with state-of-the-art personal mobility generation approaches, demonstrating the effectiveness of our approach and its potential applications in urban mobility. Overall, this study marks the pioneering work of designing an LLM agent framework for activity generation based on real-world human activity data, offering a promising tool for urban mobility analysis.

Free-space trajectory similarity calculation, e.g., DTW, Hausdorff, and Fréchet, often incur quadratic time complexity, thus learning-based methods have been proposed to accelerate the computation. The core idea is to train an encoder to transform trajectories into representation vectors and then compute vector similarity to approximate the ground truth. However, existing methods face dual challenges of effectiveness and efficiency: 1) they all utilize Euclidean distance to compute representation similarity, which leads to the severe curse of dimensionality issue - reducing the distinguishability among representations and significantly affecting the accuracy of subsequent similarity search tasks; 2) most of them are trained in triplets manner and often necessitate additional information which downgrades the efficiency; 3) previous studies, while emphasizing the scalability in terms of efficiency, overlooked the deterioration of effectiveness when the dataset size grows. To cope with these issues, we propose a simple, yet accurate, fast, scalable model that only uses a single-layer vanilla transformer encoder as the feature extractor and employs tailored representation similarity functions to approximate various ground truth similarity measures. Extensive experiments demonstrate our model significantly mitigates the curse of dimensionality issue and outperforms the state-of-the-arts in effectiveness, efficiency, and scalability.

Tensor time series (TTS) data, a generalization of one-dimensional time series on a high-dimensional space, is ubiquitous in real-world applications. Compared to modeling time series or multivariate time series, which has received much attention and achieved tremendous progress in recent years, tensor time series has been paid less effort. However, properly coping with the TTS is a much more challenging task, due to its high-dimensional and complex inner structure. In this article, we start by revealing the structure of TTS data from afn statistical view of point. Then, in line with this analysis, we perform T ensor T ime S eries forecasting via a proposed Multi-way Norm alization ( TTS-Norm ), which effectively disentangles multiple heterogeneous low-dimensional substructures from the original high-dimensional structure. Finally, we design a novel objective function for TTS forecasting, accounting for the numerical heterogeneity among different low-dimensional subspaces of TTS. Extensive experiments on two real-world datasets verify the superior performance of our proposed model. 1

Multivariate time-series (MTS) forecasting is a paramount and fundamental problem in many real-world applications. The core issue in MTS forecasting is how to effectively model complex spatial-temporal patterns. In this paper, we develop a adaptive, interpretable and scalable forecasting framework, which seeks to individually model each component of the spatial-temporal patterns. We name this framework SCNN, as an acronym ofStructuredComponent-basedNeuralNetwork. SCNN works with a pre-defined generative process of MTS, which arithmetically characterizes the latent structure of the spatial-temporal patterns. In line with its reverse process, SCNN decouples MTS data into structured and heterogeneous components and then respectively extrapolates the evolution of these components, the dynamics of which are more traceable and predictable than the original MTS. Extensive experiments are conducted to demonstrate that SCNN can achieve superior performance over state-of-the-art models on three real-world datasets. Additionally, we examine SCNN with different configurations and perform in-depth analyses of the properties of SCNN.

Perceiving and modeling urban crowd movements are of great importance to smart city-related fields. Governments and public service operators can benefit from such efforts as they can be applied to crowd management, resource scheduling, and early emergency warning. However, most prior research on urban crowd modeling has failed to describe the dynamics and continuity of human mobility, leading to inconsistent and irrelevant results when they tackle multiple homogeneous forecasting tasks as they can only be modeled independently. To overcome this drawback, we propose to model human mobility from a new perspective, which uses the citywide crowd transition process constituted by a series of transition matrices from low order to high order, to help us understand how the crowd dynamics evolve step-by-step. We further propose a Deep Transition Process Network to process and predict such new high-dimensional data, where novel grid embedding with Graph Convolutional Network, parameter-shared Convolutional LSTM, and High-Dimensional Attention mechanism are designed to learn the complicated dependencies in terms of spatial, temporal, and ordinal features. We conduct experiments on two datasets generated by a large amount of GPS data collected from a real-world smartphone application. The experiment results demonstrate the superior performance of our proposed methodology over existing approaches.

Traffic forecasting as a canonical task of multivariate time series forecasting has been a significant research topic in AI community. To address the spatio-temporal heterogeneity and non-stationarity implied in the traffic stream, in this study, we propose Spatio-Temporal Meta-Graph Learning as a novel Graph Structure Learning mechanism on spatio-temporal data. Specifically, we implement this idea into Meta-Graph Convolutional Recurrent Network (MegaCRN) by plugging the Meta-Graph Learner powered by a Meta-Node Bank into GCRN encoder-decoder. We conduct a comprehensive evaluation on two benchmark datasets (i.e., METR-LA and PEMS-BAY) and a new large-scale traffic speed dataset called EXPY-TKY that covers 1843 expressway road links in Tokyo. Our model outperformed the state-of-the-arts on all three datasets. Besides, through a series of qualitative evaluations, we demonstrate that our model can explicitly disentangle the road links and time slots with different patterns and be robustly adaptive to any anomalous traffic situations. Codes and datasets are available at https://github.com/deepkashiwa20/MegaCRN.

Spatial-temporal (ST) graph modeling, such as traffic speed forecasting and taxi demand prediction, is an important task in deep learning area. However, for the nodes in the graph, their ST patterns can vary greatly in difficulties for modeling, owning to the heterogeneous nature of ST data. We argue that unveiling the nodes to the model in a meaningful order, from easy to complex, can provide performance improvements over traditional training procedure. The idea has its root in Curriculum Learning, which suggests in the early stage of training models can be sensitive to noise and difficult samples. In this paper, we propose ST-Curriculum Dropout, a novel and easy-to-implement strategy for spatial-temporal graph modeling. Specifically, we evaluate the learning difficulty of each node in high-level feature space and drop those difficult ones out to ensure the model only needs to handle fundamental ST relations at the beginning, before gradually moving to hard ones. Our strategy can be applied to any canonical deep learning architecture without extra trainable parameters, and extensive experiments on a wide range of datasets are conducted to illustrate that, by controlling the difficulty level of ST relations as the training progresses, the model is able to capture better representation of the data and thus yields better generalization.

Urban time series data forecasting featuring significant contributions to sustainable development is widely studied as an essential task of the smart city. However, with the dramatic and rapid changes in the world environment, the assumption that data obey Independent Identically Distribution is undermined by the subsequent changes in data distribution, known as concept drift, leading to weak replicability and transferability of the model over unseen data. To address the issue, previous approaches typically retrain the model, forcing it to fit the most recent observed data. However, retraining is problematic in that it leads to model lag, consumption of resources, and model re-invalidation, causing the drift problem to be not well solved in realistic scenarios. In this study, we propose a new urban time series prediction model for the concept drift problem, which encodes the drift by considering the periodicity in the data and makes on-the-fly adjustments to the model based on the drift using a meta-dynamic network. Experiments on real-world datasets show that our design significantly outperforms state-of-the-art methods and can be well generalized to existing prediction backbones by reducing their sensitivity to distribution changes.

With the rapid development of the Intelligent Transportation System (ITS), accurate traffic forecasting has emerged as a critical challenge. The key bottleneck lies in capturing the intricate spatio-temporal traffic patterns. In recent years, numerous neural networks with complicated architectures have been proposed to address this issue. However, the advancements in network architectures have encountered diminishing performance gains. In this study, we present a novel component called spatio-temporal adaptive embedding that can yield outstanding results with vanilla transformers. Our proposed Spatio-Temporal Adaptive Embedding transformer (STAEformer) achieves state-of-the-art performance on five real-world traffic forecasting datasets. Further experiments demonstrate that spatio-temporal adaptive embedding plays a crucial role in traffic forecasting by effectively capturing intrinsic spatio-temporal relations and chronological information in traffic time series.

Next Point-of-Interest (POI) recommendation plays a crucial role in urban mobility applications. Recently, POI recommendation models based on Graph Neural Networks (GNN) have been extensively studied and achieved, however, the effective incorporation of both spatial and temporal information into such GNN-based models remains challenging. Temporal information is extracted from users' trajectories, while spatial information is obtained from POIs. Extracting distinct fine-grained features unique to each piece of information is difficult since temporal information often includes spatial information, as users tend to visit nearby POIs. To address the challenge, we propose Mobility Graph Transformer (MobGT) that enables us to fully leverage graphs to capture both the spatial and temporal features in users' mobility patterns. MobGT combines individual spatial and temporal graph encoders to capture unique features and global user-location relations. Additionally, it incorporates a mobility encoder based on Graph Transformer to extract higher-order information between POIs. To address the long-tailed problem in spatial-temporal data, MobGT introduces a novel loss function, Tail Loss. Experimental results demonstrate that MobGT outperforms state-of-the-art models on various datasets and metrics, achieving 24% improvement on average. Our codes are available at https://github.com/Yukayo/MobGT.

Tensor time series (TTS) data, a generalization of one-dimensional time series on a high-dimensional space, is ubiquitous in real-world scenarios, especially in monitoring systems involving multi-source spatio-temporal data (e.g., transportation demands and air pollutants). Compared to modeling time series or multivariate time series, which has received much attention and achieved tremendous progress in recent years, tensor time series has been paid less effort. Properly coping with the tensor time series is a much more challenging task, due to its high-dimensional and complex inner structure. In this paper, we develop a novel TTS forecasting framework, which seeks to individually model each heterogeneity component implied in the time, the location, and the source variables. We name this framework as GMRL, short for Gaussian Mixture Representation Learning. Experiment results on two real-world TTS datasets verify the superiority of our approach compared with the state-of-the-art baselines. Code and data are published on https://github.com/beginner-sketch/GMRL.

Today, thanks to the rapid developing mobile and sensor networks in IoT (Internet of Things) systems, spatio-temporal big data are being constantly generated. They have brought us a data-driven possibility to sense and understand crowd mobility on a city scale. A fundamental task towards the next-generation mobility services, such as Intelligent Transportation Systems (ITS), Mobility-as-a-Service (MaaS), is spatio-temporal predictive modeling of the geo-sensory signals. There is a recent line of research leveraging deep learning techniques to boost the forecasting performance on such tasks. While simulating the regularity of mobility behaviors (e.g., routines, periodicity) in a more sophisticated way, the existing studies ignore an important part of urban activities, i.e., events. Including holidays, extreme weathers, pandemic, accidents, various urban events happen from time to time and cause non-stationary phenomena, which by nature make the spatio-temporal forecasting task challenging. We thereby envision an event-aware urban mobility prediction model that is capable of fast adapting and making reliable predictions in different scenarios, which is crucial to decision making towards emergency response and urban resilience.

Mobility digital twin, which is a a virtual replica of the mobility in the physical world, is the key building block of modern smart city applications at a metropolitan scale, including traffic regulation, emergency management and epidemic control. To duplicate the mobility in the physical world and show the potential outcome based on either the current state or manipulated conditions, we are facing with three main challenges: 1) how to sense real-time human mobility at a large scale and assimilate different data sources to infer a dynamic city mobility state; 2) how to make an accurate prediction for the mobility replica that adapts with dynamic city mobility state; 3) how to simulate the mobility in response to different conditions. In this talk, we will present our recent studies, practices and perspectives on mobility digital twin in addressing these above challenges, with applications in the real-world scenarios with industrial connections.

Human mobility nowcasting is a fundamental research problem for intelligent transportation planning, disaster responses and management, etc. In particular, human mobility under big disasters such as hurricanes and pandemics deviates from its daily routine to a large extent, which makes the task more challenging. Existing works mainly focus on traffic or crowd flow prediction in normal situations. To tackle this problem, in this study, disaster-related Twitter data is incorporated as a covariate to understand the public awareness and attention about the disaster events and thus perceive their impacts on the human mobility. Accordingly, we propose a Meta-knowledge-Memorizable Spatio-Temporal Network (MemeSTN), which leverages memory network and meta-learning to fuse social media and human mobility data. Extensive experiments over three real-world disasters including Japan 2019 typhoon season, Japan 2020 COVID-19 pandemic, and US 2019 hurricane season were conducted to illustrate the effectiveness of our proposed solution. Compared to the state-of-the-art spatio-temporal deep models and multivariate-time-series deep models, our model can achieve superior performance for nowcasting human mobility in disaster situations at both country level and state level.

Multi-variate time series (MTS) data is a ubiquitous class of data abstraction in the real world. Any instance of MTS is generated from a hybrid dynamical system and their specific dynamics are usually unknown. The hybrid nature of such a dynamical system is a result of complex external attributes, such as geographic location and time of day, each of which can be categorized into either spatial attributes or temporal attributes. Therefore, there are two fundamental views which can be used to analyze MTS data, namely the spatial view and the temporal view. Moreover, from each of these two views, we can partition the set of data samples of MTS into disjoint forecasting tasks in accordance with their associated attribute values. Then, samples of the same task will manifest similar forthcoming pattern, which is less sophisticated to be predicted in comparison with the original single-view setting. Considering this insight, we propose a novel multi-view multi-task (MVMT) learning framework for MTS forecasting. Instead of being explicitly presented in most scenarios, MVMT information is deeply concealed in the MTS data, which severely hinders the model from capturing it naturally. To this end, we develop two kinds of basic operations, namely task-wise affine transformation and task-wise normalization, respectively. Applying these two operations with prior knowledge on the spatial and temporal view allows the model to adaptively extract MVMT information while predicting. Extensive experiments on three datasets are conducted to illustrate that canonical architectures can be greatly enhanced by the MVMT learning framework in terms of both effectiveness and efficiency. In addition, we design rich case studies to reveal the properties of representations produced at different phases in the entire prediction procedure.

Predicting the density and flow of the crowd or traffic at a citywide level becomes possible by using the big data and cutting-edge AI technologies. It has been a very significant research topic with high social impact, which can be widely applied to emergency management, traffic regulation, and urban planning. In particular, by meshing a large urban area to a number of fine-grained mesh-grids, citywide crowd and traffic information in a continuous time period can be represented with 4D tensor (Timestep, Height, Width, Channel). Based on this idea, a series of methods have been proposed to address grid-based prediction for citywide crowd and traffic. In this study, we revisit the density and in-out flow prediction problem and publish a new aggregated human mobility dataset generated from a real-world smartphone application. Comparing with the existing ones, our dataset holds several advantages including large mesh-grid number, fine-grained mesh size, and high user sample. Towards this large-scale crowd dataset, we propose a novel deep learning model called DeepCrowd by designing pyramid architectures and high-dimensional attention mechanism based on Convolutional LSTM. Lastly, thorough and comprehensive performance evaluations are conducted to demonstrate the superiority of the proposed DeepCrowd comparing to multiple state-of-the-art methods.

The outbreak of coronavirus disease (COVID-19) has swept across more than 180 countries and territories since late January 2020. As a worldwide emergency response, governments have implemented various measures and policies, such as self-quarantine, travel restrictions, work from home, and regional lockdown, to control the spread of the epidemic. These countermeasures seek to restrict human mobility because COVID-19 is a highly contagious disease that is spread by human-to-human transmission. Medical experts and policymakers have expressed the urgency to effectively evaluate the outcome of human restriction policies with the aid of big data and information technology. Thus, based on big human mobility data and city POI data, an interactive visual analytics system called Epidemic Mobility (EpiMob) was designed in this study. The system interactively simulates the changes in human mobility and infection status in response to the implementation of a certain restriction policy or a combination of policies (e.g., regional lockdown, telecommuting, screening). Users can conveniently designate the spatial and temporal ranges for different mobility restriction policies. Then, the results reflecting the infection situation under different policies are dynamically displayed and can be flexibly compared and analyzed in depth. Multiple case studies consisting of interviews with domain experts were conducted in the largest metropolitan area of Japan (i.e., Greater Tokyo Area) to demonstrate that the system can provide insight into the effects of different human mobility restriction policies for epidemic control, through measurements and comparisons.

As a decisive part in the success of Mobility-as-a-Service (MaaS), spatio-temporal predictive modeling for crowd movements is a challenging task particularly considering scenarios where societal events drive mobility behavior deviated from the normality. While tremendous progress has been made to model high-level spatio-temporal regularities with deep learning, most, if not all of the existing methods are neither aware of the dynamic interactions among multiple transport modes nor adaptive to unprecedented volatility brought by potential societal events. In this paper, we are therefore motivated to improve the canonical spatio-temporal network (ST-Net) from two perspectives: (1) design a heterogeneous mobility information network (HMIN) to explicitly represent intermodality in multimodal mobility; (2) propose a memory-augmented dynamic filter generator (MDFG) to generate sequence-specific parameters in an on-the-fly fashion for various scenarios. The enhanced event-aware spatio-temporal network, namely EAST-Net, is evaluated on several real-world datasets with a wide variety and coverage of societal events. Both quantitative and qualitative experimental results verify the superiority of our approach compared with the state-of-the-art baselines. Code and data are published on https://github.com/underdoc-wang/EAST-Net.

A core component of the Intelligent Transportation System (ITS) is road network, which forms the most basic transport infrastructure, and becomes widely applied in many traffic applications. In most traffic models, the spatial representation of road network is learned only through static graph connection while dynamic driver preference and traffic conditions in the real world are ignored. Therefore, in this paper, a novel trajectory-based road network representation is proposed. By mining vehicle trajectories, our proposed method can learn dynamic route choice through embeddings of each road in a next-hop prediction model. Then road correlations are calculated by the embeddings to build a latent correlation graph that can be applied in various traffic-related applications. Extensive experiment results prove the effectiveness and rationality of our proposed approach.

COVID-19 has dramatically changed people’s mobility patterns. This report aims to analyze the impact of COVID-19 on people’s mobility through statistics and comparing the visits of POIs (Point-Of-Interests) in New York State in 2019 and 2020. The report uses data from SafeGraph, which is a data company. The raw data contains POI visits across the United States in 2019 and 2020. Considering the analysis size and difficulty of the data, POI visits from New York State are extracted for analysis, and POI locations are classified according to the tags provided by the source data. The scale of analysis is from macro to micro, and they are the total POI visits data of New York State based on different ways in 2019 and 2020, the POI visits of CBG (Census Block Group) division in New York City, and three representative POI samples to do individual analysis. The analysis methods are: (1) use line plot and bar plot statistics to compare the trends of POI visits data from 2019 to 2020, and (2) make the spatial visualization comparison, which includes grid map, scatter map, heatmap, and OD map, between the first peak of epidemic impact in the first full week of April 2019 and April 2020, and the scope is narrowed to New York City. Wherein the OD maps are drawn based on the CBG division. Compared to related work, the analysis object includes CBG, categories, and individual POI. In addition, the analysis method combines statistical graphs and spatial visualizations and explores the policy impact of the New York City government. This report adopts more multidimensional analysis methods and objects to improve the comprehensiveness and reliability of the analysis content.

Citywide crowd prediction can be of great importance for emergency management, traffic regulation, and urban planning. By meshing a large urban area to a number of fine-grained mesh-grids as illustrated in Fig. 1 , citywide crowd in a continuous time period can be represented with a four-dimensional tensor ${\mathbb{R}^{Timestep{\text{ }}p \times {\text{ }}Height{\text{ }} \times {\text{ }}Width{\text{ }} \times {\text{ }}Channel{\text{ }}}}$ in an analogous manner to video data, where each Timestep can be seen as one video frame, Height , Width is two-dimensional index for mesh-grids, and each Channel stores an aggregated scalar value for each mesh-grid. Specifically, given historical observations of crowd density and in-out flow x d = d 1 ,…,d t , xf = f 1 ,…, f t at timestamp t , we aim to build prediction models for the next-step density and in-out flow y d = d t +1, yf = f t +1, where y d means how many people will be in each mesh-grid at the next timestamp, and yf means how many people will flow into or out from each mesh-grid in next time interval. Al-though many deep models [1] – [6] have been proposed to address such tasks, their actual effects are still not well validated on large-scale and high-quality datasets. The datasets used in most of the works so far are originally generated based on taxi or bicycle trip data, which don’t cover and reflect the citywide crowd density and flow. Thus, we first publish new crowd flow data called BousaiTYO and BousaiOSA [7] . These new datasets are created using the GPS log data collected from a popular smartphone app of Yahoo! Japan Corporation, which can well reflect the real-world crowd flow in Tokyo and Osaka. As shown by Table 1 , our dataset has: (1) larger spatial area; (2) finer mesh size; (3) higher user sample.

In this study, a method of multi-weighted graphs learning for passenger count prediction in railway networks, is presented. Traffic prediction can provide significant insights for railway system optimization, urban planning, smart city development, etc. However, affected by various factors, including spatial, temporal, and other external ones, traffic prediction on railway networks remains a critical task because of the complexity of the railway networks. To achieve high learning performance of the models and discover the correlation between the models and features, we proposed various heterogenerous weighted graphs for the passenger count prediction. Six types of weight graphs, that is, connection graph, distance graph, correlation graph, and their fused weight graphs were proposed to fully construct the spatial and geometrical features within the entire railway network. Two representative types of graph neural networks, that is, the graph convolutional network (GCN) and graph attention network (GAT) were implemented for evaluation. The evaluation results demonstrate that the proposed GAT model learning on the correlation graph achieves the best performance, as it can reduce the metrics of mean absolute error (MAE), root mean square error (RSME), and mean absolute percentage error metrics (MAPE) on average by 19.7%, 6.9%, 27.9% respectively. Finally, the importance and effectiveness of the models with corresponding weight graphs were also investigated and explained. It also provides the interpretability of the traffic prediction tasks on the railway network.

Knowing "what is happening" and "what will happen" of the mobility in a city is the building block of a data-driven smart city system. In recent years, mobility digital twin that makes a virtual replication of human mobility and predicting or simulating the fine-grained movements of the subjects in a virtual space at a metropolitan scale in near real-time has shown its great potential in modern urban intelligent systems. However, few studies have provided practical solutions. The main difficulties are four-folds: 1) the daily variation of human mobility is hard to model and predict; 2) the transportation network enforces a complex constraints on human mobility; 3) generating a rational fine-grained human trajectory is challenging for existing machine learning models; and 4) making a fine-grained prediction incurs high computational costs, which is challenging for an online system. Bearing these difficulties in mind, in this paper we propose a two-stage human mobility predictor that stratifies the coarse and fine-grained level predictions. In the first stage, to encode the daily variation of human mobility at a metropolitan level, we automatically extract citywide mobility trends as crowd contexts and predict long-term and long-distance movements at a coarse level. In the second stage, the coarse predictions are resolved to a fine-grained level via a probabilistic trajectory retrieval method, which offloads most of the heavy computations to the offline phase. We tested our method using a real-world mobile phone GPS dataset in the Kanto area in Japan, and achieved good prediction accuracy and a time efficiency of about 2 min in predicting future 1h movements of about 220K mobile phone users on a single machine to support more higher-level analysis of mobility prediction.

Predicting the density and flow of the crowd at a citywide level is significant for city management. By meshing a large urban area to a number of fine-grained mesh-grids, citywide crowd and traffic information in a continuous time period can be represented with 4D tensor (Timestep, Height, Width, Channel). Based on this, we revisit the density and in-out flow prediction problem and publish a new aggregated human mobility dataset generated from a real-world smartphone application. Compared with the existing ones, our dataset has larger mesh-grid number, finer-grained mesh size, and higher user sample. Towards such kind of large-scale crowd dataset, we propose a novel deep learning model called DeepCrowd by designing pyramid architectures and high-dimensional attention mechanism based on Convolutional LSTM. Both the datasets and codes are made available at https://github.com/deepkashiwa20/DeepCrowd.

Although software libraries promote code reuse and facilitate software development, they increase the complexity of programme analysis tasks. To effectively analyse programmes built on top of software libraries, it is essential to have specifications for the library methods that can be easily processed by analysis tools. However, the availability of such specifications is seriously limited at the moment. Manually writing the specifications can be prohibitively expensive and error-prone, while existing automated approaches to inferring the specifications seldom produce results that are strong enough to be used in programme analysis. In this work, we propose the DOC2SMT approach to generating strong functional constraints in SMT for library methods based on their documentations. DOC2SMT first applies natural language processing (NLP) techniques and a set of rules to translate a method's natural language documentation into a large number of candidate constraint clauses in OCL. Then, it utilises a manually enhanced domain model to identify OCL candidate constraint clauses that comply with the problem domain in static validation, translates well-formed OCL constraints into the SMT-LIB format, and checks whether each 5MB-LIB constraint rightly abstracts the functionalities of the method under consideration via testing in dynamic validation. In the end, it reports the first functional constraint that survives both validations to the user as the result. We have implemented the approach into a supporting tool with the same name. In experiments conducted on 451 methods from the Java Collections Framework and the Java IO library, DOC2SMT generated correct constraints for 309 methods, with the average generation time for each correct constraint being merely 2.7 min. We have also applied the generated constraints to facilitate symbolic-execution-based test generation with the Symbolic Java PathFinder (SPF) tool. For 24 utility methods manipulating Java container and IO objects, SPF with access to the generated constraints produced 51.2 times more test cases than SPF without the access.

Event crowd management has been a significant research topic with high social impact. When some big events happen such as an earthquake, typhoon, and national festival, crowd management becomes the first priority for governments (e.g., police) and public service operators (e.g., subway/bus operator) to protect people’s safety or maintain the operation of public infrastructures. However, under such event situations, human behavior will become very different from daily routines, which makes prediction of crowd dynamics at big events become highly challenging, especially at a citywide level. Therefore in this study, we aim to extract the “deep” trend only from the current momentary observations and generate an accurate prediction for the trend in the short future, which is considered to be an effective way to deal with the event situations. Motivated by these, we build an online system called DeepUrbanEvent, which can iteratively take citywide crowd dynamics from the current one hour as input and report the prediction results for the next one hour as output. A novel deep learning architecture built with recurrent neural networks is designed to effectively model these highly complex sequential data in an analogous manner to video prediction tasks. Experimental results demonstrate the superior performance of our proposed methodology to the existing approaches. Lastly, we apply our prototype system to multiple big real-world events and show that it is highly deployable as an online crowd management system.

We consider long-term path forecasting problems in crowds, where future sequence trajectories are generated given a short observation. Recent methods for this problem have focused on modeling social interactions and predicting multi-modal futures. However, it is not easy for machines to successfully consider social interactions, such as avoiding collisions while considering the uncertainty of futures under a highly interactive and dynamic scenario. In this paper, we propose a model that incorporates multiple interacting motion sequences jointly and predicts multi-modal socially acceptable distributions of futures. Specifically, we introduce a new aggregation mechanism for social interactions, which selectively models long-term inter-related dynamics between movements in a shared environment through a message passing mechanism. Moreover, we propose a loss function that not only accesses how accurate the estimated distributions of the futures are but also considers collision avoidance. We further utilize mixture density functions to describe the trajectories and learn the multi-modality of future paths. Extensive experiments over several trajectory prediction benchmarks demonstrate that our method is able to forecast socially acceptable distributions in complex scenarios.

Nowadays, with the rapid development of IoT (Internet of Things) and CPS (Cyber-Physical Systems) technologies, big spatiotemporal data are being generated from mobile phones, car navigation systems, and traffic sensors. By leveraging state-of-the-art deep learning technologies on such data, urban traffic prediction has drawn a lot of attention in AI and Intelligent Transportation System community. The problem can be uniformly modeled with a 3D tensor (T, N, C), where T denotes the total time steps, N denotes the size of the spatial domain (i.e., mesh-grids or graph-nodes), and C denotes the channels of information. According to the specific modeling strategy, the state-of-the-art deep learning models can be divided into three categories: grid-based, graph-based, and multivariate time-series models. In this study, we first synthetically review the deep traffic models as well as the widely used datasets, then build a standard benchmark to comprehensively evaluate their performances with the same settings and metrics. Our study named DL-Traff is implemented with two most popular deep learning frameworks, i.e., TensorFlow and PyTorch, which is already publicly available as two GitHub repositories https://github.com/deepkashiwa20/DL-Traff-Grid and https://github.com/deepkashiwa20/DL-Traff-Graph. With DL-Traff, we hope to deliver a useful resource to researchers who are interested in spatiotemporal data analysis.

Forecasting incident occurrences (e.g. crime, EMS, traffic accident) is a crucial task for emergency service providers and transportation agencies in performing response time optimization and dynamic fleet management. However, such events are by nature rare and sparse, which causes the label imbalance problem and inferior performance of models relying on data sufficiency. The existing studies circumvent, instead of truly solving, this issue by defining the incident prediction problem in a coarse-grained temporal (e.g. daily) setting, which leaves the proposed models unrobust to fine-grained dynamics and trivial for the real-world decision making. In this paper, we tackle the temporally fine-grained incident prediction problem in a sparse setting by explicitly exploiting the behind-the-scene chainlike triggering mechanism. Moreover, this chain effect roots in multiple domains (i.e. spatial, categorical), which further entangles with the temporal dimension and happens to be time-variant. To be specific, we propose a novel deep learning framework, namely Spatio-Temporal-Categorical Graph Neural Networks (STC-GNN), to handle the multidimensional and dynamic chain effect for performing fine-grained multi-incident co-prediction. Extensive experiments on three real-world city-level incident datasets verify the insightfulness of our perspective and effectiveness of the proposed model.

Forecasting regional ambulance demand plays a fundamental part in dynamic fleet allocation and redeployment. This topic has been gaining increasing significance, as virtually every country is experiencing an aging population, with generally higher level of vulnerability and demand for the emergency medical service (EMS). Although exploring the spatial and temporal correlations in EMS historical records, the existing methods principally consider the former time-invariant, which does not necessarily hold in reality. Moreover, this assumption ignores the fact that the behind-the-scenes dynamics are people, whose demographic profiles and activity patterns could be determinants of regional EMS demands. In this paper, we are therefore motivated to mine the collective daily routines in human mobility, to further represent the evolving spatial correlations. Particularly, we model profiled mobility groups as multiple random walkers and propose a novel bicomponent neural network, including a heterogeneous multi-graph convolution layer and spatio-temporal interlacing attention module, to perform the prediction task. Experimental results on the real-world data verify the effectiveness of introducing dynamic human mobility and the advantage of our approach over the state-of-the-art models.

Multi-variate time series (MTS) data is a ubiquitous class of data abstraction in the real world. Any instance of MTS is generated from a hybrid dynamical system with their specific dynamics normally unknown. The hybrid nature of such a dynamical system is a result of complex external impacts, which can be summarized as high-frequency and low-frequency from the temporal view, or global and local if we take the spatial view. These impacts also determine the forthcoming development of MTS making them paramount to capture in a time series forecasting task. However, conventional methods face intrinsic difficulties in disentangling the components yielded by each kind of impact from the raw data. To this end, we propose two kinds of normalization modules -- temporal and spatial normalization -- which separately refine the high-frequency component and the local component underlying the raw data. Moreover, both modules can be readily integrated into canonical deep learning architectures such as Wavenet and Transformer. Extensive experiments on three datasets are conducted to illustrate that, with additional normalization modules, the performance of the canonical architectures can be enhanced by a large margin in the application of MTS and achieves state-of-the-art results compared with existing MTS models.

Summary Although software libraries promote code reuse and facilitate software development, they increase the complexity of programme analysis tasks. To effectively analyse programmes built on top of software libraries, it is essential to have specifications for the library methods that can be easily processed by analysis tools. However, the availability of such specifications is seriously limited at the moment. Manually writing the specifications can be prohibitively expensive and error‐prone, while existing automated approaches to inferring the specifications seldom produce results that are strong enough to be used in programme analysis. In this work, we propose the Doc2smt approach to generating strong functional constraints in SMT for library methods based on their documentations. Doc2smt first applies natural language processing (NLP) techniques and a set of rules to translate a method's natural language documentation into a large number of candidate constraint clauses in OCL. Then, it utilizes a manually enhanced domain model to identify OCL candidate constraint clauses that comply with the problem domain in static validation, translates well‐formed OCL constraints into the SMT‐LIB format, and checks whether each SMB‐LIB constraint rightly abstracts the functionalities of the method under consideration via testing in dynamic validation. In the end, it reports the first functional constraint that survives both validations to the user as the result. We have implemented the approach into a supporting tool with the same name. In experiments conducted on 451 methods from the Java Collections Framework and the Java IO library, Doc2smt generated correct constraints for 309 methods, with the average generation time for each correct constraint being merely 2.7 min. We have also applied the generated constraints to facilitate symbolic‐execution‐based test generation with the Symbolic Java PathFinder (SPF) tool. For 24 utility methods manipulating Java container and IO objects, SPF with access to the generated constraints produced 51.2 times more test cases than SPF without the access.

Transportation demand forecasting is a topic of large practical value. However, the model that fits the demand of one transportation by only considering the historical data of its own could be vulnerable since random fluctuations could easily impact the modeling. On the other hand, common factors like time and region attribute, drive the evolution demand of different transportation, leading to a co-evolving intrinsic property between different kinds of transportation. In this work, we focus on exploring the co-evolution between different modes of transport, e.g., taxi demand and shared-bike demand. Two significant challenges impede the discovery of the co-evolving pattern: (1) diversity of the co-evolving correlation, which varies from region to region and time to time. (2) Multi-modal data fusion. Taxi demand and shared-bike demand are time-series data, which have different representations with the external factors. Moreover, the distribution of taxi demand and bike demand are not identical. To overcome these challenges, we propose a novel method, known as co-evolving spatial temporal neural network (CEST). CEST learns a multi-view demand representation for each mode of transport, extracts the co-evolving pattern, then predicts the demand for the target transportation based on multi-scale representation, which includes fine-scale demand information and coarse-scale pattern information. We conduct extensive experiments to validate the superiority of our model over the state-of-art models.

Accurate human path forecasting in complex and crowded scenarios is critical for collision avoidance of autonomous driving and social robots navigation. It still remains as a challenging problem because of dynamic human interaction and intrinsic multimodality of human motion. Given the observation, there is a rich set of plausible ways for an agent to walk through the circumstance. To address those issues, we propose a spatio-temporal model that can aggregate the information from socially interacting agents and capture the multimodality of the motion patterns. We use mixture density functions to describe the human path and predict the distribution of future paths with explicit density. To integrate more factors to model interacting people, we further introduce a coordinate transformation to represent the relative motion between people. Extensive experiments over several trajectory prediction benchmarks demonstrate that our method is able to forecast various plausible futures in complex scenarios and achieves state-of-the-art performance.

Nowadays, GPS devices have increased explosively and produced huge amounts of trajectory data related to people's outgoing. Through those big location data, many researches aim to analyze human mobility for urban development, such as human movement prediction/modeling, POI (Point-Of-Interest) recommendation. However, trajectory data only contains timestamp and location information. The intention of human movement is not explicit so that it is hard to understand why people go to somewhere. The intention prior to the activity could be of great significance for analyzing and predicting human mobility, which has not been taken into consideration by the existing researches until the present. Thus, in this study, we propose a brand-new concept called human intentional mobility, aiming to employ intention information to predict people's outgoing. We carefully utilize user's search query to sense his intention as well as the intensity. For instance, if a user searches a certain POI for many times in a short period, it will represent a relatively high intention to go there. Then, to fully utilize this intention representation for predicting whether user will visit searched POI or not, we specially design Dual Sequential Interaction Network (DualSIN) as a novel and unique deep-learning model, which can effectively capture the sophisticated interactions among two kinds of sequential information (i.e., search sequence and mobility sequence) and typical categorical information (i.e., user attributes). Last, we evaluate our model on real-world dataset collected from Yahoo! Japan portal application, and demonstrate that it can achieve superior satisfactory performances to the-state-of-the-art models on multiple POI search queries.

Extracting identifiable information from human trajectories is a fundamental task in many location-based services (LBS). However, various mobility patterns underlain in human trajectories are difficult to model by existing models. Moreover, we could hardly define a clear user set for user identification because the set of users are dynamic and changing everyday. Bearing these in mind, we apply a deep multiple instance learning method to handle the multimodal mobility patterns in a weak-supervised learning way, and address the dynamic user set problems via a pairwise loss with negative sampling. We utilize a multi-head attention mechanism to automatically extract multiple aspects and match the corresponding information between query trajectories and historical trajectories. Our method shows a good identification accuracy on three human GPS trajectory data sets comparing with baseline methods.

Event crowd management has been a significant research topic with high social impact. When some big events happen such as an earthquake, typhoon, and national festival, crowd management becomes the first priority for governments (e.g. police) and public service operators (e.g. subway/bus operator) to protect people's safety or maintain the operation of public infrastructures. However, under such event situations, human behavior will become very different from daily routines, which makes prediction of crowd dynamics at big events become highly challenging, especially at a citywide level. Therefore in this study, we aim to extract the deep trend only from the current momentary observations and generate an accurate prediction for the trend in the short future, which is considered to be an effective way to deal with the event situations. Motivated by these, we build an online system called DeepUrbanEvent which can iteratively take citywide crowd dynamics from the current one hour as input and report the prediction results for the next one hour as output. A novel deep learning architecture built with recurrent neural networks is designed to effectively model these highly-complex sequential data in an analogous manner to video prediction tasks. Experimental results demonstrate the superior performance of our proposed methodology to the existing approaches. Lastly, we apply our prototype system to multiple big real-world events and show that it is highly deployable as an online crowd management system.

Big human mobility data are being continuously generated through a variety of sources, some of which can be treated and used as streaming data for understanding and predicting urban dynamics. With such streaming mobility data, the online prediction of short-term human mobility at the city level can be of great significance for transportation scheduling, urban regulation, and emergency management. In particular, when big rare events or disasters happen, such as large earthquakes or severe traffic accidents, people change their behaviors from their routine activities. This means people's movements will almost be uncorrelated with their past movements. Therefore, in this study, we build an online system called DeepUrbanMomentum to conduct the next short-term mobility predictions by using (the limited steps of) currently observed human mobility data. A deep-learning architecture built with recurrent neural networks is designed to effectively model these highly complex sequential data for a huge urban area. Experimental results demonstrate the superior performance of our proposed model as compared to the existing approaches. Lastly, we apply our system to a real emergency scenario and demonstrate that our system is applicable in the real world.

As the efficiency of constraint solvers increases, constraint solving has been widely used in many applications of software engineering, such as test case generation, program synthesis and code search. However, encoding source code into constraints is not an easy task. Particularly, when it comes to complex data structures of library functions, existing work cannot generate effective constraints.

As the quality and quantity of open source code increase, semantics-based code search has become an emerging need for software developers to retrieve and reuse existing source code. We present an approach of semantics-based code search using input/output examples for the Java language. Our approach encodes Java methods in code repositories into path constraints via symbolic analysis and leverages SMT solvers to find the methods whose path constraints can satisfy the given input/output examples. Our approach extends the applicability of the semantics-based search technology to more general Java code compared with existing methods. To evaluate our approach, we encoded 1228 methods from GitHub and applied semantics-based code search on 35 queries extracted from Stack Overflow. Correct method code for 29 queries was obtained during the search and the average search time was just about 48 seconds.

Globalization leads to more and more outbound behavior. Understanding these behavior is of great significance to both outbound and inbound countries in a lot of fields. However, the previous research on outbound behavior mainly unitizes questionnaires or purely analyzes the statistical data, these data are either too difficult to acquire or unable to directly analyze public opinion. This paper aim to analyze outbound behavior through large volume of text and location information published on social network services (SNS) and look into a case study of Chinese people in Japan with the data acquired from Chinese microblog website, Sina Weibo. For text analysis, we utilize an adapted latent Dirichlet allocation model to extract the topics in text and discuss the spatial and temporal distribution of the topics. The experiment proves that the topics represent outbound behavior of tourists and foreign visitors can be found and explained in the experiment dataset, which indicates that SNS data is valid for analyzing outbound behavior.