Papers for

smart city infrastructure teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Unified system boosts license plate reading speed on small devices

A Unified Framework for License Plate Identification and Character-Recognition

Abstract: This paper presents a unified framework for license plate detection and character recognition on low-power edge hardware. The proposed network uses a shared CSP-Darknet and PANet feature encoder with an anchor-free localization head and a RoIAlign-based sequence-recognition branch. Character decoding is performed directly on aligned feature maps, avoiding image-space plate cropping and a second recognition backbone. The model is trained on an AI-assisted dataset of approximately 37,000 manually verified images covering license plate layouts from all 50 United States. The dataset includes variations in illumination, viewpoint, weather, blur, occlusion, and plate design. The framework is converted to an INT8 TensorRT engine and evaluated on an NVIDIA Jetson Orin Nano with 8 GB memory. The INT8 model shows an mAP50 of 97.8% and a full-sequence accuracy of 90.7%. It records 147 FPS at 6.8 ms end-to-end latency, with 315 MB inference memory and 9.4 W average power consumption. Relative to the YOLOv8s+CRNN baseline, the proposed deployment provides a 7.1x throughput increase and a 77.8% reduction in allocated inference memory

Sun 8 NovMachine LearningArtificial IntelligenceComputer Vision and Pattern Recognition
The gist
Reading license plates from pictures can be tricky, especially on small, low-power devices like those in cars or traffic cameras. The authors designed a single system that finds license plates and reads their letters directly from image features without extra steps. They trained it with many real license plate images from all U.S. states, including tough conditions like bad weather and blur. Their system runs very fast and uses little memory and power compared to older methods, making it better for devices with limited hardware.
Open → 2611.11289v1

Coordinate fixes boost LiDAR pedestrian detection across real and synthetic data

Coordinate Frame Recovery, Per Class NMS and Null Return Modelling for Sim2Real V2X LiDAR Detection

Abstract: We describe our approach to the V2X-Real track of the UrbanTwin Sim2Real LiDAR Challenge, in which a 3D object detector must be trained exclusively on synthetic data and evaluated on unseen real roadside LiDAR from an Ouster OS1-64 unit, while the synthetic point clouds are simultaneously scored for distributional realism. Our final entry scores 0.7457 combined, from a detection mAP of 0.4382 and a realism score of 0.9251. Three findings account for most of our result, and all three were diagnostic rather than architectural. First, a coordinate frame mismatch between our predictions and the evaluation ground truth was suppressing detection AP entirely; recovering it took our score from 0.3520 to 0.5147 in a single submission. Second, a single global non maximum suppression threshold was destroying genuine pedestrian detections, because pedestrians occur in groups and a 0.3 IoU threshold applied to a 0.7 m box suppresses distinct neighbouring people; correcting this per class raised overall mAP from 0.0751 to 0.4295. Third, approximately 31 percent of the real reference point cloud consists of null returns encoded as the literal point (0, 0, 0), a format feature entirely absent from our synthetic clouds, which pinned Earth Mover's Distance near its floor until we modelled it. Negative results are reported at equal length, and Section 9 gives a full account of data provenance and compliance. That section also accounts for every mechanism contributing to the submitted score.

Mon 2 NovComputer Vision and Pattern RecognitionRobotics
The gist
Detecting objects like pedestrians using LiDAR sensors trained only on computer-generated data can be tricky when used in real life. The authors found that fixing mismatches in how locations are reported, adjusting how overlapping detections are filtered per object type, and modeling missing sensor data significantly improved detection accuracy. Their work focused on improving these practical problems rather than building a new detection method. This led to better alignment between simulated training and real-world testing for roadside LiDAR.
Open → 2611.03940v1

3-D road hazard detection improved by linking LiDAR and camera data

ReinSeg: Retrieval-Guided Bounded Multimodal Refinement for 3-D Road Anomaly Segmentation

Abstract: 3-D road anomaly segmentation is essential for risk-aware perception in autonomous driving. Its central challenge is to rank sparsely observed hazards ahead of abundant normal road structures while preserving point-wise metric localization. LiDAR provides accurate 3-D point support, whereas surround-view imagery offers complementary semantic context for separating unexpected hazards from challenging normal structures. We propose ReinSeg, a retrieval-guided bounded multimodal refinement framework for point-wise road anomaly segmentation. Occupied LiDAR bird's-eye-view cells retrieve scene-specific context from a shared memory of eight surround-view images without supplying calibration variables to the retrieval module. A relative-norm-bounded residual returns this context to the LiDAR point features, so the frozen LIDO anomaly-scoring interface is preserved, and the refinement is optimized with in-distribution semantic labels only. On the Spotting the Unexpected (STU) benchmark, ReinSeg improves validation average precision from 27.54\% to 35.67\% and reduces the false-positive rate at 95\% anomaly recall from 34.85\% to 10.61\%, shrinking the high-recall screening set from 67.54 to 20.55 million points. Relative to REL, AP is 24.99 percentage points higher and FPR95 is 0.99 points lower. Matched interfaces and component isolation show that retrieved scene imagery and the relative-norm bound are both essential, and ReinSeg stays above LiDAR-only performance under missing, corrupted, or misaligned camera inputs while running at 21.8 frames per second. These results establish bounded retrieval of scene-specific visual evidence as an effective route to reliable 3-D road-hazard perception.

Mon 2 NovComputer Vision and Pattern Recognition
The gist
Detecting road hazards in 3-D is hard because many normal road features can look similar to dangers. The authors created ReinSeg, a method that uses LiDAR data combined with images from cameras around the vehicle to better spot unusual road obstacles. Their method searches for matching scenes in stored images and adjusts the LiDAR readings, helping the system tell dangers apart from normal road parts more accurately. Tests show this approach significantly reduces false alarms while keeping fast processing speeds.
Open → 2611.02648v1

Floodwater level detection improves with knowledge-guided cloud-edge system

Knowledge-Guided Floodwater Level Reasoning for Cloud-Edge Collaborative Monitoring

Abstract: Frequent urban floodwater poses a serious threat to the safety and efficiency of intelligent transportation systems. Existing vision-based methods for floodwater level detection face two main limitations. First, the lack of explicit reasoning leads to poor detection performance and little interpretability. Second, these methods struggle to generalize to the complex and varied real-world road scenes seen from different viewpoints. To address this, we reformulate the task as an explicit visual reasoning problem and propose HydroReasoner, a floodwater level reasoning framework designed for cloud-edge collaboration. HydroReasoner builds on Qwen2.5-VL and is trained with Group Relative Policy Optimization. To address the scarcity of reasoning process annotations, we build a floodwater domain knowledge graph HydroKG and design a novel KG path reward. Through deterministic graph traversal over HydroKG, this reward supervises the model's reasoning process at a fine granularity, which improves both reasoning and generalization. For edge deployment, we apply 4-bit quantization to the model, cutting GPU memory usage by 57% and inference latency by 43% while out-of-domain accuracy drops by only 0.2 percentage points. We also propose a confidence-based cloud-edge routing mechanism, which improves system safety and eases the computational load on the cloud. In addition, we build HydroLevel, a floodwater benchmark dataset of 3,000 images, for training and evaluating the model. Experimental results show that HydroReasoner achieves detection accuracy 4.0 and 13.6 percentage points higher than conventional vision baselines in the in-domain and out-of-domain evaluations respectively, with performance dropping by only 4.4 percentage points out of domain.

Sun 1 NovComputer Vision and Pattern RecognitionArtificial Intelligence
The gist
Floodwater on city roads can cause big problems for traffic. Vision systems that try to spot floodwater levels often do badly and can't explain their decisions well. The authors created HydroReasoner, a new system that uses explicit reasoning with a knowledge graph to better detect floodwater levels from images. It splits work between cloud and edge devices to be faster and more efficient. They also made a special dataset to train and test their system, which works better than existing methods even when conditions change.
Open → 2611.01970v1

Energy saving geographic multicast routing for cognitive wireless sensors

Energy-Efficient Geographic Multicast Routing in Cognitive Wireless Sensor Networks

Abstract: Energy consumption is a major design concern in wireless sensor networks. Cognitive radio provides opportunistic access to licensed spectrum, while multicast routing allows several destinations to share forwarding paths. In this paper, we propose two geographic multicast routing protocols for cognitive wireless sensor networks, CCast and PCast. CCast combines multicast-tree construction with a latency-oriented spectrum-selection metric. PCast uses a normalized link-power cost to select routes and can avoid regions where primary users occupy lower-frequency bands. Both protocols use local neighbor information and destination coordinates, without requiring a complete network-topology map. Simulations compare the proposed protocols with geographic multicast routing (GMP) and multiple unicast routing. In simulations, CCast and PCast reduce energy consumption by 21\% and 26\% relative to GMP, at the cost of longer per-destination routes and higher delivery delay.

Sat 31 OctNetworking and Internet Architecture
The gist
Energy use is a big problem for wireless sensor networks that send data to several places at once. This paper presents two new ways to send data efficiently in these networks that can use unused radio channels from licensed users. One method focuses on reducing delay, while the other avoids crowded channels with low frequencies to save power. Both methods only need information about nearby devices and destination locations, not the whole network. Simulations show these new methods save energy compared to previous approaches, though the data may take longer to arrive.
Open → 2611.01212v1

Vision based vehicle counting improves urban traffic signal timing

Vision-Based Density Estimation for Adaptive Traffic Control

Abstract: Traffic congestion is becoming a critical challenge in rapidly growing urban centers, particularly in developing nations where rapid urbanization and inadequate transport infrastructure place increasing strain on transport networks. In highly heterogeneous and chaotic traffic flows, camera-based traffic-state estimation is limited by severe vehicle occlusion and overlap. In this paper, we propose a framework to optimize traffic flow directly from stochastic visual observations. The architecture is structurally decoupled into three primary components: (a) vehicle density estimation, (b) predictive dynamics modeling, and (c) policy optimization. The proposed methodology introduces a convolutional neural network (CNN) that learns the mapping from camera feeds to spatial density maps, utilizing an adaptive k-nearest-neighbor (k-NN) Gaussian kernel to handle varying congestion levels. This feeds into an ST3DNet predictive module for spatiotemporal forecasting, followed by a hierarchical Graph-Transformer Multi-Agent Reinforcement Learning (MARL) algorithm to optimize signal phases. The density estimation stage was evaluated on 40 annotated intersection images from Rambagh 4, Jaipur, India, with a reported five-fold validation mean absolute error (MAE) of $3.32 \pm 0.95$ vehicles and a mean absolute percentage error (MAPE) of 14.5%. These results pertain to the perception component; empirical training and evaluation of the forecasting and control stages remain future work.

Wed 28 OctComputer Vision and Pattern Recognition
The gist
Traffic jams are a growing problem in fast-changing cities, especially where roads are crowded and chaotic. The authors created a system that uses cameras to count vehicles despite overlaps and occlusions. Their approach turns camera images into maps showing vehicle density, then predicts future traffic and suggests better traffic light timings. They tested the vehicle counting part on real images from a busy Indian intersection, showing it can estimate traffic density with reasonable accuracy. The prediction and traffic light control parts still need real-world testing.
Open → 2610.38497v1

Multi agent learning improves ev charging station assignment under traffic uncertainty

Cooperative EV Charging Assignment Using Multi-Agent Reinforcement Learning: A Comparison of QMIX and Independent Q-Learning Under Urban Traffic Uncertainty

Abstract: The allocation of electric vehicles to public charging stations is increasingly constrained by the fact that several heterogeneous facilities often serve the same stream of requests while observing only partial local information. An accept/reject decision that is defensible for one station may nevertheless reduce the efficiency of the charging district as a whole. We formulate this operational problem as a Decentralized Partially Observable Markov Decision Process (Dec-POMDP) and examine whether Centralized Training with Decentralized Execution (CTDE), instantiated through QMIX, improves assignment performance relative to a cooperative Independent Q-Learning baseline (IQL). The evaluation uses a six-station urban network simulated with SUMO, five independent random seeds, 200 training episodes, and four traffic flux scenarios ranging from 100 to 250 vehicles/hour. Under this protocol, QMIX reaches a mean reward per evaluated decision of +0.776, versus +0.578 for IQL (delta = +0.198, seed-paired Cohens dz = 2.98, exact two-sided Wilcoxon p = 0.0625, the minimum attainable value at this sample size), and raises the acceptance rate from 72.4% to 94.0% (delta = +21.6 pp, dz = 5.24, Wilcoxon p = 0.0625). A five-variant ablation suggests this gain is primarily attributable to the CTDE architecture when holding cooperative reward terms fixed, though not formally isolated with a single seed. By contrast, cooperative reward shaping and experience sharing show no clear descriptive benefit and may slightly reduce performance. These findings support the view that, for multi-station charging assignment, coordination should be embedded in the learning architecture rather than added only through heuristic cooperation terms.

Tue 27 OctMultiagent SystemsMachine Learning
The gist
Deciding how electric vehicles should share public chargers is tricky because each charging station only sees some of the information. The authors studied if a special way of training multiple agents together helps stations make better decisions than if each learns separately. Using a city traffic simulator with different vehicle volumes, they found that training agents centrally and then letting them act separately led to higher acceptance of charging requests and better overall performance. This suggests that building coordination into the learning process matters more than just adding cooperative rules after the fact.
Open → 2610.38053v1

Adaptive calibration improves radio signal mapping with fewer measurements

Adaptive Calibration of RF Digital Twins via Budget-Constrained Radio Measurement Selection

Abstract: Radio-frequency (RF) digital twins (DTs) enable site-specific wireless propagation prediction by explicitly reconstructing the physical environment and modeling electromagnetic interactions. Although physics-based DTs reduce reliance on exhaustive measurements, systematic bias relative to real radio data arises from environmental uncertainty, material inaccuracies, and geometric abstractions. Calibration using field measurements is therefore necessary but costly under limited sampling budgets. This paper studies active RF DT calibration, in which measurement locations are selected sequentially and in an online manner to estimate a calibration offset under strict measurement budget constraints. A physics-based RF DT pipeline is developed, integrating semantic geospatial data from OpenStreetMap, 3D modeling in Blender, and deterministic ray tracing in Sionna to capture dominant building and foliage effects. This RF DT is calibrated using path-loss measurements collected with software-defined radios. Multiple online calibration strategies that select radio measurement locations are compared, including random sampling, rank-structured sampling with balanced coverage, quasi-Monte Carlo ranked selection, progressive structure-aware allocation, and fully adaptive uncertainty-driven sampling with estimator stabilization. Results demonstrate that adaptive and structure-aware schemes significantly improve both mean and worst-case calibration accuracy, particularly in the low-budget regime for radio measurements.

Mon 26 OctNetworking and Internet Architecture
The gist
Wireless signals change depending on where you measure them, and building easy-to-use models to predict these signals can be tricky because the environment is complex. The authors created a system that uses maps and 3D models to simulate how signals travel but found the simulation needs correction based on some real measurements. They figured out smart ways to choose where to take these measurements to fix the model efficiently without using too many resources. Their method especially helps when only a small number of measurements can be made.
Open → 2610.36406v1

Llm guided method cuts delays in 6g vehicle infrastructure systems

LLM-Guided Dual-Stage Latency Optimization for Closed-Loop 6G V2I Digital Twins

Abstract: Wireless Digital Twins are emerging as a key enabler for predictive network control in 6G Vehicle-to-Infrastructure (V2I) systems. However, closed-loop V2I DTs face two dual-stage communication bottlenecks arising from model parameter exchange for V2I DT reconstruction and multi-service V2I traffic delivery after each reconstruction. This paper proposes a closed-loop dual-stage End-to-End (E2E) latency optimization framework that integrates Compressed Personalized Federated Learning with a unified Large-Language-Model-guided Multi-Agent Reinforcement Learning approach. A CPFL-enabled decision-oriented V2I DT is first constructed by jointly predicting Path Loss (PL) and Quality-of-Experience (QoE)-oriented Key Performance Indicators (KPIs). Then the framework comprises two temporally coupled but latency-wise distinct optimization stages: Stage I minimizes the joint Uplink (UL) and Downlink (DL) E2E transmission latency of compressed shared backbone's parameters, and Stage II minimizes the service-aware E2E latency of multi-service V2I traffic. As the two stages share aligned agent roles and action space, we develop a unified Semantic-Aware Guidance-Enhanced Multi-Agent Proximal Policy Optimization framework, in which the LLM acts as a training-time semantic planner that provides high-level traffic splitting and power allocation guidance to reduce inefficient exploration, while the MARL agents execute feasible low-level actions. Simulation results demonstrate that the proposed framework enables efficient V2I DT reconstruction and significantly reduces both E2E latency for DT reconstruction and multi-service V2I traffic. The proposed SAGE-MAPPO supports cross-stage policy reuse by sequentially solving the two stages within each DT update cycle, which accelerates convergence relative to the existing baselines and facilitates timely DT updates in dynamic V2I scenarios.

Fri 16 OctNetworking and Internet Architecture
The gist
Wireless digital twins help predict and control 6G vehicle-to-infrastructure networks, but sending model data and service traffic can cause delays. The authors developed a two-step system that uses compressed federated learning and guidance from a large language model to reduce these delays. Their approach carefully plans and controls data sharing and traffic handling to speed up network updates and responses. Simulations show this method helps keep digital twins timely and efficient in dynamic connected vehicle scenarios.
Open → 2610.22848v1

Vision based model improves fire and smoke detection in outdoor areas

Attention-Based Spatio-Temporal Approach for Fire and Smoke Detection

Abstract: In various industries, smoke and fire are two of the most important threats in the workplace. One of the common methods for detecting smoke and fire is the use of infrared thermal and smoke sensors, which cannot be used in outdoor applications. Therefore, the use of vision-based methods seems necessary. The problem of smoke and fire detection is spatiotemporal and requires spatiotemporal solutions. This paper presents a method that uses spatial features along with temporal-based features to detect smoke and fire in the scene. It consists of three main parts; the task of each part is to reduce the error of the previous part so that the final model has a robust performance. This method also uses transformer modules to increase the accuracy of the model. The results of our model show the proper performance of the proposed approach in solving the problem of smoke and fire detection and can be used to increase workplace safety.

Wed 14 OctComputer Vision and Pattern Recognition
The gist
Detecting fire and smoke is tricky, especially outdoors where usual sensors don't work well. The authors created a new method that looks at both the shapes in images and how these shapes change over time to spot smoke and fire. They used parts of AI called transformers to make the model more accurate. Their approach aims to help keep workplaces safer by detecting dangers early.
Open → 2610.20141v1

Physics inspired network improves wireless signal loss prediction accuracy

HankelKANet: Lightweight Path Loss Estimation via Physics-Inspired Hankel Convolution

Abstract: Accurate path loss (PL) estimation is essential for wireless network planning and optimization. This paper proposes HankelKANet, a lightweight physics-inspired network for PL estimation that structurally embeds electromagnetic propagation characteristics into the neural architecture. Built on a single residual UNet backbone, the proposed model integrates urban layout features with a convolutional Kolmogorov-Arnold network(KAN). Specifically, the convolutional kernels are parameterized by the radial amplitudes of the zeroth- and first-order Hankel functions. This formulation acts as a physics-inspired nonlinear distance encoding, enabling the network to capture dominant attenuation characteristics directly without requiring additional physics-based loss terms. Experimental results demonstrate that HankelKANet achieves state-of-the-art accuracy and efficiency on both the simulated RadioMapSeer dataset and the real-world Huawei RSRPSet. Compared with the leading baseline, PEFNet, it reduces the parameter count by 78% and lowers the RMSE by 16.96% on the simulated data, while further exhibiting robust cross-dataset transferability and high data efficiency during target-domain adaptation. To interpret these empirical gains, we mathematically establish a parameter-conditional spatial gradient bound. Using discrete Dirichlet energy (DDE) and Jacobian sensitivity heatmaps, we empirically verify that this physical encoding intrinsically constrains the network's spatial derivatives. This structural smoothness effectively eliminates non-physical spatial ripples, preventing overfitting to simulator artifacts and serving as the fundamental driver for robust real world generalization.

Wed 14 OctMachine LearningNetworking and Internet Architecture
The gist
Wireless signals weaken as they travel through cities, and knowing exactly how much they weaken helps with planning networks. The authors designed HankelKANet, a lightweight AI that mimics physical laws of wave travel to predict this weakening more accurately and efficiently. This model uses special math functions to represent distances, which helps it avoid learning unrealistic patterns and improves its usefulness in the real world. Tests show HankelKANet beats previous methods, needing fewer resources while giving better predictions on both fake and real data.
Open → 2610.19524v1

Digital twins enable smarter closed loop control in wireless networks

Tutorial on Closed-Loop and Application-Aware Digital Twins for Future Wireless Networks

Abstract: To support emerging applications such as extended reality, autonomous systems, and large-scale cyber-physical infrastructures, next-generation wireless networks must evolve beyond traditional communication paradigms toward tightly integrated and closed-loop systems. Digital Twin Networks (DTNs) have recently emerged as a key enabler of this transformation by bridging the physical and digital worlds through real-time, bidirectional interactions and adaptive control mechanisms. In this tutorial, we provide a comprehensive overview of DTN design and optimization across multiple system dimensions. Specifically, we first introduce the fundamental design principles of DTN creation, including high-fidelity twin modeling, telemetry integration, and cyber-physical synchronization. We then present resource optimization and semantic-driven DTNs, where application-aware objectives and semantic abstractions guide efficient data processing, communication, and decision-making. Subsequently, we explore physical-layer optimization techniques for DTNs, emphasizing joint sensing, communication, and adaptive wireless resource control. Furthermore, we examine DTN-guided network management and feedback optimization, focusing on closed-loop learning, real-time control, and reinforcement-driven adaptation between digital twins and physical networks. Finally, we review ongoing standardization efforts and enabling tools that support the practical deployment of DTNs in real-world systems. This tutorial aims to provide a unified perspective on DTNs as an emerging paradigm for application-aware, AI-native, and closed-loop wireless systems, and to highlight key challenges and future research directions in this rapidly evolving field.

Sat 10 OctNetworking and Internet Architecture
The gist
Wireless networks will need to work better and faster to support future tech like virtual reality and driverless cars. The authors explain how digital twin networks create virtual copies of real networks that communicate back and forth in real time. These digital twins help optimize resources, manage wireless signals, and improve decision-making using AI and feedback loops. Their tutorial covers designing these twins, controlling wireless resources, and standards for using them in the future.
Open → 2610.15120v1

VbotNav-0 enables seamless indoor outdoor robot navigation with dual policies

VbotNav-0: Asynchronous Cloud-Edge Dual-Policy Navigation across Indoor and Outdoor Environments

Abstract: Continuous indoor-outdoor navigation connects destination-level reasoning with motion decisions that must respond to current observations. We present VbotNav-0, a cloud-edge dual-policy framework that links these processes through asynchronous policy-to-policy conditioning. The cloud policy produces subgoal guidance from visual history, navigation instructions, and route tokens; the edge policy combines available guidance with current RGB and LiDAR observations to generate local trajectories. This formulation allows the local perception-action loop to continue between cloud updates. A common route representation connects indoor paths and outdoor map routes, and the training recipe combines navigation demonstrations, corrective data, and grounded visual question answering (VQA). In the reported closed-loop evaluation of 258 episodes across 145 scenes, VbotNav-0 achieves 55.04% success rate and 55.03% success weighted by path length.

Sat 10 OctRobotics
The gist
Robots often struggle to navigate smoothly when moving between indoor and outdoor spaces because these environments require different kinds of decisions. The researchers created VbotNav-0, a system where a powerful cloud-based controller plans general routes, while a local device uses nearby camera and LiDAR data to make quick movement decisions. This way, the robot keeps moving safely and accurately even when updates from the cloud are delayed. They tested this approach in many diverse environments and showed it successfully completed routes more than half the time.
Open → 2610.14634v1

Unified score matching improves video anomaly detection and anticipation

A Unified Score Matching Paradigm for Video Anomaly Detection and Anticipation

Abstract: Video anomaly detection (VAD) is a fundamental and safety-critical task in computer vision. Recent generative approaches detect anomalies from a distributional perspective, but remain limited by local anomaly modes. Meanwhile, video anomaly anticipation (VAA), as a proactive extension beyond post-hoc detection, introduces additional challenges. In particular, the contrastive inference paradigm in VAD, which relies on ground-truth frames, is not applicable to VAA, hindering its development. To address these challenges, we propose a unified score-driven framework, termed Uni-DSM, based on denoising score matching (DSM), which models anomaly patterns through likelihood estimation and score functions over the learned data distribution. Within this unified framework, we adopt a shared noise-conditioned score transformer backbone with scene-dependent embeddings and motion-aware weighting for distribution-level modeling. Instead of introducing separate architectures, Uni-DSM unifies VAD and VAA through different inference and supervision paradigms built upon the same score-based formulation. For VAD, we instantiate an autoregressive denoising score matching (ADSM) mechanism, which progressively accumulates anomalous evidence via autoregressive denoising, enabling enhanced perception of local modes beyond visual cues. For VAA, we extend the same architecture by incorporating a lightweight auxiliary decoder and a novel self-distilled denoising score matching (SDSM) mechanism. By constructing supervision from output discrepancies instead of relying on unavailable future ground truth, our method achieves efficient training suitable or early anomaly anticipation. Extensive experiments on multiple benchmark datasets demonstrate state-of-the-art performance in both VAD and VAA while maintaining high efficiency, establishing a unified and scalable pipeline from anomaly detection to anticipation.

Thu 8 OctComputer Vision and Pattern Recognition
The gist
Detecting unusual events in videos is important for safety, but current methods struggle with some local details and predicting problems before they happen. The authors propose a single method called Uni-DSM that uses a score-matching approach to better understand normal video patterns and spot when something unusual occurs or is about to occur. Their method shares parts of the model for both finding and anticipating anomalies, overcoming previous limitations that needed future video frames to learn. Tests on several video datasets show their method performs better and faster than existing ones for both detecting and predicting unexpected events.
Open → 2610.11149v1

Citydeploy bench tests better ways to place city wireless transmitters

CityDeploy-Bench: Benchmarking Physics-Grounded Spatial Set Planning for Multi-Transmitter Network Deployment

Abstract: Automating city-scale wireless deployment remains challenging under complex urban propagation and network-wide interference. We introduce \textbf{CityDeploy-Bench}, a benchmark that reframes multi-transmitter deployment as \emph{physics-grounded spatial set planning} under a unified ray-tracing verifier. The benchmark separates utility representation from planning dynamics, enabling controlled comparison between direct scalar rewards, relational models, and higher-order interaction structures across diverse planners. Our experiments reveal a clear transition in planning behavior as physical coupling grows. Deployment quality becomes increasingly dependent on whether the learned utility captures collective transmitter interactions, whereas stronger search alone cannot compensate for missing relational structure. This establishes multi-transmitter deployment as a coordination problem over physically interacting sets rather than a collection of independent spatial decisions. We release CityDeploy-Data and the benchmark framework as a reproducible testbed for research linking decision learning with physically grounded wireless network design.

Thu 8 OctMachine Learning
The gist
Setting up many wireless transmitters across a city is tricky because signals can block or interfere with each other. The authors created CityDeploy-Bench, a tool to test how well different planning methods place transmitters considering real-world physics of signal paths. They found that planning works best when methods understand how transmitters interact together, not just individually. This shows that placing transmitters is a teamwork problem needing to think about their combined effects. The authors also provide data and tools to help others improve city wireless network setups.
Open → 2610.11065v1

Drone delivery fleets monitor city traffic while making food deliveries

Synergizing Drone Delivery Order Pooling and Road Network Monitoring through Monitoring-Task Orderization

Abstract: This paper investigates the real-time dispatch of a shared drone fleet for on-demand food delivery and urban road network monitoring. We consider a courier-drone collaborative setting in which couriers transport orders to launchpads and drones complete the final delivery leg to kiosks. Drones may consolidate multiple origin-destination orders within one flight and make monitoring-aware route adjustments to collect real-time traffic information subject to delivery-time constraints. This yields a joint decision problem coupling dynamic order-to-drone matching, multi-order pooling, routing, and time-varying monitoring under fleet-level competition and uncertainty. We propose monitoring-task orderization, which periodically converts road-network nodes with high congestion and stale information into virtual monitoring orders. Pooling these virtual tasks with food-delivery orders creates a unified heterogeneous task set and transforms the coupled matching-and-routing problem into an order-level decision process. Building on this abstraction, we formulate a decentralized graph-interdependent Multi-Agent Markov Decision Process and develop Graph Multi-Agent Q-Learning (Graph-MAQL), which captures localized inter-agent dependencies through bipartite match coordination graphs. Agent-task value estimates are then used as edge weights in a dynamic heterogeneous bipartite matching program for globally feasible execution. Experiments using real-world data reveal strong operational synergy between delivery and monitoring. Monitoring-task orderization improves monitoring performance by 25.1% with less than a 1% reduction in delivery performance, while Graph-MAQL improves the aggregate objective by up to 20.8%, reduces deadline violations by over 40%, and transfers zero-shot to higher demand intensity without retraining.

Sun 4 OctMultiagent SystemsMachine Learning
The gist
This paper studies how to use a shared fleet of drones for two jobs at once: delivering food orders and watching traffic in the city. The authors consider a system where human couriers bring orders to drone launch spots, and drones then finish deliveries while also collecting traffic data along their routes. By turning traffic-monitoring tasks into virtual delivery orders, they combine both jobs into one scheduling problem. They develop a learning method that helps drones decide which orders and monitoring tasks to do next, improving traffic monitoring without slowing down deliveries much.
Open → 2610.05270v1

Adaptive system improves online learning for changing data streams

AIGS: Adaptive Incremental Gating System for Online Representation Learning in Non-Stationary Data Streams

Abstract: Real-time data streams in Web of Things (WoT) and edge computing environments often evolve through latent regime changes. For online representation learning under strict computational constraints, the central problem is resolving the stability-plasticity dilemma: keeping useful historical knowledge while rapidly reacting to concept drift. Existing methods employ fixed update schedules or rolling windows. However, they suffer from parameter ossification during sudden shifts and waste computational resources when the stream remains stable. This paper proposes the Adaptive Incremental Gating System (AIGS), a lightweight closed-loop state-aware adaptation framework. AIGS introduces the Shock Ratio, an endogenous residual feedback mechanism that normalizes current reconstruction error against recent variation. This signal drives a Continuous Plasticity Controller that smoothly interpolates between learning plasticity and memory retention. By treating representation learning as a closed-loop control mechanism, AIGS avoids catastrophic forgetting and maintains a strictly linear $\mathcal{O}\left(k\cdot d\right)$ per-step complexity suitable for latency-sensitive edge devices. Experiments on real-world smart city dynamic streams-spanning traffic networks, meteorological systems, and industrial infrastructure-demonstrate distinct domain-dependent advantages. On Electricity Transformer Temperature datasets, AIGS achieves preventative early-warning lead times of 8.31 (ETTm1) and 9.88 (ETTm2) steps under gradual degradation. On Performance Measurement System traffic datasets, it shows significantly faster post-shift recovery after abrupt mutations. On the highly noisy Weather dataset, it improves anomaly recall while resisting stochastic noise overfitting. These findings establish AIGS as a practical, plug-and-play adapter for resource-constrained edge monitoring systems.

Fri 2 OctMachine Learning
The gist
Data from sensors often changes over time, making it hard for computers to keep learning without forgetting old information. The authors created a system that can quickly adjust how much it learns based on how much the data is changing. This helps it react faster to sudden shifts while still remembering important past patterns. Their system works well on different real-world data, like city traffic and weather, and runs efficiently on devices with limited computing power.
Open → 2610.02661v1

Audio visual descriptions improve understanding of urban scenes

AVSD-Scenes: A Dataset for Audio-Visual Description of Urban Scenes

Abstract: Natural language descriptions can provide rich semantic representations of audio-visual urban scenes, yet datasets that jointly describe both auditory and visual information remain limited. In this paper, we introduce AVSD-Scenes, a paired audio-visual scene description dataset for urban environments. The dataset contains 12,291 audio-visual scene descriptions generated from the TAU Urban Audio-Visual Scenes dataset. To construct the dataset, we first generate audio- and visual-based descriptions using Qwen2-Audio-7B and Qwen2.5-VL-7B, respectively. These modality-specific descriptions are then combined using large language models, namely Qwen3-14B, Mistral-Small-3.2-24B-Instruct-2506, and Gemma-3-27B-it, to produce multimodal descriptions that capture complementary information from both modalities. We benchmark AVSD-Scenes using semantic alignment, cross-modal retrieval, scene classification, LLM-as-a-judge evaluation, and human subjective assessment. Results show that multimodal descriptions improve semantic alignment and cross-modal retrieval performance compared with modality-specific descriptions while preserving strong scene-discriminative information. The generated descriptions achieve up to 94.5% accuracy in urban scene classification, while combining audio, visual, and description embeddings further improves accuracy to 95.4%. Furthermore, the descriptions remain highly scene-discriminative even when scene labels are removed from the prompting instructions, indicating that they capture semantic information derived from the audio-visual content rather than merely reflecting label information.

Thu 1 OctArtificial IntelligenceSound
The gist
Understanding a busy city scene can be easier if you use both sounds and pictures together. This paper presents a new collection of city scene descriptions that combine what you see and hear. The authors created these descriptions using specialized AI models for sound and images, then merged them to capture more details. Their tests show that using both sounds and visuals helps computers recognize and describe city scenes better than using either alone.
Open → 2610.01861v1

Closure-guided communication cuts data and boosts vehicle perception

SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception

Abstract: Vehicle-to-Everything (V2X) cooperative perception improves 3-D detection by sharing intermediate features, but dense remote features may repeat context that the ego agent can infer locally. Most communication-efficient designs optimize masks or codes empirically, leaving a more basic question open: which remote evidence is indispensable given the receiver's own observation? We introduce a closure-fidelity perspective on ego conditioned remote perception. Under a finite deductive abstraction and explicit conditions, its rate--distortion function decomposes over an irredundant core, and the exact zero-distortion rate becomes $P_A H(π_A)$. This analysis suggests a concrete design principle: transmit compact evidence and recover derivable context with bounded receiver-side inference. Guided by this principle, SemRD-V2X is an operational neural proxy that combines exact-budget BEV support selection, pointwise channel compression, and masked shared-weight reconstruction before standard fusion. Experiments on simulated V2XSet and real-world DAIR-V2X validate the resulting design. In a controlled five-run V2XSet comparison against a locally reproduced V2X-ViT-v1 baseline on one Tesla V100, SemRD-V2X reduces the analytical feature payload by $26.6\times$ while improving AP@0.5/AP@0.7 by 4.13/8.57 points, with 3.81\% additional mean compute latency. These results position closure fidelity as both an analytical lens and an actionable design principle for communication-efficient cooperative perception.

Mon 28 SeptArtificial IntelligenceInformation Theory
The gist
Vehicles working together to detect objects around them can send lots of data, often repeating what each car already knows. The authors study how to pick only the most crucial shared information that can't be guessed from a vehicle's own sensors. They design a system called SemRD-V2X that smartly compresses and shares only this essential data, letting each car fill in the rest by itself. Tests show this method cuts data sent by over 26 times while improving detection accuracy and only slightly slowing down computation.
Open → 2609.34353v1

TrafficFab autonomously manages city traffic using edge and cloud AI

TrafficFab: An Autonomic Edge-Cloud Testbed Fabric forAI-Driven Traffic Management

Abstract: Traffic management in emerging megacities requires real-time analytics over thousands of CCTV video streams under latency, bandwidth, compute and energy constraints. We present TrafficFab, an autonomic edge--cloud testbed for AI-driven traffic management, designed to validate a representative slice of a megacity deployment. TrafficFab combines RTSP stream emulation, heterogeneous edge inference using DNNs, cloud-based nowcasting and forecasting using Spatio-Temporal Graph Neural Network (ST-GNN), and continual model adaptation through foundation-model (FM)-assisted Federated Learning (FL). Its autonomic control enables fine-grained scale-out/in of edge inference through energy- and migration-aware scheduling, elastic scale-up/down of GNN forecasting on public clouds, and periodic adaptation of the DNN on edge accelerators and private cloud, without centralized video collection. We evaluate TrafficFab on a Bangalore-city inspired deployment, spanning Raspberry Pis, Jetson accelerators, GPU fogs, private cloud servers, and cloud VMs, sustaining real-time analytics for $\approx 400$ live camera streams (10% of Bangalore) and analytically characterize larger setups. The results demonstrate that TrafficFab offers a practical validation-scale platform for closed-loop traffic analytics, short-term operational decision support, and longer-horizon planning analyses in megacity scales.

Thu 24 SeptDistributed, Parallel, and Cluster Computing
The gist
Managing traffic in huge cities is challenging because it requires very fast analysis of many live video streams from traffic cameras. The authors created TrafficFab, a system that uses smart cameras and cloud computers to watch and predict traffic patterns without sending all video to one place. It automatically controls where and how AI models run on small devices near cameras and big cloud servers to save energy and computing power. They tested TrafficFab using a setup inspired by Bangalore city with hundreds of cameras to show it can work at city scale.
Open → 2609.29223v1

Domain adaptation method improves segmentation in bad weather

ICM: Intra-class Mixing for Domain Adaptation in Adverse Weather

Abstract: Unsupervised domain adaptation (UDA) for semantic segmentation remains challenging under adverse weather conditions because severe appearance changes enlarge the domain gap and degrade the reliability of pseudo labels in the target domain. To address this problem, we propose an Intra-Class Mixing Consistency (ICM) framework that enforces prediction consistency between an intra-class mixed image and its original counterpart. Unlike previous mixing-based consistency methods that combine regions across different images or domains and may introduce unrealistic semantic inconsistencies, ICM performs mixing within the same image and semantic class, preserving realistic semantic layout for consistency regularization. With ICM, we establish a new state-of-the-art performance for clear-to-adverse-weather unsupervised domain adaptation (UDA) in semantic segmentation. On the Cityscapes $\rightarrow$ ACDC benchmark, our method achieves 75.7\% mIoU, outperforming the previous state of the art by +1.9 pp, demonstrating its effectiveness in mitigating class confusion under challenging environmental conditions. The code is provided in the supplementary material.

Wed 23 SeptComputer Vision and Pattern Recognition
The gist
Semantic segmentation is a way for computers to understand images by identifying each part of a scene. This task gets harder in bad weather because the images look very different, making the computer confused. The authors created a new method called intra-class mixing consistency (ICM), which mixes parts of the same object within an image to help the computer learn better without confusing different things. Their method improved accuracy on a well-known test for understanding images taken in bad weather, showing it helps computers see more clearly when conditions are tough.
Open → 2609.27533v1

Trust management challenges and features in edge-enabled IoT security

Trust in Edge-Enabled IoT Security: Features, Challenges and Research Directions

Abstract: Providing autonomous intelligence, pervasive connectivity and usability to human life and industry has led to the emergence of the Internet of Things (IoT). To support time-sensitive and resource-constrained applications, IoT systems nowadays increasingly rely on edge computing. This brings computation and decision-making closer to end devices. In edge-enabled IoT architecture, latency and communication overhead are reduced, but interactions among a larger and more diverse set of devices, edge nodes, services, and data sources are introduced as well. In such environments, security and privacy mechanisms provide the foundation for protection, while trust management can assess the reliability of interacting entities and adapting secure decisions. In this paper, we systematically review the current state of trust management in edge-enabled IoT. To this end, we propose a comprehensive taxonomy that maps physical, network, and application architectural IoT layers against the consumer, commercial, industrial, and infrastructure IoT domains. We further investigate state-of-art research based on their trust design, how trust integrated into secure IoT operations, the attacks that effect trust management process. Based on these findings, we identify key gaps in current research and outline future directions for context-aware and adaptive trust management in edge-enabled IoT.

Mon 21 SeptCryptography and SecurityArtificial IntelligenceComputers and Society
The gist
As many everyday devices connect to the internet through the Internet of Things (IoT), keeping these devices secure becomes more complicated when edge computing is used to process data closer to the devices. This paper by the authors explains how trust management helps decide which devices and services can be trusted in these complex networks. They review existing research on how trust is designed and attacked in edge-enabled IoT and highlight gaps that need more work. Their findings guide future research toward making trust systems more adaptive and aware of different contexts in IoT setups.
Open → 2609.24669v1

Vast toolchain boosts testing of cooperative autonomous driving systems

VAST: V2X/Dynamic Map-Aware Autonomous Driving Systems Validation Toolchain

Abstract: Cooperative autonomous driving in the IoT-to-Edge-to-Cloud continuum requires system-level validation across vehicles, infrastructure sensors, edge-side Dynamic Map services, and in-vehicle autonomous-driving stacks. This paper presents VAST, a V2X/Dynamic Map-aware validation toolchain that connects Scenic, Scenario Simulator v2, AWSIM, Autoware, and SIM-LDM. VAST does not introduce a new search algorithm; instead, it addresses interoperability challenges, including Lanelet2-to-Scenic mapping, ROS 2-based co-simulation through SS2, Dynamic Map object injection into Autoware, and collection of TTC, PET, collision, timeout, and performance measurements. In occluded-intersection scenarios, Lanelet2-compatible constrained sampling increases the edge-case discovery rate from 40.0% to 80.0% and reduces the average time per discovered edge case from 259.7 s to 110.4 s. Under the same generated scenario distribution, Dynamic Map availability reduces the collision rate from 78.0% to 40.0% and increases non-collision outcomes from 22.0% to 60.0%, with statistically significant TTC/PET shifts. A throughput study with 1-16 NPCs shows that sampling remains below 0.1 s, whereas AWSIM/Autoware execution and restart overhead dominate runtime. These results position VAST as a practical validation infrastructure for cooperative autonomous-driving CPSs.

Thu 17 SeptRobotics
The gist
Testing self-driving cars that communicate with other vehicles and traffic systems is complicated because many different software parts and sensors have to work together. The authors made VAST, a toolchain that connects several simulators and autonomous-driving software, to check how these systems behave together in different scenarios. Their tool helps find tricky situations where accidents could happen faster and more often, and shows that having detailed dynamic maps can make driving safer. VAST also runs efficiently even with multiple simulated cars around, making it a useful platform for testing real-world cooperative driving technology.
Open → 2609.19681v1

Multi-view pedestrian tracking improves with fewer cameras

GRACE: Geometry- and Ray-Aware Camera-Efficient Multi-View Pedestrian Tracking

Abstract: Reducing the number of cameras reduces the deployment cost but removes views that correct BEV responses stretched away from true pedestrian positions by projection and short score drops that can split tracks} in Bird's-Eye View (BEV) tracking. We introduce GRACE, a camera-efficient multi-view tracker with three components. Volumetric-Guided Fusion combines homography-based BEV features with features lifted through 3D space. Ray Conditioning exposes each camera's viewing direction to the fusion network. Its tracking component, BEV Track Recovery (BTR), uses low-confidence detections only to continue existing tracks. The same detections cannot start new tracks. With two WildTrack cameras, GRACE improves MOTA from 83.54 for TrackTacular, our baseline, to 91.07.

Tue 15 SeptComputer Vision and Pattern Recognition
The gist
Tracking people using cameras is easier with many cameras, but using fewer cameras cuts costs. The authors introduce GRACE, a method that helps track pedestrians accurately even with just two cameras. It combines different types of 3D and bird’s-eye view features and conditions on each camera’s direction to improve tracking. Their method reduces errors and keeps track of people better than earlier approaches.
Open → 2609.16872v1

Multi sensor fusion improves vehicle network beam prediction accuracy

Robust Beam Prediction for V2X Networks with Multi-Modal Sensing

Abstract: Integrated sensing and communication (ISAC) provides a promising foundation for beam prediction in future vehicle-to-everything (V2X) networks. However, existing sensing-assisted beamforming methods still rely heavily on radio-frequency sensing, which may become unreliable in complex vehicular environments. Meanwhile, the growing availability of heterogeneous sensors, such as cameras and LiDAR, offers new opportunities to improve beam prediction through richer environmental perception. Motivated by this, this paper proposes a multi-modal beam prediction framework for V2X networks. Specifically, we develop BeamTransFuser, a hierarchical Transformer-based architecture that progressively fuses camera, LiDAR, radar, and GPS observations for robust beam prediction. In addition, to handle possible missing modalities in practical deployment, we introduce a generative module that reconstructs missing modality features from the available observations. Experimental results on a real-world multi-modal V2X dataset show that the proposed framework consistently outperforms representative baselines, while the generative module further improves robustness under incomplete sensing conditions.

Wed 9 SeptMachine Learning
The gist
In vehicle communication networks, it's important to focus signals accurately to maintain good connections. The authors noticed current methods rely mostly on radio signals, which can fail in tricky environments. They designed a system called BeamTransFuser that combines data from cameras, LiDAR, radar, and GPS to better guess where to direct signals. Their system also copes when some sensors aren’t available by filling in missing data from the others. Tests on real vehicle data showed this approach predicts signal directions more reliably than before.
Open → 2609.10200v1

LiDAR detection improves urban vehicle and pedestrian tracking

Solution for UCF UrbanTwin V2X-Real Track: Sim-to-Real Urban LiDAR 3D Object Detection

Abstract: Bridging the simulation-to-reality gap in roadside LiDAR requires addressing several coupled discrepancies, including scene geometry, sampling density, return patterns, and pedestrian scale. This report presents a multi-source collaborative training and class-aware fusion framework for Sim2Real 3D detection. The method organizes digital-twin scans, diffusion-redrawn scans, density-stabilized scans, and pedestrian morphology-aligned samples into a unified training pool with complementary roles. Within a common DSVT detection formulation, source-specialized expert branches preserve those roles while optimizing for the same detection objective. At inference, a predefined class-aware fusion pathway integrates geometry-stable and calibration-aware branches for vehicles, sampling-complementary branches for trucks, and morphology-consistent evidence for pedestrians. A label-free point-cloud center blend then refines geometric localization. On the UrbanTwin V2X-Real hidden test set, the unified system achieves a combined score of 0.7421, with 3D mAP@0.5 of 0.4518 and a realism score of 0.8871. The results indicate that a stable, interpretable collaboration among data sources is more valuable than unconstrained aggregation of model outputs.

Mon 7 SeptComputer Vision and Pattern Recognition
The gist
Detecting objects like vehicles and pedestrians using 3D LiDAR scanners in cities is tricky because real-world data differs from simulated data. The authors created a system that combines different types of simulated scans to better match real-world conditions. They use specialized parts of the system for different object types and then fuse the results in a smart way to improve accuracy. Their approach works well on a challenging urban dataset, showing that carefully mixing data sources is better than just combining many model outputs randomly.
Open → 2609.07608v1