Papers

The Schrodinger Bridge and Bass (SBB) formulation, which jointly controls drift and volatility, is an established extension of the classical Schrodinger Bridge (SB). Building on this framework, we introduce LightSBB-M, an algorithm that computes the optimal SBB transport plan in only a few iterations. The method exploits a dual representation of the SBB objective to obtain analytic expressions for the optimal drift and volatility, and it incorporates a tunable parameter beta greater than zero that interpolates between pure drift (the Schrodinger Bridge) and pure volatility (Bass martingale transport). We show that LightSBB-M achieves the lowest 2-Wasserstein distance on synthetic datasets against state-of-the-art SB and diffusion baselines with up to 32 percent improvement. We also illustrate the generative capability of the framework on an unpaired image-to-image translation task (adult to child faces in FFHQ). These findings demonstrate that LightSBB-M provides a scalable, high-fidelity SBB solver that outperforms existing SB and diffusion baselines across both synthetic and real-world generative tasks. The code is available at https://github.com/alexouadi/LightSBB-M.

We propose a method for estimating the magnitude distribution of an acoustic field from spatially sparse magnitude measurements. Such a method is useful when phase measurements are unreliable or inaccessible. Physics-informed neural networks (PINNs) have shown promise for sound field estimation by incorporating constraints derived from governing partial differential equations (PDEs) into neural networks. However, they do not extend to settings where phase measurements are unavailable, as the loss function based on the governing PDE relies on phase information. To remedy this, we propose a phase-retrieval-based PINN for magnitude field estimation. By representing the magnitude and phase distributions with separate networks, the PDE loss can be computed based on the reconstructed complex amplitude. We demonstrate the effectiveness of our phase-retrieval-based PINN through experimental evaluation.

Neural video compression (NVC) has demonstrated superior compression efficiency, yet effective rate control remains a significant challenge due to complex temporal dependencies. Existing rate control schemes typically leverage frame content to capture distortion interactions, overlooking inter-frame rate dependencies arising from shifts in per-frame coding parameters. This often leads to suboptimal bitrate allocation and cascading parameter decisions. To address this, we propose a reinforcement-learning (RL)-based rate control framework that formulates the task as a frame-by-frame sequential decision process. At each frame, an RL agent observes a spatiotemporal state and selects coding parameters to optimize a long-term reward that reflects rate-distortion (R-D) performance and bitrate adherence. Unlike prior methods, our approach jointly determines bitrate allocation and coding parameters in a single step, independent of group of pictures (GOP) structure. Extensive experiments across diverse NVC architectures show that our method reduces the average relative bitrate error to 1.20% and achieves up to 13.45% bitrate savings at typical GOP sizes, outperforming existing approaches. In addition, our framework demonstrates improved robustness to content variation and bandwidth fluctuations with lower coding overhead, making it highly suitable for practical deployment.

Stabilizing a dynamical system is a fundamental problem that serves as a cornerstone for many complex tasks in the field of control systems. The problem becomes challenging when the system model is unknown. Among the Reinforcement Learning (RL) algorithms that have been successfully applied to solve problems pertaining to unknown linear dynamical systems, the policy gradient (PG) method stands out due to its ease of implementation and can solve the problem in a model-free manner. However, most of the existing works on PG methods for unknown linear dynamical systems assume full-state feedback. In this paper, we take a step towards model-free learning for partially observed linear dynamical systems with output feedback and focus on the fundamental stabilization problem of the system. We propose an algorithmic framework that stretches the boundary of PG methods to the problem without global convergence guarantees. We show that by leveraging zeroth-order PG update based on system trajectories and its convergence to stationary points, the proposed algorithms return a stabilizing output feedback policy for discrete-time linear dynamical systems. We also explicitly characterize the sample complexity of our algorithm and verify the effectiveness of the algorithm using numerical examples.

In this work, we revisit outlier hypothesis testing and propose exponentially consistent, low-complexity fixed-length tests that achieve a better tradeoff between detection performance and computational complexity than existing exhaustive-search methods. In this setting, the goal is to identify outlying sequences from a set of observed sequences, where most sequences are i.i.d. from a nominal distribution and outliers are i.i.d. from a different anomalous distribution. While prior work has primarily focused on discrete-valued sequences, we extend the results of Bu et al. (TSP 2019) to continuous-valued sequences and develop a distribution-free test based on the MMD metric. Our framework handles both known and unknown numbers of outliers. In the unknown-count case, we bound the detection performance and characterize the tradeoff among the exponential decay rates of three types of error probabilities. Finally, we quantify the performance penalty incurred when the number of outliers is unknown.

Purpose: To simulate effective transverse relaxation ($T_2^*$) as a part of MR simulation. $T_2^*$ consists of reversible ($T_2^{\prime}$) and irreversible ($T_2$) components. Whereas simulations of $T_2$ are easy, $T_2^{\prime}$ is not easily simulated if only magnetizations of individual isochromats are simulated. Theory and Methods: Efficient methods for simulating $T_2^{\prime}$ were proposed. To approximate the Lorentzian function of $T_2^{\prime}$ realistically, conventional simulators require 100+ isochromats. This approximation can be avoided by utilizing a linear phase model for simulating an entire Lorentzian function directly. To represent the linear phase model, the partial derivatives of the magnetizations with respect to the frequency axis were also simulated. To accelerate the simulations with these partial derivatives, the proposed methods introduced two techniques: analytic solutions, and combined transitions. For understanding the fundamental mechanism of the proposed method, a simple one-isochromat simulation was performed. For evaluating realistic cases, several pulse sequences were simulated using two phantoms with and without $T_2^{\prime}$ simulations. Results: The one-isochromat simulation demonstrated that $T_2^{\prime}$ simulations were possible. In the realistic cases, $T_2^{\prime}$ was recovered as expected without using 100+ isochromats for each point. The computational times with $T_2^{\prime}$ simulations were only 2.0 to 2.7 times longer than those without $T_2^{\prime}$ simulations. When the above-mentioned two techniques were utilized, the analytic solutions accelerated 19 times, and the combined transitions accelerated up to 17 times. Conclusion: Both theory and results showed that the proposed methods simulated $T_2^{\prime}$ efficiently by utilizing a linear model with a Lorentzian function, analytic solutions, and combined transitions.

Bluetooth Low Energy (BLE) is widely used in wearable devices but has rarely been considered for ingestible electronics because of severe tissue attenuation at 2.4 GHz. In this work, we systematically reevaluate BLE for ingestible applications by benchmarking it against representative Sub-GHz schemes in terms of tissue attenuation, power consumption, throughput, reliability, bidirectional communication, latency, and system integration. Water-phantom and in vivo measurements confirm stronger attenuation at 2.4 GHz, by approximately 40 dB; however, front-end-module (FEM)-assisted BLE compensates for this loss while maintaining practical received signal levels and energy efficiency. For ingestible sensing applications requiring below 100 kbps, BLE with a 20 dBm transmission power achieves lower power consumption than the evaluated Sub-GHz alternatives. BLE also provides protocol-level reliability through retransmission and in-order packet delivery, whereas Sub-GHz links can suffer packet loss under high attenuation. In addition, BLE supports bidirectional communication with limited throughput degradation, while the evaluated Sub-GHz implementation shows a substantial reduction. End-to-end measurements further demonstrate lower BLE latency because of its native compatibility with modern computing infrastructure. Finally, BLE offers advantages in antenna form factor, ecosystem support, and system integration. These results demonstrate that properly configured BLE is a practical communication solution for next-generation ingestible electronics.

Wireless capsule endoscopy (WCE) is fundamentally constrained by limited wireless bandwidth, resulting in low imaging resolution and frame rate, which can cause motion blur and missed lesions. Although adaptive frame-rate schemes have been explored to accommodate transient gastrointestinal (GI) motility, these approaches typically require sacrificing image resolution. The use of higher-frequency communication bands is further limited by increased tissue attenuation. To address these challenges, we propose a hybrid Bluetooth Low Energy (BLE) and WiFi communication architecture that combines the low-power operation of BLE with the high data throughput of WiFi. We systematically evaluate the performance of BLE and WiFi under tissue-mimicking conditions by measuring throughput, received signal strength indicator (RSSI), and power consumption. The results demonstrate that amplified BLE with an adaptive transmission power control strategy provides a stable frame rate at low power consumption, while 2.4 GHz WiFi operating in station mode is the most suitable high-throughput communication configuration for WCE. Compared with WiFi, BLE reduces power consumption by approximately ten times, whereas WiFi achieves up to ten times higher throughput. To reconcile these complementary trade-offs, we further introduce a hybrid system with a frame-boundary-synchronized switching mechanism to ensure lossless data transmission during BLE and WiFi transitions. Experimental results show that the switching latency from BLE to WiFi is approximately 92.66 ms, which is longer than the WiFi-to-BLE switching latency of 15.49 ms when transmitting 10 kB image payloads. Overall, the proposed hybrid BLE and WiFi system enables robust, lossless, and energy-efficient mode switching, supports adaptive imaging, and advances the development of next-generation autonomous WCE platforms.

Speaker-attributed automatic speech recognition (ASR) in multi-speaker environments remains a major challenge. While some approaches achieve strong performance when fine-tuned on specific domains, few systems generalize well across out-of-domain datasets. Our prior work, Diarization-Conditioned Whisper (DiCoW), leverages speaker diarization outputs as conditioning information and, with minimal fine-tuning, demonstrated strong multilingual and multi-domain performance. In this paper, we address a key limitation of DiCoW: ambiguity in Silence-Target-Non-target-Overlap (STNO) masks, where two or more fully overlapping speakers may have nearly identical conditioning despite differing transcriptions. We introduce SE-DiCoW (Self-Enrolled Diarization-Conditioned Whisper), which uses diarization output to locate an enrollment segment anywhere in the conversation where the target speaker is most active. This enrollment segment is used as fixed conditioning via cross-attention at each encoder layer. We further refine DiCoW with improved data segmentation, model initialization, and augmentation. Together, these advances yield substantial gains: SE-DiCoW reduces macro-averaged tcpWER by 52.4% relative to the original DiCoW on the EMMA MT-ASR benchmark.

Vision-aided wireless sensing is emerging as a cornerstone of 6G mobile computing. While data-driven approaches have advanced rapidly, establishing a precise geometric correspondence between ego-centric visual data and radio propagation remains a challenge. Existing paradigms typically either associate 2D topology maps and auxiliary information with radio maps, or provide 3D perspective views limited by sparse radio data. This spatial representation flattens the complex vertical interactions such as occlusion and diffraction that govern signal behavior in urban environments, rendering the task of cross-view signal inference mathematically ill-posed. To resolve this geometric ambiguity, we introduce SynthRM, a scalable synthetic data platform. SynthRM implements a Visible-Aligned-Surface simulation strategy: rather than probing a global volumetric grid, it performs ray-tracing directly onto the geometry exposed to the sensor. This approach ensures pixel-level consistency between visual semantics and electromagnetic response, transforming the learning objective into a physically well-posed problem. We demonstrate the platform's capabilities by presenting a diverse, city-scale dataset derived from procedurally generated environments. By combining efficient procedural synthesis with high-fidelity electromagnetic modeling, SynthRM provides a transparent, accessible foundation for developing next-generation mobile systems for environment-aware sensing and communication.

Fluorescence microscopy (FM) imaging is a fundamental technique for observing live cell division, one of the most essential processes in the cycle of life and death. Observing 3D live cells requires scanning through the cell volume while minimizing lethal phototoxicity. That limits acquisition time and results in sparsely sampled volumes with anisotropic resolution and high noise. Existing image restoration methods, primarily based on inverse problem modeling, assume known and stable degradation processes and struggle under such conditions, especially in the absence of high-quality reference volumes. In this paper, from a new perspective, we propose a novel tensor completion framework tailored to the nature of FM imaging, which inherently involves nonlinear signal degradation and incomplete observations. Specifically, FM imaging with equidistant Z-axis sampling is essentially a tensor completion task under a uniformly random sampling condition. On one hand, we derive the theoretical lower bound for exact cell tensor completion, validating the feasibility of accurately recovering 3D cell tensor. On the other hand, we reformulate the tensor completion problem as a mathematically equivalent score-based generative model. By incorporating structural consistency priors, the generative trajectory is effectively guided toward denoised and geometrically coherent reconstructions. Our method demonstrates state-of-the-art performance on SR-CACO-2 and three real in vivo cellular datasets, showing substantial improvements in both signal-to-noise ratio and structural fidelity.

Most universal sound extraction algorithms focus on isolating a target sound event from single-channel audio mixtures. However, the real world is three-dimensional, and binaural audio, which mimics human hearing, can capture richer spatial information, including sound source location. This spatial context is crucial for understanding and modeling complex auditory scenes, as it inherently informs sound detection and extraction. In this work, we propose a language-driven universal sound extraction network that isolates text-described sound events from binaural mixtures by effectively leveraging the spatial cues present in binaural signals. Additionally, we jointly predict the direction of arrival (DoA) of the target sound using spatial features from the extraction network. This dual-task approach exploits complementary location information to improve extraction performance while enabling accurate DoA estimation. Experimental results on the in-the-wild AudioCaps dataset show that our proposed LuSeeL model significantly outperforms single-channel and uni-task baselines.

This paper summarizes the opportunities of in-network collective operations (INC) for accelerated collective operations in AI workloads. We provide sufficient detail to make this important field accessible to non-experts in AI or networking, fostering a connection between these communities. Consider two types of INC: Edge-INC, where the system is implemented at the node level, and Core-INC, where the system is embedded within network switches. We outline the potential performance benefits as well as six key obstacles in the context of both Edge-INC and Core-INC that may hinder their adoption. Finally, we present a set of predictions for the future development and application of INC.

This study treats transmission scheduling for remote state estimation over unreliable channels with a hidden mode. A local Kalman estimator selects scheduling actions, such as power allocation and resource usage, and communicates with a remote estimator based on acknowledgement feedback, balancing estimation performance and communication cost. The resulting problem is naturally formulated as a partially observable Markov decision process (POMDP). In settings with observable channel modes, it is well known that monotonicity of the value function can be established via investigating order-preserving property of transition kernels. In contrast, under partial observability, the transition kernels generally lack this property, which prevents the direct application of standard monotonicity arguments. To overcome this difficulty, we introduce a novel technique, referred to as state-space folding, which induces transformed transition kernels recovering order preservation on the folded space. This transformation enables a rigorous monotonicity analysis in the partially observable setting. As a representative implication, we focus on an associated optimal stopping formulation and show that the resulting optimal scheduling policy admits a threshold structure.

Most audio-visual speaker extraction methods rely on synchronized lip recording to isolate the speech of a target speaker from a multi-talker mixture. However, in natural human communication, co-speech gestures are also temporally aligned with speech, often emphasizing specific words or syllables. These gestures provide complementary visual cues that can be especially valuable when facial or lip regions are occluded or distant. In this work, we move beyond lip-centric approaches and propose SeLG, a model that integrates both lip and upper-body gesture information for robust speaker extraction. SeLG features a cross-attention-based fusion mechanism that enables each visual modality to query and selectively attend to relevant speech features in the mixture. To improve the alignment of gesture representations with speech dynamics, SeLG also employs a contrastive InfoNCE loss that encourages gesture embeddings to align more closely with corresponding lip embeddings, which are more strongly correlated with speech. Experimental results on the YGD dataset, containing TED talks, demonstrate that the proposed contrastive learning strategy significantly improves gesture-based speaker extraction, and that our proposed SeLG model, by effectively fusing lip and gesture cues with an attention mechanism and InfoNCE loss, achieves superior performance compared to baselines, across both complete and partial (i.e., missing-modality) conditions.

Color quantization represents an image using a fraction of its original number of colors while only minimally losing its visual quality. The $k$-means algorithm is commonly used in this context, but has mostly been applied in the machine-based RGB colorspace composed of the three primary colors. However, some recent studies have indicated its improved performance in human perception-based colorspaces. We investigated the performance of $k$-means color quantization at four quantization levels in the RGB, CIE-XYZ, and CIE-LUV/CIE-HCL colorspaces, on 148 varied digital images spanning a wide range of scenes, subjects and settings. The Visual Information Fidelity (VIF) measure numerically assessed the quality of the quantized images, and showed that in about half of the cases, $k$-means color quantization is best in the RGB space, while at other times, and especially for higher quantization levels ($k$), the CIE-XYZ colorspace is where it usually does better. There are also some cases, especially at lower $k$, where the best performance is obtained in the CIE-LUV colorspace. Further analysis of the performances in terms of the distributions of the hue, chromaticity and luminance in an image presents a nuanced perspective and characterization of the images for which each colorspace is better for $k$-means color quantization.

Passive daytime radiative cooling (PDRC) provides an energy-free approach to suppress surface temperatures by reflecting solar irradiation while emitting thermal radiation through the mid-infrared atmospheric window. Despite rapid progress in optical performance, most PDRC systems remain limited by rigid, fragile, or planar substrates, restricting their use on flexible, curved, or wearable surfaces. Here, we report a biocompatible and structurally robust PDRC system integrated onto a commercial rapid-curing fiberglass cast, a conformal substrate widely used in orthopedic and industrial applications. The cooling architecture adopts a bilayer polymer design consisting of a polyvinyl alcohol (PVA) adhesion layer and a polymethyl methacrylate (PMMA) protective layer, both embedded with calcium pyrophosphate (CPP) ceramic particles derived from processed animal bone waste. The bio-derived CPP simultaneously enables broadband solar scattering and high mid-infrared emittance, while offering sustainability and biocompatibility advantages. The resulting composite exhibits over 90% solar reflectance and achieves up to 15 C sub-ambient cooling under direct outdoor sunlight.

Universal speech enhancement aims at handling inputs with various speech distortions and recording conditions. In this work, we propose a novel hybrid architecture that synergizes the signal fidelity of discriminative modeling with the reconstruction capabilities of generative modeling. Our system utilizes the discriminative TF-GridNet model with the Sampling-Frequency-Independent strategy to handle variable sampling rates universally. In parallel, an autoregressive model combined with spectral mapping modeling generates detail-rich speech while effectively suppressing generative artifacts. Finally, a fusion network learns adaptive weights of the two outputs under the optimization of signal-level losses and the comprehensive Speech Quality Assessment (SQA) loss. Our proposed system is evaluated in the ICASSP 2026 URGENT Challenge (Track 1) and ranks the third place.

As load varies continuously over time, it is essential to provide continuous-time price signals that accurately reflect supply-demand balance. However, conventional discrete-time economic dispatch fails to capture the intra-temporal variations in load and generation. Its dual solution–the marginal price–may distort economic signals, leading to inefficient market incentives. To analyze these issues, this paper develops a continuous-time dispatch model and derives its dual formulation for price analysis. The continuous-time dispatch produces dual variables that can be interpreted as price signals. Piecewise time-indexed generation and price trajectories are then constructed through a parametric programming approach. The resulting price, represented by the Lagrange multiplier of the system-wide power balance constraint, evolves piecewise along the continuous-time load profile. Each segment corresponds to a critical region characterized by a set of active constraints. Furthermore, we discuss the impact of unit-specific ramping constraints on price implications. Results indicate that continuous-time generation and price trajectories provide deeper insights into the price distortions and inefficient incentives inherent in discrete-time formulations. The proposed methodology is validated on an illustrative 5-bus system and a modified IEEE RTS-2019.

This paper presents a coded-aperture reflector for indoor mmWave coverage enhancement in obstructed or blocked LoS settings. We model the reflecting aperture using an equivalent array-factor formulation, where each passive reflecting cell contributes a reradiated field with phase set by the incident and departure directions. Building on this model, we develop two fabrication-friendly passive synthesis methods: (i) binary (1-bit) spatial coding that enables deterministic non-specular beam formation and multibeam patterns by selecting cell participation on a dense {\lambda}/2 lattice via an ON/OFF metallization mask, and (ii) diffraction-order (periodic) steering that exploits aperture periodicity to place selected diffraction orders at prescribed angles. We analytically characterize the proposed cosine-threshold quantization rule, including its asymptotic activation ratio and a distribution-free lower bound on non-specular gain relative to ideal continuous-phase control. To validate the proposed designs, we fabricate and metallize low-cost prototypes in-house using a copper-backed 3D-printed "inkwell" substrate with stencil-guided conductive ink deposition. 60 GHz over-the-air measurements show non-specular power enhancements on the order of +14-20 dB relative to passive, non-engineered (all-ON) reflector baselines. Results also demonstrate that fully passive, binary-coded apertures can deliver beam control with rapid in-lab manufacturability and offer a practical alternative to power-consuming reconfigurable surfaces for static indoor mmWave links.

This paper deals with the leader-follower attitude synchronization problem for a group of heterogeneous rigid body systems on $SO(3)$ under an undirected, connected, and acyclic graph communication topology. The proposed distributed control strategy, endowed with almost global asymptotic stability guarantees, allows the synchronization of the rigid body systems to a constant desired orientation known only to a single rigid body. Some simulation results are also provided to validate the theoretical developments and illustrate the performance of the proposed control strategy.

Designing optimal non-terrestrial network (NTN) constellations is essential for maximizing throughput and ensuring fair resource distribution. This paper presents FTA-NTN (Fairness and Throughput Assurance in Non-Terrestrial Networks), a multi-objective optimization framework that jointly maximizes throughput and fairness under realistic system constraints. The framework integrates multi-layer Walker Delta constellations, a parametric mobility model for user distributions across Canadian land regions, adaptive K-Means clustering for beamforming and user association, and Bayesian optimization for parameter tuning. Simulation results with 500 users show that FTA-NTN achieves over 9.88 Gbps of aggregate throughput with an average fairness of 0.42, corresponding to an optimal configuration of 9 planes with 15 satellites per plane in LEO and 7 planes with 3 satellites per plane in MEO. These values align with 3GPP NTN evaluation scenarios and representative system assumptions, confirming their relevance for realistic deployments. Overall, FTA-NTN demonstrates that throughput and fairness can be jointly optimized under practical constraints, advancing beyond throughput-centric designs in the literature and offering a scalable methodology for next-generation NTN deployments that supports efficient and equitable global connectivity.

Using actual flight data from a 50-cm-class microsatellite whose mission and operations have already been completed, this study re-evaluates satellite attitude determination performance and the error characteristics of onboard attitude sensors. While conventional approaches rely on batch estimation or Kalman filtering based on predefined physical models and white-noise assumptions, this research introduces a machine-learning-based approach to extract and correct structural and nonlinear error patterns embedded in real observational data. In this study, high-quality attitude determination results obtained from star sensors and a fiber optical gyro (FOG) are treated as ground truth, and machine learning is applied to coarse attitude sensor data consisting of Sun sensors and magnetic field sensors. A one-dimensional convolutional neural network (Conv1D) is employed to regressively predict attitude from short sequences of time-series sensor measurements. The model is trained and evaluated using five sets of on-orbit observation logs, with four passes used for training and one independent pass used for testing. The results show that, while conventional coarse attitude determination using the TRIAD method yields attitude errors on the order of 7 deg RMS, the proposed machine-learning approach achieves RMS errors of approximately 0.7 deg for the training data and 2-3 deg for the test data, depending on the sensor combination.

Play behaviour serves as a positive welfare indicator in dairy calves, yet the influence of space allowance under commercial conditions remains poorly characterized, particularly at intermediate-to-high allowances (6-20 m2 per calf). This study investigated the relationship between space allowance and play behaviour in 60 group-housed dairy calves across 14 commercial farms in the Netherlands (space range: 2.66-17.98 m2 per calf), and developed an automated computer vision pipeline for scalable monitoring. Video observations were analyzed using a detailed ethogram, with play expressed as percentage of observation period (%OP). Statistical analysis employed linear mixed models with farm as a random effect. A computer vision pipeline was trained on manual annotations from 108 hours on 6 farms and validated on held-out test data. The computer vision classifier achieved 97.6% accuracy with 99.4% recall for active play detection. Calves spent on average 1.0% of OP playing reflecting around 10 minutes per 17-hour period. The space-play relationship was non-linear, with highest play levels at 8-10 m2 per calf (1.6% OP) and lowest at 6-8 m2 and 12-14 m2 (<0.6% OP). Space remained significant after controlling for age, health, and group size. In summary, these findings suggest that 8-10 m2 per calf represents a practical target balancing welfare benefits with economic feasibility, and demonstrate that automated monitoring can scale small annotation projects to continuous welfare assessment systems.

Presentation attack detection (PAD) subsystems are an important part of effective and user-friendly remote identity validation (RIV) systems. However, ensuring robust performance across diverse environmental and procedural conditions remains a critical challenge. This paper investigates the impact of low-light conditions and automated image acquisition on the robustness of commercial PAD systems using a scenario test of RIV. Our results show that PAD systems experience a significant decline in performance when utilized in low-light or auto-capture scenarios, with a model-predicted increase in error rates by a factor of about four under low-light conditions and a doubling of those odds under auto-capture workflows. Specifically, only one of the tested systems was robust to these perturbations, maintaining a maximum bona fide presentation classification error rate below 3% across all scenarios. Our findings emphasize the importance of testing across diverse environments to ensure robust and reliable PAD performance in real-world applications.

To meet the requirements for managing unauthorized UAVs in the low-altitude economy, a multi-modal UAV trajectory prediction method based on the fusion of LiDAR and millimeter-wave radar information is proposed. A deep fusion network for multi-modal UAV trajectory prediction, termed the Multi-Modal Deep Fusion Framework, is designed. The overall architecture consists of two modality-specific feature extraction networks and a bidirectional cross-attention fusion module, aiming to fully exploit the complementary information of LiDAR and radar point clouds in spatial geometric structure and dynamic reflection characteristics. In the feature extraction stage, the model employs independent but structurally identical feature encoders for LiDAR and radar. After feature extraction, the model enters the Bidirectional Cross-Attention Mechanism stage to achieve information complementarity and semantic alignment between the two modalities. To verify the effectiveness of the proposed model, the MMAUD dataset used in the CVPR 2024 UG2+ UAV Tracking and Pose-Estimation Challenge is adopted as the training and testing dataset. Experimental results show that the proposed multi-modal fusion model significantly improves trajectory prediction accuracy, achieving a 40% improvement compared to the baseline model. In addition, ablation experiments are conducted to demonstrate the effectiveness of different loss functions and post-processing strategies in improving model performance. The proposed model can effectively utilize multi-modal data and provides an efficient solution for unauthorized UAV trajectory prediction in the low-altitude economy.

Real-time voice agents face a dilemma: end-to-end models often lack deep reasoning, while cascaded pipelines incur high latency by executing ASR, LLM reasoning, and TTS strictly in sequence, unlike human conversation where listeners often start thinking before the speaker finishes. Since cascaded architectures remain the dominant choice for complex tasks, existing cascaded streaming strategies attempt to reduce this latency via mechanical segmentation (e.g., fixed chunks, VAD-based splitting) or speculative generation, but they frequently either break semantic units or waste computation on predictions that must be rolled back. To address these challenges, we propose LTS-VoiceAgent, a Listen-Think-Speak framework that explicitly separates when to think from how to reason incrementally. It features a Dynamic Semantic Trigger to detect meaningful prefixes, and a Dual-Role Stream Orchestrator that coordinates a background Thinker (for state maintenance) and a foreground Speaker (for speculative solving). This parallel design enables "thinking while speaking" without blocking responses. We also introduce a Pause-and-Repair benchmark containing natural disfluencies to stress-test streaming robustness. Experiments across VERA, Spoken-MQA, BigBenchAudio, and our benchmark show that LTS-VoiceAgent achieves a stronger accuracy-latency-efficiency trade-off than serial cascaded baselines and existing streaming strategies.

Symbolic music representation is a fundamental challenge in computational musicology. While grid-based representations effectively preserve pitch-time spatial correspondence, their inherent data sparsity leads to low encoding efficiency. Discrete-event representations achieve compact encoding but fail to adequately capture structural invariance and spatial locality. To address these complementary limitations, we propose Pianoroll-Event, a novel encoding scheme that describes pianoroll representations through events, combining structural properties with encoding efficiency while maintaining temporal dependencies and local spatial patterns. Specifically, we design four complementary event types: Frame Events for temporal boundaries, Gap Events for sparse regions, Pattern Events for note patterns, and Musical Structure Events for musical metadata. Pianoroll-Event strikes an effective balance between sequence length and vocabulary size, improving encoding efficiency by 1.36\times to 7.16\times over representative discrete sequence methods. Experiments across multiple autoregressive architectures show models using our representation consistently outperform baselines in both quantitative and human evaluations.

Automated piano performance evaluation traditionally relies on symbolic (MIDI) representations, which capture note-level information but miss the acoustic nuances that characterize expressive playing. I propose using pre-trained audio foundation models, specifically MuQ and MERT, to predict 19 perceptual dimensions of piano performance quality. Using synthesized audio from PercePiano MIDI files (rendered via Pianoteq), I compare audio and symbolic approaches under controlled conditions where both derive from identical source data. The best model, MuQ layers 9-12 with Pianoteq soundfont augmentation, achieves R^2 = 0.537 (95% CI: [0.465, 0.575]), representing a 55% improvement over the symbolic baseline (R^2 = 0.347). Statistical analysis confirms significance (p < 10^-25) with audio outperforming symbolic on all 19 dimensions. I validate the approach through cross-soundfont generalization (R^2 = 0.534 +/- 0.075), difficulty correlation with an external dataset (rho = 0.623), and multi-performer consistency analysis. Analysis of audio-symbolic fusion reveals high error correlation (r = 0.738), explaining why fusion provides minimal benefit: audio representations alone are sufficient. I release the complete training pipeline, pretrained models, and inference code.

The prosthetic socket is an essential part in ensuring comfort and stability for the overall prosthesis system. This study proposes a multi-material/thickness individualized transtibial prosthetic socket that focuses on providing comfort. This study aims to identify the proper material and thickness to protect Calf areas and reduce the pressure around bone areas. First, the socket is divided into four parts depending on the pressure-sensitive/tolerant regions. After identifying the thickness range for each concerned area, the thickness and material are selected based on the Pressure-Pain Threshold (PPT) test, and the finalized design is then prototyped. The prototyped individualized socket (i-Socket) is 22% lighter than the participant's own socket. Results of the pilot experiments with an amputee participant showed that the pressure inside the socket decreased by 45% and 31% for the Tibia and Fibula regions, respectively. Additionally, the self-selected CoM velocity for the walking experiment is increased by 15% compared to the similar studies in the literature. Regarding the kinematic results, symmetry in the knee and ankle joints increased by 65% and 2% when using the i-Socket compared to the results with the participant's own socket.