Papers

This paper presents an adaptive observer design for semilinear hyperbolic rolling contact ODE-PDE systems with uncertain friction characteristics parameterized by a matrix of unknown coefficients appearing in the nonlinear (and possibly non-smooth) PDE source terms. Under appropriate assumptions of forward completeness and boundary sensing, an adaptive observer is synthesized to simultaneously estimate the lumped and distributed states, as well as the uncertain friction parameters, using only boundary measurements. The observer combines a finite-dimensional parameter estimator with an infinite-dimensional description of the state error dynamics, and achieves exponential convergence under persistent excitation. The effectiveness of the proposed design is demonstrated in simulation by considering a relevant example borrowed from road vehicle dynamics.

This paper proposes a novel paradigm centered on Artificial Intelligence (AI)-empowered propagation channel prediction to address the limitations of traditional channel modeling. We present a comprehensive framework that deeply integrates heterogeneous environmental data and physical propagation knowledge into AI models for site-specific channel prediction, which referred to as channel inference. By leveraging AI to infer site-specific wireless channel states, the proposed paradigm enables accurate prediction of channel characteristics at both link and area levels, capturing spatio-temporal evolution of radio propagation. Some novel strategies to realize the paradigm are introduced and discussed, including AI-native and AI-hybrid inference approaches. This paper also investigates how to enhance model generalization through transfer learning and improve interpretability via explainable AI techniques. Our approach demonstrates significant practical efficacy, achieving an average path loss prediction root mean square error (RMSE) of $\sim$ 4 dB and reducing training time by 60%-75%. This new modeling paradigm provides a foundational pathway toward high-fidelity, generalizable, and physically consistent propagation channel prediction for future communication networks.

Purpose: The model allocates the system components orders to the suppliers to minimize the parts price and the system construction delay penalties and maximize the system availability during its use. It considers the quantity-based discount and variation of delivery lead time by ordering similar components. The model also reflects the prerequisite relationships between construction activities and calculates the delay penalty resulting from parts delivery lead time. Design/methodology/approach: This research presents a model for selecting suppliers of components of an industrial series-parallel multi-state system. A nonlinear binary mathematical program uses the Markov process results to select system components. It minimizes the total system construction phase costs, including the components' price, and the system construction delay penalty, and the system exploitation phase costs, including the system shutdown and working at half capacity. Findings: The model allocates the optimal orders for a typical industrial system's components, composing four elements. The proposed approach combines the nonlinear binary program and the Markov process results to optimize the system life cycle parameters, including the system construction cost and operational availability. Originality/value: Using the Markov chain results in binary nonlinear mathematical programming, this study attempts to strike the right balance between the construction phase's objectives and an industrial unit's operation phase.

Accurate precoding in massive multiple-input multiple-output (MIMO) frequency-division duplexing (FDD) systems relies on efficient channel state information (CSI) acquisition. End-to-end learning frameworks improve performance by jointly optimizing this process, but they lack scalability and fail to generalize across different system configurations, such as varying numbers of antennas and users. To overcome this limitation, we introduce WiFo-E, a wireless foundation model designed for scalable end-to-end precoding. WiFo-E employs multi-task pretraining on a diverse set of configurations to learn transferable representations of underlying wireless principles. Central to the model is a sparse Mixture-of-Experts (MoE) Transformer architecture, which mitigates task interference and enhances training efficiency by activating specialized parameter subsets adaptively. Extensive simulations demonstrate that WiFo-E outperforms conventional per-configuration training and shows strong generalization to unseen system configurations, providing a flexible and efficient foundation for adaptive massive MIMO precoding.

The emerging convergence of next-generation wireless networks and agentic artificial intelligence (AI) is inspiring a new vision: embodied intelligent network entities utilize environmental sensing to refine their physical-layer (PHY) actions. Despite a growing body of preliminary work, prevailing small and task-specific AI models require extensive manual design of data pre-processing, network architecture, and fine-tuning, leaving them tightly coupled to particular PHY actions, system configurations, and deployment scenarios. To address this, we propose a paradigm shift with WiFo-M$^2$, a foundation model that enables environment sensing to be easily integrated into PHY actions, delivering universal performance gains. To extract generalizable out-of-band (OOB) channel-aware features from environment sensing, we introduce ContraSoM, a contrastive pre-training strategy. Once pre-trained, WiFo-M$^2$ infers future OOB channel-aware features from historical sensory data and strengthens feature robustness via modality-specific data augmentation. Experiments show that WiFo-M$^2$ improves the performance of a comprehensive suite of fundamental PHY actions, demonstrating strong generalization to unseen scenarios.

Grant-free (GF) access is essential for massive connectivity but faces collision risks due to uncoordinated transmissions. While user-side sensing can mitigate these collisions by enabling autonomous transmission decisions, conventional methods become ineffective in overloaded scenarios where active streams exceed receive antennas. To address this problem, we propose a differential stream sensing framework that reframes the problem from estimating the total stream count to isolating newly activated streams via covariance differencing. We analyze the covariance deviation induced by channel variations to establish a theoretical bound based on channel correlation for determining the sensing window size. To mitigate residual interference from finite sampling, a deep learning (DL) classifier is integrated. Simulations across both independent and identically distributed flat Rayleigh fading and standardized channel environments demonstrate that the proposed method consistently outperforms non-DL baselines and remains robust in overloaded scenarios.

We propose a probabilistic semantic filtering framework in which parameters of a dynamical system are inferred and associated with a closed set of semantic classes in a map. We extend existing methods to a multi-parameter setting using a posterior that tightly couples semantics with the parameter likelihoods, and propose a filter to compute this posterior sequentially, subject to dynamics in the map's state. Using Bayesian moment matching, we show that the computational complexity of measurement updates scales linearly in the dimension of the parameter space. Finally, we demonstrate limitations of applying existing methods to a problem from the driving domain, and show that the proposed framework better captures time-varying parameter-to-semantic associations.

This letter proposes a block sparse Bayesian learning (BSBL) algorithm of non-circular (NC) signals for direction-of-arrival (DOA) estimation, which is suitable for arbitrary unknown NC phases. The block sparse NC signal representation model is constructed through a permutation strategy, capturing the available intra-block structure information to enhance recovery performance. After that, we create the sparse probability model and derive the cost function under BSBL framework. Finally, the fast marginal likelihood maximum (FMLM) algorithm is introduced, enabling the rapid implementation of signal recovery by the addition and removal of basis functions. Simulation results demonstrate the effectiveness and the superior performance of our proposed method.

Vision Transformers (ViTs) have gained rapid adoption in computational pathology for their ability to model long-range dependencies through self-attention, addressing the limitations of convolutional neural networks that excel at local pattern capture but struggle with global contextual reasoning. Recent pathology-specific foundation models have further advanced performance by leveraging large-scale pretraining. However, standard ViTs remain inherently non-equivariant to transformations such as rotations and reflections, which are ubiquitous variations in histopathology imaging. To address this limitation, we propose Equi-ViT, which integrates an equivariant convolution kernel into the patch embedding stage of a ViT architecture, imparting built-in rotational equivariance to learned representations. Equi-ViT achieves superior rotation-consistent patch embeddings and stable classification performance across image orientations. Our results on a public colorectal cancer dataset demonstrate that incorporating equivariant patch embedding enhances data efficiency and robustness, suggesting that equivariant transformers could potentially serve as more generalizable backbones for the application of ViT in histopathology, such as digital pathology foundation models.

The move to next-generation wireless communications with extremely large-scale antenna arrays (ELAAs) brings the communications into the radiative near-field (RNF) region, where distance-aware focusing is feasible. However, high-frequency RNF links are highly vulnerable to blockage in indoor environments dominated by half-space obstacles (walls, corners) that create knife-edge shadows. Conventional near-field focused beams offer high gain in line-of-sight (LoS) scenarios but suffer from severe energy truncation and effective-rank collapse in shadowed regions, often necessitating the deployment of auxiliary hardware such as Reconfigurable Intelligent Surfaces (RIS) to restore connectivity. We propose a beamforming strategy that exploits the auto-bending property of Airy beams to mitigate half-space blockage without additional hardware. The Airy beam is designed to “ride” the diffraction edge, accelerating its main lobe into the shadow to restore connectivity. Our contributions are threefold: (i) a Green's function-based RNF multi-user channel model that analytically reveals singular-value collapse behind knife-edge obstacles; (ii) an Airy analog beamforming scheme that optimizes the bending trajectory to recover the effective channel rank; and (iii) an Airy null-steering method that aligns oscillatory nulls with bright-region users to suppress interference in mixed shadow/bright scenarios. Simulations show that the proposed edge-riding Airy strategy achieves a Signal-to-Noise Ratio (SNR) improvement of over 20 dB and restores full-rank connectivity in shadowed links compared to conventional RNF focusing, virtually eliminating outage in geometric shadows and increasing multi-user spectral efficiency by approximately 35% under typical indoor ELAA configurations. These results demonstrate robust RNF multi-user access in half-space blockage scenarios without relying on RIS.

Split Federated Learning (SFL) enables collaborative training between resource-constrained edge devices and a compute-rich server. Communication overhead is a central issue in SFL and can be mitigated with auxiliary networks. Yet, the fundamental client-side computation challenge remains, as back-propagation requires substantial memory and computation costs, severely limiting the scale of models that edge devices can support. To enable more resource-efficient client computation and reduce the client-server communication, we propose HERON-SFL, a novel hybrid optimization framework that integrates zeroth-order (ZO) optimization for local client training while retaining first-order (FO) optimization on the server. With the assistance of auxiliary networks, ZO updates enable clients to approximate local gradients using perturbed forward-only evaluations per step, eliminating memory-intensive activation caching and avoiding explicit gradient computation in the traditional training process. Leveraging the low effective rank assumption, we theoretically prove that HERON-SFL's convergence rate is independent of model dimensionality, addressing a key scalability concern common to ZO algorithms. Empirically, on ResNet training and language model (LM) fine-tuning tasks, HERON-SFL matches benchmark accuracy while reducing client peak memory by up to 64% and client-side compute cost by up to 33% per step, substantially expanding the range of models that can be trained or adapted on resource-limited devices.

The Access Traffic Steering, Switching, and Splitting (ATSSS) defined in the latest 3GPP Release 19 enables traffic flow over the multiple access paths to achieve the lower-latency End-to-end (E2E) delivery for 6G time-sensitive services. However, the existing E2E multi-path operation often falls short of more stringent QoS requirements for 6G time-sensitive services. This work proposes a Reconfigurable Intelligent Surfaces (RIS)-aided E2E multi-path uplink (UL) transmission architecture that explicitly accounts for both radio link latency and N3 backhaul latency, via the coupled designs of the UL traffic-splitting ratio, transmit power, receive combining, and RIS phase shift under practical constraints to achieve the minimum average E2E latency. We develop an alternating optimization framework that updates the above target parameters to be optimized. The simulations were conducted to compare the effectiveness of the proposed E2E optimization framework that lowers the average E2E latency up to 43% for a single user and 32% for the whole system compared with baselines in our prior work [1].

Medical imaging datasets often suffer from class imbalance and limited availability of pathology-rich cases, which constrains the performance of machine learning models for segmentation, classification, and vision-language tasks. To address this challenge, we propose POWDR, a pathology-preserving outpainting framework for 3D MRI based on a conditioned wavelet diffusion model. Unlike conventional augmentation or unconditional synthesis, POWDR retains real pathological regions while generating anatomically plausible surrounding tissue, enabling diversity without fabricating lesions. Our approach leverages wavelet-domain conditioning to enhance high-frequency detail and mitigate blurring common in latent diffusion models. We introduce a random connected mask training strategy to overcome conditioning-induced collapse and improve diversity outside the lesion. POWDR is evaluated on brain MRI using BraTS datasets and extended to knee MRI to demonstrate tissue-agnostic applicability. Quantitative metrics (FID, SSIM, LPIPS) confirm image realism, while diversity analysis shows significant improvement with random-mask training (cosine similarity reduced from 0.9947 to 0.9580; KL divergence increased from 0.00026 to 0.01494). Clinically relevant assessments reveal gains in tumor segmentation performance using nnU-Net, with Dice scores improving from 0.6992 to 0.7137 when adding 50 synthetic cases. Tissue volume analysis indicates no significant differences for CSF and GM compared to real images. These findings highlight POWDR as a practical solution for addressing data scarcity and class imbalance in medical imaging. The method is extensible to multiple anatomies and offers a controllable framework for generating diverse, pathology-preserving synthetic data to support robust model development.

We present the Universal Latent Homeomorphic Manifold (ULHM), a framework that unifies semantic representations (e.g., human descriptions, diagnostic labels) and observation-driven machine representations (e.g., pixel intensities, sensor readings) into a single latent structure. Despite originating from fundamentally different pathways, both modalities capture the same underlying reality. We establish homeomorphism, a continuous bijection preserving topological structure, as the mathematical criterion for determining when latent manifolds induced by different semantic-observation pairs can be rigorously unified. This criterion provides theoretical guarantees for three critical applications: (1) semantic-guided sparse recovery from incomplete observations, (2) cross-domain transfer learning with verified structural compatibility, and (3) zero-shot compositional learning via valid transfer from semantic to observation space. Our framework learns continuous manifold-to-manifold transformations through conditional variational inference, avoiding brittle point-to-point mappings. We develop practical verification algorithms, including trust, continuity, and Wasserstein distance metrics, that empirically validate homeomorphic structure from finite samples. Experiments demonstrate: (1) sparse image recovery from 5% of CelebA pixels and MNIST digit reconstruction at multiple sparsity levels, (2) cross-domain classifier transfer achieving 86.73% accuracy from MNIST to Fashion-MNIST without retraining, and (3) zero-shot classification on unseen classes achieving 78.76% on CIFAR-10. Critically, the homeomorphism criterion determines when different semantic-observation pairs share compatible latent structure, enabling principled unification into universal representations and providing a mathematical foundation for decomposing general foundation models into domain-specific components.

We aim to investigate the impact of image and signal properties on visual attention mechanisms during a signal detection task in digital images. The application of insight yielded from this work spans many areas of digital imaging where signal or pattern recognition is involved in complex heterogenous background. We used simulated tomographic breast images as the platform to investigate this question. While radiologists are highly effective at analyzing medical images to detect and diagnose diseases, misdiagnosis still occurs. We selected digital breast tomosynthesis (DBT) images as a sample medical images with different breast densities and structures using digital breast phantoms (Bakic and XCAT). Two types of lesions (with distinct spatial frequency properties) were randomly inserted in the phantoms during projections to generate abnormal cases. Six human observers participated in observer study designed for a locating and detection of an 3-mm sphere lesion and 6-mm spicule lesion in reconstructed in-plane DBT slices. We collected eye-gaze data to estimate gaze metrics and to examine differences in visual attention mechanisms. We found that detection performance in complex visual environments is strongly constrained by later perceptual stages, with decision failures accounting for the largest proportion of errors. Signal detectability is jointly influenced by both target morphology and background complexity, revealing a critical interaction between local signal features and global anatomical noise. Increased fixation duration on spiculated lesions suggests that visual attention is differentially engaged depending on background and signal spatial frequency dependencies.

Ultra-High Field MRI (UHF-MRI) is increasingly used in large-scale neuroimaging studies, yet automatic brain segmentation and cortical parcellation remain challenging due to signal inhomogeneities, heterogeneous contrasts and resolutions, and the limited availability of tools optimized for UHF data. Standard software packages such as FastSurferVINN and SynthSeg+ often yield suboptimal results when applied directly to UHF images, thereby restricting region-based quantitative analyses. To address this need, we introduce GOUHFI 2.0, an updated implementation of GOUHFI that incorporates increased training data variability and additional functionalities, including cortical parcellation and volumetry. GOUHFI 2.0 preserves the contrast- and resolution-agnostic design of the original toolbox while introducing two independently trained 3D U-Net segmentation tasks. The first performs whole-brain segmentation into 35 labels across contrasts, resolutions, field strengths and populations, using a domain-randomization strategy and a training dataset of 238 subjects. Using the same training data, the second network performs cortical parcellation into 62 labels following the Desikan-Killiany-Tourville (DKT) protocol. Across multiple datasets, GOUHFI 2.0 demonstrated improved segmentation accuracy relative to the original toolbox, particularly in heterogeneous cohorts, and produced reliable cortical parcellations. In addition, the integrated volumetry pipeline yielded results consistent with standard volumetric workflows. Overall, GOUHFI 2.0 provides a comprehensive solution for brain segmentation, parcellation and volumetry across field strengths, and constitutes the first deep-learning toolbox enabling robust cortical parcellation at UHF-MRI.

This work introduces ABE-VVS, a framework that performs attribute based selective coordinate encryption for point cloud based volumetric video streaming, enabling lightweight yet effective digital rights management (DRM). Rather than encrypting entire point cloud frames, our approach encrypts only selected subsets of coordinates ($X, Y, Z$, or combinations), lowering computational overhead and latency while still producing strong visual distortion that prevents meaningful unauthorized viewing. Our experiments show that encrypting only the $X$ coordinates achieves effective obfuscation while reducing encryption and decryption times by up to 50% and 80%, respectively, compared to full-frame encryption. To our knowledge, this is the first work to provide a novel end-to-end evaluation of a DRM-enabled secure point cloud streaming system. We deployed a point cloud video streaming setup on the CloudLab testbed and evaluated three HTTP-based Attribute-Based Encryption (ABE) granularities - ABE-XYZ (encrypting all $X,Y,Z$ coordinates), ABE-XY, and ABE-X against conventional HTTPS/TLS secure streaming as well as an HTTP-only baseline without any security. Our streaming evaluation demonstrates that ABE-based schemes reduce server-side CPU load by up to 80% and cache CPU load by up to 63%, comparable to HTTP-only, while maintaining similar cache hit rates. Moreover, ABE-XYZ and ABE-XY exhibit lower client-side rebuffering than HTTPS, and ABE-X achieves zero rebuffering comparable to HTTP-only. Although ABE-VVS increases client-side CPU usage, the overhead is not large enough to affect streaming quality and is offset by its broader benefits, including simplified key revocation, elimination of per-client encryption, and reduced server and cache load.

The cell-free networking paradigm constitutes a revolutionary architecture for future generations of wireless networks, which has been recently considered in synergy with Reconfigurable Intelligent Surfaces (RISs), a promising physical-layer technology for signal propagation programmability. In this paper, we focus on wideband cell-free multi-RIS-empowered Multiple-Input Single-Output (MISO) systems and present a decentralized cooperative active and passive beamforming scheme, aiming to provide an efficient alternative towards the cooperation overhead of available centralized schemes depending on central processing unit. Considering imperfect channel information availability and realistic frequency selectivity behavior of each RIS's element response, we devise a distributed optimization approach based on consensus updates for the RISs' phase configurations. Our simulation results showcase that the proposed distributed design is superior to centralized schemes that are based on various Lorentzian-type wideband modeling approaches for the RISs.

Full-duplex communication substantially enhances spectral efficiency by enabling simultaneous transmission and reception on the same time-frequency resources. However, its practical deployment remains hindered by strong residual self-interference and inter-user interference, which severely degrade system performance. This work investigates a full-duplex MISO network that leverages movable-antenna base stations (MA-BS) and movable-element reconfigurable intelligent surfaces (ME-RIS) to overcome these limitations in next-generation 6G systems. Unlike conventional fixed-geometry architectures, the proposed framework jointly optimizes antenna and RIS element positions, together with RIS phase shifts, to strengthen desired links while suppressing interference. Our design objective is to maximize the system sum rate through the joint optimization of transmit and receive beamforming vectors, uplink transmit powers, RIS phase shifts, and the spatial locations of both the BS antennas and RIS elements. To solve this challenging nonconvex problem, an alternating optimization algorithm is developed, employing semidefinite relaxation for beamforming design and successive convex approximation for position optimization. Simulation results demonstrate that the proposed ME-RIS-assisted architecture with movable BS antennas offers substantial gains over conventional fixed-position full-duplex networks. These findings highlight the potential of integrating movable antennas with movable RIS elements as a key enabler for high-performance full-duplex operation in future 6G wireless systems.

Medical image fusion integrates complementary information from multiple imaging modalities to improve clinical interpretation. However, existing deep learningbased methods, including recent spatial-frequency frameworks such as AdaFuse and ASFE-Fusion, often suffer from a fundamental trade-off between global statistical similaritymeasured by correlation coefficient (CC) and mutual information (MI)and local structural fidelity. This paper proposes W-DUALMINE, a reliability-weighted dual-expert fusion framework designed to explicitly resolve this trade-off through architectural constraints and a theoretically grounded loss design. The proposed method introduces dense reliability maps for adaptive modality weighting, a dual-expert fusion strategy combining a global-context spatial expert and a wavelet-domain frequency expert, and a soft gradient-based arbitration mechanism. Furthermore, we employ a residual-to-average fusion paradigm that guarantees the preservation of global correlation while enhancing local details. Extensive experiments on CT-MRI, PET-MRI, and SPECT-MRI datasets demonstrate that W-DUALMINE consistently outperforms AdaFuse and ASFE-Fusion in CC and MI metrics while

Machine learning approaches for fringe projection profilometry (FPP) are hindered by the lack of large, diverse datasets and standardized benchmarking protocols. This paper introduces the first open-source, photorealistic synthetic dataset for FPP, generated using NVIDIA Isaac Sim, comprising 15,600 fringe images and 300 depth reconstructions across 50 objects. We apply this dataset to single-shot FPP, where models predict 3D depth maps directly from individual fringe images without temporal phase shifting. Through systematic ablation studies, we identify optimal learning configurations for long-range (1.5-2.1 m) depth prediction. We compare three depth normalization strategies and show that individual normalization, which decouples object shape from absolute scale, yields a 9.1x improvement in object reconstruction accuracy over raw depth. We further show that removing background fringe patterns severely degrades performance across all normalizations, demonstrating that background fringes provide essential spatial phase reference rather than noise. We evaluate six loss functions and identify Hybrid L1 loss as optimal. Using the best configuration, we benchmark four architectures and find UNet achieves the strongest performance, though errors remain far above the sub-millimeter accuracy of classical FPP. The small performance gap between architectures indicates that the dominant limitation is information deficit rather than model design: single fringe images lack sufficient information for accurate depth recovery without explicit phase cues. This work provides a standardized benchmark and evidence motivating hybrid approaches combining phase-based FPP with learned refinement. The dataset is available at https://huggingface.co/datasets/aharoon/fpp-ml-bench and code at https://github.com/AnushLak/fpp-ml-bench.

The received in-phase and quadrature (I/Q) baseband signals inherently encode physical-layer and channel characteristics of wireless links. Learning robust and transferable representations directly from such raw signals, however, remains challenging due to heterogeneous communication systems, diverse propagation environments, and limited labeled data. To address this, we present LWM-Spectro, a transformer-based foundation model pretrained on large-scale I/Q data represented as time-frequency spectrograms. The model leverages self-supervised masked modeling, contrastive learning, and a mixture-of-experts (MoE) architecture to learn general-purpose wireless representations. These representations transfer effectively to downstream tasks such as modulation classification and joint SNR/mobility recognition, even with minimal supervision. Across tasks, LWM-Spectro consistently outperforms state-of-the-art deep learning baselines in both few-shot and data-rich regimes, providing a unified foundation for wireless learning.

Generative recommendation systems have achieved significant advances by leveraging semantic IDs to represent items. However, existing approaches that tokenize each modality independently face two critical limitations: (1) redundancy across modalities that reduces efficiency, and (2) failure to capture inter-modal interactions that limits item representation. We introduce FusID, a modality-fused semantic ID framework that addresses these limitations through three key components: (i) multimodal fusion that learns unified representations by jointly encoding information across modalities, (ii) representation learning that brings frequently co-occurring item embeddings closer while maintaining distinctiveness and preventing feature redundancy, and (iii) product quantization that converts the fused continuous embeddings into multiple discrete tokens to mitigate ID conflict. Evaluated on a multimodal next-song recommendation (i.e., playlist continuation) benchmark, FusID achieves zero ID conflicts, ensuring that each token sequence maps to exactly one song, mitigates codebook underutilization, and outperforms baselines in terms of MRR and Recall@k (k = 1, 5, 10, 20).

Chain-of-Thought (CoT) reasoning has proven effective in enhancing large language models by encouraging step-by-step intermediate reasoning, and recent advances have extended this paradigm to Multimodal Large Language Models (MLLMs). In the medical domain, where diagnostic decisions depend on nuanced visual cues and sequential reasoning, CoT aligns naturally with clinical thinking processes. However, current benchmarks for medical image understanding generally focus on the final answer while ignoring the reasoning path. Such opaque reasoning processes lack reliable bases for judgment, making it difficult to assist doctors in diagnosis. To address this gap, we introduce a new M3CoTBench benchmark specifically designed to evaluate the correctness, efficiency, impact, and consistency of CoT reasoning in medical image understanding. M3CoTBench features 1) a diverse, multi-level difficulty dataset covering 24 examination types, 2) 13 varying-difficulty tasks, 3) a suite of CoT-specific evaluation metrics (correctness, efficiency, impact, and consistency) tailored to clinical reasoning, and 4) a performance analysis of multiple MLLMs. M3CoTBench systematically evaluates CoT reasoning across diverse medical imaging tasks, revealing current limitations of MLLMs in generating reliable and clinically interpretable reasoning, and aims to foster the development of transparent, trustworthy, and diagnostically accurate AI systems for healthcare. Project page at https://juntaojianggavin.github.io/projects/M3CoTBench/.

The rapid growth of urban populations and the increasing need for sustainable transportation solutions have prompted a shift towards electric buses in public transit systems. However, the effective management of mixed fleets consisting of both electric and diesel buses poses significant operational challenges. One major challenge is coping with dynamic electricity pricing, where charging costs vary throughout the day. Transit agencies must optimize charging assignments in response to such dynamism while accounting for secondary considerations such as seating constraints. This paper presents a comprehensive mixed-integer linear programming (MILP) model to address these challenges by jointly optimizing charging schedules and trip assignments for mixed (electric and diesel bus) fleets while considering factors such as dynamic electricity pricing, vehicle capacity, and route constraints. We address the potential computational intractability of the MILP formulation, which can arise even with relatively small fleets, by employing a hierarchical approach tailored to the fleet composition. By using real-world data from the city of Chattanooga, Tennessee, USA, we show that our approach can result in significant savings in the operating costs of the mixed transit fleets.

We propose a novel piecewise smooth image model with piecewise constant local parameters that are automatically adapted to each image. Technically, the model is formulated in terms of factor graphs with NUP (normal with unknown parameters) priors, and the pertinent computations amount to iterations of conjugate-gradient steps and Gaussian message passing. The proposed model and algorithms are demonstrated with applications to denoising and contrast enhancement.

Many areas in science and engineering now have access to technologies that enable the rapid collection of overwhelming data volumes. While these datasets are vital for understanding phenomena from physical to biological and social systems, the sheer magnitude of the data makes even simple storage, transmission, and basic processing highly challenging. To enable efficient and accurate execution of these data processing tasks, we require new dimensionality reduction tools that 1) do not need expensive, time-consuming training, and 2) preserve the underlying geometry of the data that has the information required to understand the measured system. Specifically, the geometry to be preserved is that induced by the fact that in many applications, streaming high-dimensional data evolves on a low-dimensional attractor manifold. Importantly, we may not know the exact structure of this manifold a priori. To solve these challenges, we present randomized filtering (RF), which leverages a specific instantiation of randomized dimensionality reduction to provably preserve non-linear manifold structure in the embedded space while remaining data-independent and computationally efficient. In this work we build on the rich theoretical promise of randomized dimensionality reduction to develop RF as a real, practical approach. We introduce novel methods, analysis, and experimental verification to illuminate the practicality of RF in diverse scientific applications, including several simulated and real-data examples that showcase the tangible benefits of RF.

Thoracic aortic dissection and aneurysms are the most lethal diseases of the aorta. The major hindrance to treatment lies in the accurate analysis of the medical images. More particularly, aortic segmentation of the 3D image is often tedious and difficult. Deep-learning-based segmentation models are an ideal solution, but their inability to deliver usable outputs in difficult cases and their computational cost cause their clinical adoption to stay limited. This study presents an innovative approach for efficient aortic segmentation using targeted region of interest (ROI) detection. In contrast to classical detection models, we propose a simple and efficient detection model that can be widely applied to detect a single ROI. Our detection model is trained as a multi-task model, using an encoder-decoder architecture for segmentation and a fully connected network attached to the bottleneck for detection. We compare the performance of a one-step segmentation model applied to a complete image, nnU-Net and our cascade model composed of a detection and a segmentation step. We achieve a mean Dice similarity coefficient of 0.944 with over 0.9 for all cases using a third of the computing power. This simple solution achieves state-of-the-art performance while being compact and robust, making it an ideal solution for clinical applications.

Tabla Stroke Transcription (TST) is central to the analysis of rhythmic structure in Hindustani classical music, yet remains challenging due to complex rhythmic organization and the scarcity of strongly annotated data. Existing approaches largely rely on fully supervised learning with onset-level annotations, which are costly and impractical at scale. This work addresses TST in a weakly supervised setting, using only symbolic stroke sequences without temporal alignment. We propose a framework that combines a CTC-based acoustic model with sequence-level rhythmic rescoring. The acoustic model produces a decoding lattice, which is refined using a \textbf{$T\bar{a}la$}-Independent Static–Dynamic Rhythmic Model (TI-SDRM) that integrates long-term rhythmic structure with short-term adaptive dynamics through an adaptive interpolation mechanism. We curate a new real-world tabla solo dataset and a complementary synthetic dataset, establishing the first benchmark for weakly supervised TST in Hindustani classical music. Experiments demonstrate consistent and substantial reductions in stroke error rate over acoustic-only decoding, confirming the importance of explicit rhythmic structure for accurate transcription.

Particle-based communication using diffusion and advection has emerged as an alternative signaling paradigm recently. While most existing studies assume constant flow conditions, real macro-scale environments such as atmospheric winds exhibit time-varying behavior. In this work, airborne particle communication under time-varying advection is modeled as a linear time-varying (LTV) channel, and a closed-form, time-dependent channel impulse response is derived using the method of moving frames. Based on this formulation, the channel is characterized through its power delay profile, leading to the definition of channel dispersion time as a physically meaningful measure of channel memory and a guideline for symbol duration selection. System-level simulations under directed, time-varying wind conditions show that waveform design is critical for performance, enabling multi-symbol modulation using a single particle type when dispersion is sufficiently controlled. To quantify waveform distortion and guide the design of orthogonal signaling waveforms, the Orthogonality Loss Ratio (OLR) is introduced as a structural metric. The results demonstrate that time-varying diffusion-advection channels can be systematically modeled and engineered using communication-theoretic tools, providing a realistic foundation for particle-based communication in complex flow environments.