Many multi-agent reinforcement learning (MARL) algorithms are trained in fixed simulation environments, making them brittle when deployed in real-world scenarios with more complex and uncertain conditions. Contextual MARL (cMARL) addresses this by parameterizing environments with context variables and training a context-agnostic policy that performs well across all environment configurations. Existing cMARL methods attempt to use curriculum learning to help train and evaluate context-agnostic policies, but they often rely on unreliable proxy signals, such as value estimates or generalized advantage estimates that are noisy and unstable in multi-agent settings due to inter-agent dynamics and partial observability. To address these issues, we propose Contextual Multi-Agent LLM-Guided Curriculum Learning with Diversity-Based Context Blending (cMALC-D), a framework that uses Large Language Models (LLMs) to generate semantically meaningful curricula and provide a more robust evaluation signal. To prevent mode collapse and encourage exploration, we introduce a novel diversity-based context blending mechanism that creates new training scenarios by combining features from prior contexts. Experiments in traffic signal control domains demonstrate that cMALC-D improves both generalization and sample efficiency compared to existing curriculum learning baselines.

Explaining Graph Neural Networks (GNNs) has garnered significant attention due to the need for interpretability, enabling users to understand the behavior of these black-box models better and extract valuable insights from their predictions.While numerous post-hoc instance-level explanation methods have been proposed to interpret GNN predictions, the reliability of these explanations remains uncertain, particularly in out-of-distribution or unknown test datasets.In this paper, we address this challenge by introducing an explainer framework with the confidence scoring module (ConfExplainer), grounded in theoretical principle, which is a generalized graph information bottleneck with confidence constraint (GIB-CC), that quantifies the reliability of generated explanations.Experimental results demonstrate the superiority of our approach, highlighting the effectiveness of the confidence score in enhancing the trustworthiness and robustness of GNN explanations.

KDD ’25, August 3–7, 2025, Toronto, ON, Canada

Uncertainty quantification (UQ) enhances the reliability of Large Language Models (LLMs) by estimating confidence in outputs, enabling risk mitigation and selective prediction. However, traditional UQ methods struggle with LLMs due to computational constraints and decoding inconsistencies. Moreover, LLMs introduce unique uncertainty sources, such as input ambiguity, reasoning path divergence, and decoding stochasticity, that extend beyond classical aleatoric and epistemic uncertainty. To address this, we introduce a new taxonomy that categorizes UQ methods based on computational efficiency and uncertainty dimensions, including input, reasoning, parameter, and prediction uncertainty. We evaluate existing techniques, summarize existing benchmarks and metrics for UQ, assess their real-world applicability, and identify open challenges, emphasizing the need for scalable, interpretable, and robust UQ approaches to enhance LLM reliability.

The field of information retrieval has significantly transformed with the integration of AI technologies. AI agents, especially those leveraging LLMs and vast computational power, have revolutionized in- formation retrieval, processing, and presentation. LLM agents, with advanced memory, reasoning, and planning capabilities, can perform complex tasks, engage in coherent conversations, and provide personalized responses. Despite these advancements, challenges such as ensuring relevance and accuracy, mitigating biases, providing real-time responses, and maintaining data security remain. This workshop aims to explore these challenges, share innovative solutions, and discuss future directions. It will provide a platform to bring together researchers and practitioners to discuss the latest theoretical advancements and practical implementations of AI agents in information retrieval. Topics include AI in search, recommendation, and personalization systems. By gathering a diverse group of experts, the workshop seeks to deepen the understanding of AI agents in information retrieval, advance the field, and enhance its societal impact. Participants will gain insights into cutting-edge research and emerging trends, and foster knowledge exchange and collaboration within the community.

Uncertainty reasoning and quantification play a critical role in decision making across various domains, prompting increased attention from both academia and industry. As real-world applications become more complex and data-driven, effectively handling uncertainty becomes paramount for accurate and reliable decision making. This workshop focuses on the critical topics of uncertainty reasoning and quantification in decision making. It provides a platform for experts and researchers from diverse backgrounds to exchange ideas on cutting-edge techniques and challenges in this field. The interdisciplinary nature of uncertainty reasoning and quantification, spanning artificial intelligence, machine learning, statistics, risk analysis, and decision science, will be explored. The workshop aims to address the need for robust and interpretable methods for modeling and quantifying uncertainty, fostering reasoning decision-making in various domains. Participants will have the opportunity to share research findings and practical experiences, promoting collaboration and advancing decision-making practices under uncertainty.

The privacy concerns associated with the use of Large Language Models (LLMs) have grown dramatically with the development of pioneer LLMs such as ChatGPT. Differential Privacy (DP) techniques that utilize DP-SGD are explored in existing work to mitigate their privacy risks at the cost of generalization degradation. Our paper reveals that the flatness of DP-SGD trained models’ loss landscape plays an essential role in the trade-off between their privacy and generalization. We further propose a holistic framework Privacy-Flat to enforce appropriate weight flatness, which substantially improves model generalization with promising privacy protection. It innovates from three coarse-to-grained levels: Perturbation-aware min-max optimization within a layer, flatness-guided sparse prefix-tuning across layers, and weight knowledge distillation between private & non-private weights copies. We empirically demonstrate that our framework Privacy-Flat outperforms vanilla private training baseline while protecting privacy from membership inference attacks (MIA). Comprehensive experiments of both black-box and white-box scenarios are conducted to demonstrate the effectiveness of our proposal in enhancing generalization. The code link is provided at https://github.com/tiejin98/Privacy_ Flatness.

The integration of autonomous vehicles into urban traffic has great potential to improve efficiency by reducing congestion and optimizing traffic flow systematically. In this paper, we introduce CoMAL (Collaborative Multi-Agent LLMs), a framework designed to address the mixed-autonomy traffic problem by collaboration among autonomous vehicles to optimize traffic flow. CoMAL is built upon large language models and operates in an interactive traffic simulation environment. Specifically, It utilizes a Perception Module to observe surrounding agents and a Memory Module to store strategies for each agent. The overall workflow includes a Collaboration Module that encourages autonomous vehicles to discuss the effective strategy and allocate roles, a reasoning engine to determine optimal behaviors based on assigned roles, and an Execution Module that controls vehicle actions using a hybrid approach combining rule-based models. Experimental results demonstrate that CoMAL achieves superior performance on the Flow benchmark. Additionally, we evaluate the impact of different language models and compare our framework with reinforcement learning approaches. It highlights the strong cooperative capability of LLM agents and presents a promising solution to the mixed-autonomy traffic challenge. The code is available at https://github.com/Hyan-Yao/CoMAL

Due to detector malfunctions and communication failures, missing data is ubiquitous during the collection of traffic data. Therefore, it is of vital importance to impute the missing values to facilitate data analysis and decision-making for Intelligent Transportation System (ITS) . However, existing imputation methods generally perform zero pre-filling techniques to initialize missing values, introducing inevitable noise. Moreover, we observe prevalent over-smoothed interpolations, falling short in revealing the intrinsic spatio-temporal correlations of incomplete traffic data. To this end, we propose Mask-Aware Graph Imputation Network (MagiNet) . Our method designs an adaptive mask spatio-temporal encoder to learn the latent representations of incomplete data, eliminating the reliance on pre-filling missing values. Furthermore, we devise a spatio-temporal decoder that stacks multiple blocks to capture the inherent spatial and temporal dependencies within incomplete traffic data, alleviating over-smoothed imputation. Extensive experiments demonstrate that our method outperforms state-of-the-art imputation methods on five real-world traffic datasets, yielding an average improvement of 4.31% in RMSE and 3.72% in MAPE under Missing Completely at Random (MCAR) pattern. Code is available at https://github.com/JeremyChou28/MagiNet .

Urban traffic is subject to disruptions that cause extended waiting time and safety issues at signalized intersections. While numerous studies have addressed the issue of intelligent traffic systems in the context of various disturbances, traffic signal malfunction, a common real-world occurrence with significant repercussions, has received comparatively limited attention. The primary objective of this research is to mitigate the adverse effects of traffic signal malfunction, such as traffic congestion and collision, by optimizing the control of neighboring functioning signals. To achieve this goal, this paper presents a novel traffic signal control framework (MalLight), which leverages an Influence-aware State Aggregation Module (ISAM) and an Influence-aware Reward Aggregation Module (IRAM) to achieve coordinated control of surrounding traffic signals. To the best of our knowledge, this study pioneers the application of a Reinforcement Learning(RL)-based approach to address the challenges posed by traffic signal malfunction. Empirical investigations conducted on real-world datasets substantiate the superior performance of our proposed methodology over conventional and deep learning-based alternatives in the presence of signal malfunction, with reduction of throughput alleviated by as much as 48.6%.

Multi-modal transportation leverages the advantages of various transportation modes, leading to more efficient urban traveling services. Accurately predicting transfer times between different modes provides guidance for tasks such as trip planning and transportation management. Most existing transfer time prediction works rely on strong assumptions, e.g., predetermined routes, assumed speeds, and predefined downstream transportation timetables. However, these assumptions are hard to hold in practice due to internal factors like individual preferences and external factors like dynamic traffic conditions. These factors are dynamic and vary with location and time, presenting a significant challenge. To address this, we introduce an adaptive transfer time prediction framework, AdaTrans, to forecast personalized transfer times between upstream and downstream transportation modes. Firstly, an attribute learning module is designed to model the trends of internal factors. Then a spatial-temporal adaptive learning component is designed to learn dynamic external factors. Finally, an aggregation component with a capsule network is employed to fuse the influences of these factors. The extensive evaluation results in two real-world datasets demonstrate that AdaTrans effectively harnesses insights from internal and external factors, outperforming state-of-the-art methods by ~20%.

Effective multi-intersection collaboration is pivotal for reinforcement-learning-based traffic signal control to alleviate congestion. Existing work mainly chooses neighboring intersections as collaborators. However, quite a lot of congestion, even some wide-range congestion, is caused by non-neighbors failing to collaborate. To address these issues, we propose to separate the collaborator selection as a second policy to be learned, concurrently being updated with the original signal-controlling policy. Specifically, the selection policy in real-time adaptively selects the best teammates according to phase- and intersection-level features. Empirical results on both synthetic and real-world datasets provide robust validation for the superiority of our approach, offering significant improvements over existing state-of-the-art methods. Code is available at https://github.com/bonaldli/CoSLight.

Sentiment Analysis (SA) is a fundamental task in natural language processing, which is widely used in public decision-making. Recently, deep learning have demonstrated great potential to deal with this task. However, prior works have mostly treated SA as a deterministic classification problem, and meanwhile, without quantifying the predictive uncertainty. This presents a serious problem in the SA, different annotator, due to the differences in beliefs, values, and experiences, may have different perspectives on how to label the text sentiment. Such situation will lead to inevitable data uncertainty and make the deterministic classification models feel puzzle to make decision. To address this issue, we propose a new SA paradigm with the consideration of uncertainty and conduct an expensive empirical study. Specifically, we treat SA as the regression task and introduce uncertainty quantification to obtain confidence intervals for predictions, which enables the risk assessment ability of the model and can improve the credibility of SA-aids decision-making. Experiments on five datasets show that our proposed new paradigm effectively quantifies uncertainty in SA while remaining competitive performance to point estimation, in addition to being capable of Out-Of-Distribution~(OOD) detection.

Traffic prediction is a crucial topic because of its broad scope of applications in the transportation domain. Though recent studies have achieved promising results, most of them cannot adequately deal with positions with no historical data, which is common due to limited resources in real life. Apart from this, the lack of uncertainty measurements also makes current models unable to manage risks, especially for the downstream tasks involving decision-making. Inspired by the previous inductive graph neural network, we proposed an uncertainty-aware framework to 1) extend prediction to locations with no historical records and significantly extend spatial coverage of prediction while reducing sensor deployment and 2) generate probabilistic prediction with uncertainty quantification to help the risk management. The experiment results show that our method achieved promising results on prediction tasks, and the uncertainty quantification gives consistent results that highly correlate with the locations with and without historical data. We also show that our model could help support sensor deployment tasks in the transportation field to achieve higher accuracy with a limited sensor deployment budget.

Graph Neural Networks (GNNs) have received increasing attention due to their ability to learn from graph-structured data. However, their predictions are often not interpretable. Post-hoc instance-level explanation methods have been proposed to understand GNN predictions. These methods seek to discover substructures that explain the prediction behavior of a trained GNN. In this paper, we shed light on the existence of the distribution shifting issue in existing methods, which affects explanation quality, particularly in applications on real-life datasets with tight decision boundaries. To address this issue, we introduce a generalized Graph Information Bottleneck (GIB) form that includes a label-independent graph variable, which is equivalent to the vanilla GIB. Driven by the generalized GIB, we propose a graph mixup method, MixupExplainer, with a theoretical guarantee to resolve the distribution shifting issue. We conduct extensive experiments on both synthetic and real-world datasets to validate the effectiveness of our proposed mixup approach over existing approaches. We also provide a detailed analysis of how our proposed approach alleviates the distribution shifting issue.

Traffic is the pulse of the city. Transportation systems can involve humans, vehicles, shipments, information technology, and the physical infrastructure, all interacting in complex ways. Intelligent transportation enables the city to function in a more efficient and effective way. A wide range of city data become increasingly available, such as taxi trips, surveillance camera data, human mobility data from mobile phones or location-based services, events from social media, car accident reports, bike-sharing information, Points-Of-Interest, traffic sensors, public transportation data, and many more. This abundance of data poses a grand challenge to the CIKM research community: How to utilize such data toward city intelligence, across various transportation tasks? The 3rd workshop of "Data-driven Intelligent Transportation" welcomes articles and presentations in the areas of transportation systems, data mining, and artificial intelligence, conveying new advances and developments in theory, modeling, simulation, testing, case studies, as well as large-scale deployment.

Modeling how network-level traffic flow changes in the urban environment is useful for decision-making in transportation, public safety and urban planning. The traffic flow system can be viewed as a dynamic process that transits between states (e.g., traffic volumes on each road segment) over time. In the real-world traffic system with traffic operation actions like traffic signal control or reversible lane changing, the system's state is influenced by both the historical states and the actions of traffic operations. In this paper, we consider the problem of modeling network-level traffic flow under a real-world setting, where the available data is sparse (i.e., only part of the traffic system is observed). We present DTIGNN, an approach that can predict network-level traffic flows from sparse data. DTIGNN models the traffic system as a dynamic graph influenced by traffic signals, learns the transition models grounded by fundamental transition equations from transportation, and predicts future traffic states with imputation in the process. Through comprehensive experiments, we demonstrate that our method outperforms state-of-the-art methods and can better support decision-making in transportation.

Understanding city-wide traffic problems may benefit many downstream applications, such as city planning and public transportation development. One key step to understand traffic is to reveal how many people travel from one location to another during one period (we call TOD, short for temporal origin-destination). With TOD, we can rebuild the city-wide traffic by simulating the volume and speed on each road segment.Frequently used mobility data, e.g., GPS trajectories, surveillance cameras, can only cover a subset of vehicles or selected regions of the city. Hence, we propose to use pervasive speed data to recover TOD, and use other mobility data as auxiliary data. To the best of our knowledge, we are the first to work on this challenging problem. It is highly challenging because the speed is generated from a complex process from TOD, and there exists multiple TOD distributions that may generate similar city-wide road speed observations. We propose a new method that models the complex process via separate modules and takes auxiliary data to eliminate infeasible solutions. Extensive experiments on synthetic and real datasets have shown the superior performance of our model over baselines.

Cooperation among the traffic signals enables vehicles to move through intersections more quickly. Conventional transportation approaches implement cooperation by pre-calculating the offsets between two intersections. Such pre-calculated offsets are not suitable for dynamic traffic environments. To enable cooperation of traffic signals, in this paper, we propose a model, CoLight, which uses graph attentional networks to facilitate communication. Specifically, for a target intersection in a network, CoLight can not only incorporate the temporal and spatial influences of neighboring intersections to the target intersection, but also build up index-free modeling of neighboring intersections. To the best of our knowledge, we are the first to use graph attentional networks in the setting of reinforcement learning for traffic signal control and to conduct experiments on the large-scale road network with hundreds of traffic signals. In experiments, we demonstrate that by learning the communication, the proposed model can achieve superior performance against the state-of-the-art methods.

Increasingly available city data and advanced learning techniques have empowered people to improve the efficiency of our city functions. Among them, improving urban transportation efficiency is one of the most prominent topics. Recent studies have proposed to use reinforcement learning (RL) for traffic signal control. Different from traditional transportation approaches which rely heavily on prior knowledge, RL can learn directly from the feedback. However, without a careful model design, existing RL methods typically take a long time to converge and the learned models may fail to adapt to new scenarios. For example, a model trained well for morning traffic may not work for the afternoon traffic because the traffic flow could be reversed, resulting in very different state representation. In this paper, we propose a novel design called FRAP, which is based on the intuitive principle of phase competition in traffic signal control: when two traffic signals conflict, priority should be given to one with larger traffic movement (i.e., higher demand). Through the phase competition modeling, our model achieves invariance to symmetrical cases such as flipping and rotation in traffic flow. By conducting comprehensive experiments, we demonstrate that our model finds better solutions than existing RL methods in the complicated all-phase selection problem, converges much faster during training, and achieves superior generalizability for different road structures and traffic conditions.

Traffic signal control is essential for transportation efficiency in road networks. It has been a challenging problem because of the complexity in traffic dynamics. Conventional transportation research suffers from the incompetency to adapt to dynamic traffic situations. Recent studies propose to use reinforcement learning (RL) to search for more efficient traffic signal plans. However, most existing RL-based studies design the key elements - reward and state - in a heuristic way. This results in highly sensitive performances and a long learning process. To avoid the heuristic design of RL elements, we propose to connect RL with recent studies in transportation research. Our method is inspired by the state-of-the-art method max pressure (MP) in the transportation field. The reward design of our method is well supported by the theory in MP, which can be proved to be maximizing the throughput of the traffic network, i.e., minimizing the overall network travel time. We also show that our concise state representation can fully support the optimization of the proposed reward function. Through comprehensive experiments, we demonstrate that our method outperforms both conventional transportation approaches and existing learning-based methods.

The proliferation of publicly accessible urban data provide new insights on various urban tasks. A frequently used approach is to treat each region as a data sample and build a model over all the regions to observe the correlations between urban features (e.g., demographics) and the target variable (e.g., crime count). To define regions, most existing studies use fixed grids or pre-defined administrative boundaries (e.g., census tracts or community areas). In reality, however, definitions of regions should be different depending on tasks (e.g., regional crime count prediction vs. real estate prices estimation). In this paper, we propose a new problem of task-specific city region partitioning, aiming to find the best partition in a city w.r.t. a given task. We prove this is an NP-hard search problem with no trivial solution. To learn the partition, we first study two variants of Markov Chain Monte Carlo (MCMC). We further propose a reinforcement learning scheme for effective sampling the search space. We conduct experiments on two real datasets in Chicago (i.e., crime count and real estate price) to demonstrate the effectiveness of our proposed method.

Ride-hailing applications have been offering convenient ride services for people in need. However, such applications still suffer from the issue of supply-demand disequilibrium, which is a typical problem for traditional taxi services. With effective predictions on passenger demands, we can alleviate the disequilibrium by pre-dispatching, dynamic pricing or avoiding dispatching cars to zero-demand areas. Existing studies of demand predictions mainly utilize limited data sources, trajectory data, or orders of ride services or both of them, which also lacks a multi-perspective consideration. In this article, we present a unified framework with a new combined model and a road-network-based spatial partition to leverage multi-source data and model the passenger demands from temporal, spatial, and zero-demand-area perspectives. In addition, our framework realizes offline training and online predicting, which can satisfy the real-time requirement more easily. We analyze and evaluate the performance of our combined model using the actual operational data from UCAR. The experimental results indicate that our model outperforms baselines on both Mean Absolute Error and Root Mean Square Error on average.

The intelligent traffic light control is critical for an efficient transportation system. While existing traffic lights are mostly operated by hand-crafted rules, an intelligent traffic light control system should be dynamically adjusted to real-time traffic. There is an emerging trend of using deep reinforcement learning technique for traffic light control and recent studies have shown promising results. However, existing studies have not yet tested the methods on the real-world traffic data and they only focus on studying the rewards without interpreting the policies. In this paper, we propose a more effective deep reinforcement learning model for traffic light control. We test our method on a large-scale real traffic dataset obtained from surveillance cameras. We also show some interesting case studies of policies learned from the real data.

Chauffeured car service based on mobile applications like Uber or Didi suffers from supply-demand disequilibrium, which can be alleviated by proper prediction on the distribution of passenger demand. In this paper, we propose a Zero-Grid Ensemble Spatio Temporal model (ZEST) to predict passenger demand with four predictors: a temporal predictor and a spatial predictor to model the influences of local and spatial factors separately, an ensemble predictor to combine the results of former two predictors comprehensively and a Zero-Grid predictor to predict zero demand areas specifically since any cruising within these areas costs extra waste on energy and time of driver. We demonstrate the performance of ZEST on actual operational data from ride-hailing applications with more than 6 million order records and 500 million GPS points. Experimental results indicate our model outperforms 5 other baseline models by over 10% both in MAE and sMAPE on the three-month datasets.