Urban mobility data has significant connections with economic growth and plays an essential role in various smart-city applications. Due to privacy concerns and substantial data collection costs, fine-grained human mobility trajectories are challenging to make publicly available on a large scale. A promising solution to address this issue is trajectory synthesizing, which generates synthetic trajectories that preserve aggregate spatiotemporal distributions. However, existing works often neglect the road network structural constraints or rely on instance-level external supervision, thus limiting their scalability in generating fine-grained and high-fidelity trajectories. In this paper, we propose Cardiff, a coarse-to-fine Cascaded hybrid diffusion-based framework for fine-grained and structure-plausible trajectory generation. By leveraging the hierarchical nature of urban mobility, Cardiff decomposes the generation process into two cascaded levels, i.e., discrete road segment-level and continuous fine-grained GPS-level: (i) At the segment level, to reduce computational costs and redundancy in raw trajectories, we first encode the discrete road segments into low-dimensional latent embeddings and design a diffusion transformer-based latent denoising network for segment-level synthesis. (ii) Taking the first stage of generation as conditions, we then design a fine-grained GPS-level conditional denoising network with a noise augmentation mechanism to achieve road-network-constrained and fine-grained generation. The cascaded progressive generation yields high-fidelity fine-grained trajectories while adhering to road geometry and topology constraints. Experimental results on three large real-world trajectory datasets demonstrate that our method outperforms state-of-the-art baselines in various metrics. The code is available at~ https://github.com/urban-mobility-generation/Cardiff.

Autonomous Mobility-on-Demand (AMoD) systems, powered by advances in robotics, control, and Machine Learning (ML), offer a promising paradigm for future urban transportation. AMoD offers fast and personalized travel services by leveraging centralized control of autonomous vehicle fleets to optimize operations and enhance service performance. However, the rapid growth of this field has outpaced the development of standardized practices for evaluating and reporting results, leading to significant challenges in reproducibility. As AMoD control algorithms become increasingly complex and data-driven, a lack of transparency in modeling assumptions, experimental setups, and algorithmic implementation hinders scientific progress and undermines confidence in the results. This paper presents a systematic study of reproducibility in AMoD research. We identify key components across the research pipeline, spanning system modeling, control problems, simulation design, algorithm specification, and evaluation, and analyze common sources of irreproducibility. We survey prevalent practices in the literature, highlight gaps, and propose a structured framework to assess and improve reproducibility. While focused on AMoD, the principles and practices we advocate generalize to a broader class of cyber-physical systems that rely on networked autonomy and data-driven control. This work aims to lay the foundation for a more transparent and reproducible research culture in the design and deployment of intelligent mobility systems.

Graph Neural Networks (GNNs) deliver strong classification results but often suffer from poor calibration performance, leading to overconfidence or underconfidence. This is particularly problematic in high-stakes applications where accurate uncertainty estimates are essential. Existing post-hoc methods, such as temperature scaling, fail to effectively utilize graph structures, while current GNN calibration methods often overlook the potential of leveraging diverse input information and model ensembles jointly. In the paper, we propose Graph Ensemble Temperature Scaling (GETS), a novel calibration framework that combines input and model ensemble strategies within a Graph Mixture-of-Experts (MoE) architecture. GETS integrates diverse inputs, including logits, node features, and degree embeddings, and adaptively selects the most relevant experts for each node’s calibration procedure. Our method outperforms state-of-the-art calibration techniques, reducing expected calibration error (ECE) by $\geq$ 25% across 10 GNN benchmark datasets. Additionally, GETS is computationally efficient, scalable, and capable of selecting effective input combinations for improved calibration performance. The implementation is available at https://github.com/ZhuangDingyi/GETS/.

Simulating society with large language models (LLMs), we argue, requires more than generating plausible behavior; it demands cognitively grounded reasoning that is structured, revisable, and traceable. LLM-based agents are increasingly used to emulate individual and group behavior, primarily through prompting and supervised fine-tuning. Yet current simulations remain grounded in a behaviorist “demographics in, behavior out” paradigm, focusing on surface-level plausibility. As a result, they often lack internal coherence, causal reasoning, and belief traceability—making them unreliable for modeling how people reason, deliberate, and respond to interventions.To address this, we present a conceptual modeling paradigm, Generative Minds (GenMinds), which draws from cognitive science to support structured belief representations in generative agents. To evaluate such agents, we introduce the RECAP (REconstructing CAusal Paths) framework, a benchmark designed to assess reasoning fidelity via causal traceability, demographic grounding, and intervention consistency. These contributions advance a broader shift: from surface-level mimicry to generative agents that simulate thought—not just language—for social simulations.

Spatiotemporal systems are ubiquitous in a large number of scientific areas, representing underlying knowledge and patterns in the data. Here, a fundamental question usually arises as how to understand and characterize these spatiotemporal systems with a certain data-driven machine learning framework. In this work, we introduce an unsupervised pattern discovery framework, namely, dynamic autoregressive tensor factorization. Our framework is essentially built on the fact that the spatiotemporal systems can be well described by the time-varying autoregression on multivariate or even multidimensional data. In the modeling process, tensor factorization is seamlessly integrated into the time-varying autoregression for discovering spatial and temporal modes/patterns from the spatiotemporal systems in which the spatial factor matrix is assumed to be orthogonal. To evaluate the framework, we apply it to several real-world spatiotemporal datasets, including fluid flow dynamics, international import/export merchandise trade, and urban human mobility. On the international trade dataset with dimensions {country/region, product type, year}, our framework can produce interpretable import/export patterns of countries/regions, while the low-dimensional product patterns are also important for classifying import/export merchandise and understanding systematical differences between import and export. On the ridesharing mobility dataset with dimensions {origin, destination, time}, our framework is helpful for identifying the shift of spatial patterns of urban human mobility that changed between 2019 and 2022. Empirical experiments demonstrate that our framework can discover interpretable and meaningful patterns from the spatiotemporal systems that are both time-varying and multidimensional.

This study addresses the problem of convolutional kernel learning in univariate, multivariate, and multidimensional time series data, which is crucial for interpreting temporal patterns in time series and supporting downstream machine learning tasks. First, we propose formulating convolutional kernel learning for univariate time series as a sparse regression problem with a non-negative constraint, leveraging the properties of circular convolution and circulant matrices. Second, to generalize this approach to multivariate and multidimensional time series data, we use tensor computations, reformulating the convolutional kernel learning problem in the form of tensors. This is further converted into a standard sparse regression problem through vectorization and tensor unfolding operations. In the proposed methodology, the optimization problem is addressed using the existing non-negative subspace pursuit method, enabling the convolutional kernel to capture temporal correlations and patterns. To evaluate the proposed model, we apply it to several real-world time series datasets. On the multidimensional ridesharing and taxi trip data from New York City and Chicago, the convolutional kernels reveal interpretable local correlations and cyclical patterns, such as weekly seasonality. For the monthly temperature time series data in North America, the proposed model can quantify the yearly seasonality and make it comparable across different decades. In the context of multidimensional fluid flow data, both local and nonlocal correlations captured by the convolutional kernels can reinforce tensor factorization, leading to performance improvements in fluid flow reconstruction tasks. Thus, this study lays an insightful foundation for automatically learning convolutional kernels from time series data, with an emphasis on interpretability through sparsity and non-negativity constraints.

Quantifying uncertainty is crucial for robust and reliable predictions. However, existing spatiotemporal deep learning mostly focuses on deterministic prediction, overlooking the inherent uncertainty in such prediction. Particularly, highly-granular spatiotemporal datasets are often sparse, posing extra challenges in prediction and uncertainty quantification. To address these issues, this paper introduces a novel post-hoc Sparsity-aware Uncertainty Calibration (SAUC) framework, which calibrates uncertainty in both zero and non-zero values. To develop SAUC, we firstly modify the state-of-the-art deterministic spatiotemporal Graph Neural Networks (ST-GNNs) to probabilistic ones in the pre-calibration phase. Then we calibrate the probabilistic ST-GNNs for zero and non-zero values using quantile approaches. Through extensive experiments, we demonstrate that SAUC can effectively fit the variance of sparse data and generalize across two real-world spatiotemporal datasets at various granularities. Specifically, our empirical experiments show a 20% reduction in calibration errors in zero entries on the sparse traffic accident and urban crime prediction. Overall, this work demonstrates the theoretical and empirical values of the SAUC framework, thus bridging a significant gap between uncertainty quantification and spatiotemporal prediction.

Public transit systems are the backbone of urban mobility systems in the era of urbanization. The design of transit schedules is important for the efficient and sustainable operation of public transit. However, limited studies have considered demand uncertainties when designing transit schedules. To better address demand uncertainty issues inherent in public transit systems, this paper utilizes the robust optimization (RO) framework to generate robust transit schedules against demand uncertainty. A nominal (non-robust) optimization model for the transit frequency setting problem (TFSP) under a single transit line setting is first proposed. The model is then extended to the RO-based formulation to incorporate demand uncertainty, which has not been considered in the literature. The large-scale origin-destination (OD) matrices for real-world transit problems bring computational challenges in solving the optimization problem. To efficiently generate robust transit schedules, a Transit Downsizing (TD) approach is proposed to reduce the dimensionality of the problem. The proposed models are tested with real-world transit lines and data from the Chicago Transit Authority (CTA). Meanwhile, a stochastic programming (SP) framework is used to construct a benchmark stochastic TFSP model. Compared to the current transit schedule implemented by the CTA, the nominal TFSP model without considering demand uncertainty reduces passengers’ wait times while increasing in-vehicle travel times. After incorporating demand uncertainty, both stochastic and robust TFSP models reduce passengers’ wait times and in-vehicle travel times simultaneously. The robust transit schedules outperform the benchmark stochastic transit schedules by reducing both wait and in-vehicle travel times when demand is significantly uncertain.

Recent studies have significantly improved the prediction accuracy of travel demand using graph neural networks. However, these studies largely ignored uncertainty that inevitably exists in travel demand prediction. To fill this gap, this study proposes a framework of probabilistic graph neural networks (Prob-GNN) to quantify the spatiotemporal uncertainty of travel demand. This Prob-GNN framework is substantiated by deterministic and probabilistic assumptions, and empirically applied to the task of predicting the transit and ridesharing demand in Chicago. We found that the probabilistic assumptions (e.g. distribution tail, support) have a greater impact on uncertainty prediction than the deterministic ones (e.g. deep modules, depth). Among the family of Prob-GNNs, the GNNs with truncated Gaussian and Laplace distributions achieve the highest performance in transit and ridesharing data. Even under significant domain shifts, Prob-GNNs can predict the ridership uncertainty in a stable manner, when the models are trained on pre-COVID data and tested across multiple periods during and after the COVID-19 pandemic. Prob-GNNs also reveal the spatiotemporal pattern of uncertainty, which is concentrated on the afternoon peak hours and the areas with large travel volumes. Overall, our findings highlight the importance of incorporating randomness into deep learning for spatiotemporal ridership prediction. Future research should continue to investigate versatile probabilistic assumptions to capture behavioral randomness, and further develop methods to quantify uncertainty to build resilient cities.

Previous work on misbehavior detection and trust management for vehicle-to-everything (V2X) communication security is effective in identifying falsified and malicious V2X data. Each vehicle in a given region can be a witness to report on the misbehavior of other nearby vehicles, which will then be added to a “blacklist.” However, there may not exist enough witness vehicles that are willing to opt-in in the early stage of connected-vehicle deployment. In this article, we propose a “whitelisting” approach to V2X security, titled proof-of-travel (POT), which leverages the support of roadside infrastructure. Our goal is to transform the power of cryptography techniques embedded within vehicle-to-infrastructure (V2I) protocols into game-theoretic mechanisms to incentivize connected-vehicle data sharing and validate data trustworthiness simultaneously. The key idea is to determine the reputation and the contribution made by a vehicle based on its distance traveled and the information it shared through V2I channels. In particular, the total vehicle miles traveled for a vehicle must be testified by digital signatures signed by each infrastructure component along the path of its movement. While building a chain of proofs of spatial movement creates burdens for malicious vehicles, acquiring proofs does not result in extra costs for normal vehicles, which naturally want to move from the origin to the destination. The POT protocol is used to enhance the security of previous voting-based data validation algorithms for V2I crowdsensing applications. For the POT-enhanced voting, we prove that all vehicles choosing to cheat are not a pure Nash equilibrium using game-theoretic analysis. Simulation results suggest that the POT-enhanced voting is more robust to malicious data.

Origin-Destination (O-D) travel demand prediction is a fundamental challenge in transportation. Recently, spatial-temporal deep learning models demonstrate the tremendous potential to enhance prediction accuracy. However, few studies tackled the uncertainty and sparsity issues in fine-grained O-D matrices. This presents a serious problem, because a vast number of zeros deviate from the Gaussian assumption underlying the deterministic deep learning models. To address this issue, we design a Spatial-Temporal Zero-Inflated Negative Binomial Graph Neural Network (STZINB-GNN) to quantify the uncertainty of the sparse travel demand. It analyzes spatial and temporal correlations using diffusion and temporal convolution networks, which are then fused to parameterize the probabilistic distributions of travel demand. The STZINB-GNN is examined using two real-world datasets with various spatial and temporal resolutions. The results demonstrate the superiority of STZINB-GNN over benchmark models, especially under high spatial-temporal resolutions, because of its high accuracy, tight confidence intervals, and interpretable parameters. The sparsity parameter of the STZINB-GNN has physical interpretation for various transportation applications.

Recent technological developments have shown significant potential for transforming urban mobility. Considering first- and last-mile travel and short trips, the rapid adoption of dockless bike-share systems showed the possibility of disruptive change, while simultaneously presenting new challenges, such as fleet management or the use of public spaces. In this paper, we evaluate the operational characteristics of a new class of shared vehicles that are being actively developed in the industry: scooters with self-repositioning capabilities. We do this by adapting the methodology of shareability networks to a large-scale dataset of dockless bike-share usage, giving us estimates of ideal fleet size under varying assumptions of fleet operations. We show that the availability of self-repositioning capabilities can help achieve up to 10 times higher utilization of vehicles than possible in current bike-share systems. We show that actual benefits will highly depend on the availability of dedicated infrastructure, a key issue for scooter and bicycle use. Based on our results, we envision that technological advances can present an opportunity to rethink urban infrastructures and how transportation can be effectively organized in cities.

Individual mobility is driven by demand for activities with diverse spatiotemporal patterns, but existing methods for mobility prediction often overlook the underlying activity patterns. Knowledge of activity patterns can improve the performance and interpretability of existing individual mobility models, leading to more informed policy design and better user experience in intelligent transportation systems. This study develops an activity-based modeling framework for individual mobility prediction in mass transit systems. Specifically, an input-output hidden Markov model (IOHMM) approach is proposed to simultaneously predict the (continuous) time and (discrete) location of an individual’s next trip using transit smart card data. The prediction task can be transformed into predicting the hidden activity duration and end location. Based on a case study of Hong Kong’s metro system, we show that the proposed model can achieve similar prediction performance as the state-of-the-art long short-term memory (LSTM) model. Unlike LSTM, the proposed IOHMM approach can also be used to analyze hidden activity patterns, which provides meaningful behavioral interpretation for why an individual makes a certain trip. Therefore, the activity-based prediction framework offers a way to preserve the predictive power of advanced machine learning methods while enhancing our ability to generate insightful behavioral explanations, which is useful for user-centric policy design and intelligent transportation applications such as personalized traveler information.

Short-term demand predictions, typically defined as less than an hour into the future, are essential for implementing dynamic control strategies and providing useful customer information in transit applications. Knowing the expected demand enables transit operators to deploy real-time control strategies in advance of the demand surge, and minimize the impact of abnormalities on the service quality and passenger experience. One of the most useful applications of demand prediction models in transit is in predicting the congestion on station platforms and crowding on vehicles. These require information about the origin-destination (OD) demand, providing a detailed profile of how and when passengers enter and exit the service. However, existing work in the literature is limited and overwhelmingly focuses on forecasting passengerarrivalsat stations. This information, while useful, is incomplete for many practical applications. We address this gap by developing a scalable methodology for real-time, short-term OD demand prediction in transit systems. Our proposed model consists of three modules: multi-resolution spatial feature extraction module for capturing the local spatial dependencies with a channel-wise attention block, auxiliary information encoding module (AIE) for encoding the exogenous information, and a module for capturing the temporal evolution of demand. The OD demand at time$t$, represented as a$N \times N$matrix, is processed in two separate branches. In one branch we use the discrete wavelet transform (DWT) to decompose the demand into its different time and frequency variations, detecting patterns that are not visible in the raw data. In the other, three convolutional neural network (CNN) layers are utilized to learn the spatial dependencies from the OD demand directly. Instead of treating each channel of the resultant transformation equally, we use asqueeze-and-excitationlayer to weight feature maps based on their contribution to the final prediction. A Convolutional Long Short-term Memory network (ConvLSTM) is then used to capture the temporal evolution of demand. The approach is demonstrated through a case study using 2 months of Automated Fare Collection (AFC) data from the Hong Kong Mass Transit Railway (MTR) system. The extensive evaluation of the model shows the superiority of our proposed model compared to the other compared methods.

With millions of people using ride-hailing platforms for daily travel, estimated time of arrival (ETA) has become a significant problem in intelligent transportation systems and attracted considerable attention recently. Deep learning-based ETA methods have achieved promising results using massive spatial-temporal data. However, we find that the prediction accuracy is not satisfactory in practical applications due to the prevalent data sparsity problems. Instead of focusing on the average prediction performance as many other methods, this study aims to alleviate the data sparsity problems in ETA to enhance user experience. In general, the data sparsity problems arise from two aspects. The first is the road network, where many links are only traversed by few floating cars. The second aspect is drivers, where many drivers’ trajectories are too scarce (e.g., with only 3 trip records). To alleviate the sparsity in road network, we propose a Road Network Metric Learning framework for ETA (RNML-ETA), where an auxiliary metric learning task is used to improve the link-embedding, especially for links with insufficient data. A novel triangle loss is proposed to improve metric learning effectiveness for links. Experiments on massive real-world data show that RNML-ETA outperforms competing methods by promoting the cold links with limited data. Furthermore, we propose a novel unified framework to Alleviate Data Sparsity problems in ETA (ADS-ETA) by extending RNML-ETA with an additional auxiliary task for driver ID embedding. Results with extensive experiments demonstrate that ADS-ETA can effectively alleviate the data sparsity problems caused by road network and driver sparsity.

Traffic congestion, dominated by single-occupancy vehicles, reflects not only transportation system inefficiency and negative externalities but also a sociological state of human isolation. Advances in information and communication technology are enabling the growth of real-time ridesharing to improve system efficiency. While most ridesharing algorithms optimize fellow passenger matching based on efficiency criteria (maximum number of paired trips, minimum total vehicle-time, or vehicle-distance traveled), very few explicitly consider passengers' preference for their peers as the matching objective. The existing literature either considers the bipartite driver-passenger matching problem, which is structurally different from the monopartite passenger-passenger matching, or only considers the passenger-passenger problem in a simplified one-origin-multiple-destination setting. We formulate a general monopartite passenger matching model in a road network and illustrate the model by pairing 301 430 taxi trips in Manhattan in two scenarios: one considering 1000 randomly generated preference orders and the other considering four sets of group-based preference orders. In both scenarios, compared with efficiency-based matching models, preference-based matching improves the average ranking of paired fellow passenger to the near-top position of people's preference orders with only a small efficiency loss at the individual level and a moderate loss at the aggregate level. The near-top-ranking results fall in a narrow range even with the random variance of passenger preference as inputs.

Regularity is an important property of individual travel behavior, and the ability to measure it enables advances in behavior modeling, mobility prediction, and customer analytics. In this paper, we propose a methodology to measure travel behavior regularity based on the order in which trips or activities are organized. We represent individuals' travel over multiple days as sequences of “travel events”-discrete and repeatable behavior units explicitly defined based on the research question and the available data. We then present a metric of regularity based on entropy rate, which is sensitive to both the frequency of travel events and the order in which they occur. The methodology is demonstrated using a large sample of pseudonymised transit smart card transaction records from London, U.K. The entropy rate is estimated with a procedure based on the Burrows-Wheeler transform. The results confirm that the order of travel events is an essential component of regularity in travel behavior. They also demonstrate that the proposed measure of regularity captures both conventional patterns and atypical routine patterns that are regular but not matched to the 9-to-5 working day or working week. Unlike existing measures of regularity, our approach is agnostic to calendar definitions and makes no assumptions regarding periodicity of travel behavior. The proposed methodology is flexible and can be adapted to study other aspects of individual mobility using different data sources.

Arrival time predictions for the next available bus or train are a key component of modern traveler information systems (TISs). A great deal of research has been conducted within the intelligent transportation system community in developing an assortment of different algorithms that seek to increase the accuracy of these predictions. However, the inherent stochastic and nonlinear nature of these systems, particularly in the case of bus transport, means that these predictions suffer from variable sources of error, stemming from variations in weather conditions, bus bunching, and numerous other sources. In this paper, we tackle the issue of uncertainty in bus arrival time predictions using an alternative approach. Rather than endeavor to develop a superior method for prediction, we take existing predictions from a TIS and treat the algorithm generating them as a black box. The presence of heteroscedasticity in the predictions is demonstrated and then a metamodel approach is deployed, which augments existing predictive systems using quantile regression to place bounds on the associated error. As a case study, this approach is applied to data from a real-world TIS in Boston. This method allows bounds on the predicted arrival time to be estimated, which give a measure of the uncertainty associated with the individual predictions. This represents to the best of our knowledge the first application of methods to handle the uncertainty in bus arrival times that explicitly takes into account the inherent heteroscedasticity. The metamodel approach is agnostic to the process generating the predictions, which ensures the methodology is implementable in any system.