Papers

Dynamic spectrum slicing is a critical enabler for 6G Radio Access Networks (RANs), allowing the coexistence of heterogeneous services. However, optimizing resource allocation in dense, interference-limited deployments remains challenging due to non-stationary channel dynamics, strict Quality-of-Service (QoS) requirements, and the need for data privacy. In this paper, we propose SliceFed, a novel Federated Constrained Multi-Agent Deep Reinforcement Learning (F-MADRL) framework. SliceFed formulates the slicing problem as a Constrained Markov Decision Process (CMDP) where autonomous gNB agents maximize spectral efficiency while explicitly satisfying inter-cell interference budgets and hard ultra-reliable low-latency communication (URLLC) latency deadlines. We employ a Lagrangian primal-dual approach integrated with Proximal Policy Optimization (PPO) to enforce constraints, while Federated Averaging enables collaborative learning without exchanging raw local data. Extensive simulations in a dense multi-cell environment demonstrate that SliceFed converges to a stable, safety-aware policy. Unlike heuristic and unconstrained baselines, SliceFed achieves nearly 100% satisfaction of 1~ms URLLC latency deadlines and exhibits superior robustness to traffic load variations, verifying its potential for reliable and scalable 6G spectrum management.

Automated phoneme-level pronunciation assessment is vital for scalable speech therapy and language learning, yet validated tools for Arabic remain scarce. We present Harf-Speech, a modular system scoring Arabic pronunciation at the phoneme level on a clinical scale. It combines an MSA phonetizer, a fine-tuned speech-to-phoneme model, Levenshtein alignment, and a blended scorer using longest common subsequence and edit-distance metrics. We fine-tune three ASR architectures on Arabic phoneme data and benchmark them with zero-shot multimodal models; the best, OmniASR-CTC-1B-v2, achieves 8.92% phoneme error rate. Three certified speech-language pathologists independently scored 40 utterances for clinical validation. Harf-Speech attains a Pearson correlation of 0.791 and ICC(2,1) of 0.659 with mean expert scores, outperforming existing end-to-end assessment frameworks. These results show Harf-Speech yields clinically aligned, interpretable scores comparable to inter-rater expert agreement.

Most nonlinear filters used in spacecraft navigation are based on a linear approximation of the optimal minimum mean square error estimator. The Unscented Kalman Filter (UKF) handles nonlinear dynamics through a sigma-point transform, but the resulting state estimate remains a linear function of the measurement. This paper proposes a polynomial approximation of the optimal Bayesian update, leading to a Polynomial Unscented Kalman Filter that retains the structure of the standard UKF but enriches the measurement update with higher-order (polynomial) terms. To compute the moments required by this polynomial estimator, we employ a Conjugate Unscented Transformation (CUT), which accurately captures higher-order central moments of the state and measurement. Numerical examples, including Clohessy-Wiltshire and Circular Restricted 3-Body dynamics with non-Gaussian measurement noise, illustrate that the resulting polynomial-CUT filters improve both state estimation accuracy and covariance consistency when compared with their linear counterparts.

Reliable detection of event-related potentials (ERPs) at the single-trial level remains a major challenge due to the low signal-to-noise ratio EEG recordings. In this work, we investigate whether incorporating prior knowledge about ERP templates into deep learning models can improve detection performance. We employ the Deep-Match framework for ERP detection using multi-channel EEG signals. The model is trained in two stages. First, an encoder-decoder architecture is trained to reconstruct input EEG signals, enabling the network to learn compact signal representations. In the second stage, the decoder is replaced with a detection module, and the network is fine-tuned for ERP identification. Two model variants are evaluated: a standard model with randomly initialized filters and a Deep-MF model in which input kernels are initialized using ERP templates. Model performance is assessed on a single-trial ERP detection task using leave-one-subject-out validation. The proposed Deep-MF model slightly outperforms the detector with standard kernel initialization for most held-out subjects. Despite substantial inter-subject variability, Deep-MF achieves a higher average F1-score (0.37) compared to the standard network (0.34), indicating improved robustness to cross-subject differences. The best performance obtained by Deep-MF reaches an F1-score of 0.71, exceeding the maximum score achieved by the standard model (0.59). These results demonstrate that ERP-informed kernel initialization can provide consistent improvements in subject-independent single-trial ERP detection. Overall, the findings highlight the potential of integrating domain knowledge with deep learning architectures for EEG analysis. The proposed approach represents a step toward practical wearable EEG and passive brain-computer interface systems capable of real-time monitoring of cognitive processes.

Autonomous science agents built on large language models (LLMs) are increasingly used to generate hypotheses, design experiments, and produce reports. However, prior work mainly targets open-ended scientific problems with subjective outputs that are difficult to evaluate. Scientific coding benchmarks, by contrast, provide executable outputs for objective assessment. Existing approaches remain engineering-driven pipelines, revealing the need for structured, end-to-end science agent frameworks for scientific coding tasks. We address this gap by focusing on scientific coding tasks, where evaluation can be made rigorously, and introducing an agent framework SciNav (Scientific Navigator) that enables more effective solution exploration. Our framework is designed to operate under constrained search budgets, moving beyond reliance on pre-defined success metrics and prolonged search cycles. Inspired by findings that comparative judgments often reveal finer-grained quality differences and therefore provide greater discriminative power than absolute scoring, our framework leverages pairwise relative judgments within a tree search process to select top-K promising solution branches, prune low-potential ones, and progressively narrow down the solution candidates on the selected branches guided by relative comparisons. We demonstrate our agent's effectiveness across different types of tasks on two benchmarks. Experiments show that SciNav significantly outperforms direct prompting and prior agents like OpenHands and Self-Debug across different base models, task types, and difficulty levels, and exceeds different frontier comparators such as random selection and LLM absolute scoring. These results confirm the strength of our agent design and highlight the effectiveness of relative judgment-guided top-K search for high-quality scientific coding, marking a step toward more practical science agents.

Speech-based AI educational applications have gained significant interest in recent years, particularly for children. However, children speech research remains limited due to the lack of publicly available datasets, especially for low-resource languages such as Arabic.This paper presents Abjad-Kids, an Arabic speech dataset designed for kindergarten and primary education, focusing on fundamental learning of alphabets, numbers, and colors. The dataset consists of 46397 audio samples collected from children aged 3 - 12 years, covering 141 classes. All samples were recorded under controlled specifications to ensure consistency in duration, sampling rate, and format. To address high intra-class similarity among Arabic phonemes and the limited samples per class, we propose a hierarchical audio classification based on CNN-LSTM architectures. Our proposed methodology decomposes alphabet recognition into a two-stage process: an initial grouping classification model followed by specialized classifiers for each group. Both strategies: static linguistic-based grouping and dynamic clustering-based grouping, were evaluated. Experimental results demonstrate that static linguistic-based grouping achieves superior performance. Comparisons between traditional machine learning with deep learning approaches, highlight the effectiveness of CNN-LSTM models combined with data augmentation. Despite achieving promising results, most of our experiments indicate a challenge with overfitting, which is likely due to the limited number of samples, even after data augmentation and model regularization. Thus, future work may focus on collecting additional data to address this issue. Abjad-Kids will be publicly available. We hope that Abjad-Kids enrich children representation in speech dataset, and be a good resource for future research in Arabic speech classification for kids.

To tackle the huge computational demand of large foundation models, activation-aware compression techniques without retraining have been introduced. However, since these methods highly rely on calibration data, domain shift issues may arise for unseen downstream tasks. We propose a test-time quantization (TTQ) framework which compresses large models on the fly at inference time to resolve this issue. With an efficient online calibration, instant activation-aware quantization can adapt every prompt regardless of the downstream tasks, yet achieving inference speedup. Several experiments demonstrate that TTQ can improve the quantization performance over state-of-the-art baselines.

Simultaneous speech-to-speech translation (SimulS2S) is essential for real-time multilingual communication, with increasing integration into meeting and streaming platforms. Despite this, SimulS2S remains underexplored in research, where current solutions often rely on resource-intensive training procedures and operate on short-form, pre-segmented utterances, failing to generalize to continuous speech. To bridge this gap, we propose SimulU, the first training-free policy for long-form SimulS2S. SimulU adopts history management and speech output selection strategies that exploit cross-attention in pre-trained end-to-end models to regulate both input history and output generation. Evaluations on MuST-C across 8 languages show that SimulU achieves a better or comparable quality-latency trade-off against strong cascaded models. By eliminating the need for ad-hoc training, SimulU offers a promising path to end-to-end SimulS2S in realistic, long-form scenarios.

Deep learning dominates speech processing but relies on massive datasets, global backpropagation-guided weight updates, and produces entangled representations. Assembly Calculus (AC), which models sparse neuronal assemblies via Hebbian plasticity and winner-take-all competition, offers a biologically grounded alternative, yet prior work focused on discrete symbolic inputs. We introduce an AC-based speech processing framework that operates directly on continuous speech by combining three key contributions:(i) neural encoding that converts speech into assembly-compatible spike patterns using probabilistic mel binarisation and population-coded MFCCs; (ii) a multi-area architecture organising assemblies across hierarchical timescales and classes; and (iii) cross-area update schemes for downstream tasks. Applied to two core tasks of boundary detection and segment classification, our framework detects phone (F1=0.69) and word (F1=0.61) boundaries without any weight training, and achieves 47.5% and 45.1% accuracy on phone and command recognition. These results show that AC-based dynamical systems are a viable alternative to deep learning for speech processing.

Self-attention scales quadratically with sequence length, limiting transformer-based speech models on edge devices. We introduce the Learnable Pulse Accumulator (LPA), an O(n) replacement that substitutes key-query dot products with learned gating functions: content-dependent rectangular pulses, periodic windows, and position-dependent basis functions. An MSE diagnostic sweep determines per-layer replacement difficulty and ordering. Replacing 8 of 12 wav2vec2-base layers yields 10.61% word error rate (WER) on LibriSpeech test-clean, +7.24 percentage points (pp) over the 3.37% baseline, with 3.27x speedup at 120s audio on Apple M4 Pro via an optimized MLX inference path. Cross-domain validation on SepFormer speech enhancement shows all 16 intra-chunk attention layers can be replaced without collapse, suggesting the depth wall arises from linguistic computation rather than an LPA limitation. LPA's near-binary gates at inference enable dense GPU computation with no CPU-GPU synchronization, and all operations map to mobile neural accelerators.

End-to-end automatic speech recognition often degrades on domain-specific data due to scarce in-domain resources. We propose a synthetic-data-based domain adaptation framework with two contributions: (1) a large language model (LLM)-based text augmentation pipeline with a filtering strategy that balances lexical diversity, perplexity, and domain-term coverage, and (2) phonetic respelling augmentation (PRA), a novel method that introduces pronunciation variability through LLM-generated orthographic pseudo-spellings. Unlike conventional acoustic-level methods such as SpecAugment, PRA provides phonetic diversity before speech synthesis, enabling synthetic speech to better approximate real-world variability. Experimental results across four domain-specific datasets demonstrate consistent reductions in word error rate, confirming that combining domain-specific lexical coverage with realistic pronunciation variation significantly improves ASR robustness.

Chest X-ray imaging remains the primary diagnostic tool for pulmonary and cardiac disorders worldwide, yet its accuracy is hampered by radiologist shortages and inter-observer variability. This study presents a systematic comparative evaluation of seven deep learning architectures for multi-class chest disease classification: ConvNeXt-Tiny, DenseNet121, DenseNet201, ResNet50, ViT-B/16, EfficientNetV2-M, and MobileNetV2. A balanced dataset of 18,080 chest X-ray images spanning five disease categories (Cardiomegaly, COVID-19, Normal, Pneumonia, and Tuberculosis) was constructed from three public repositories and partitioned at the patient level to prevent data leakage. All models were trained under identical conditions using ImageNet-pretrained weights, standardized preprocessing, and consistent hyperparameters. All seven architectures exceeded 90% test accuracy. ConvNeXt-Tiny achieved the highest performance (92.31% accuracy, 95.70% AUROC), while MobileNetV2 emerged as the most parameter-efficient model (3.5M parameters, 90.42% accuracy, 94.10% AUROC), completing training in 48 minutes. Tuberculosis and COVID-19 classification was near-perfect (AUROC >= 99.97%) across all architectures, while Normal, Cardiomegaly, and Pneumonia presented greater challenges due to overlapping radiographic features. Grad-CAM visualizations confirmed clinically consistent attention patterns across disease categories. These findings demonstrate that high-accuracy multi-disease chest X-ray classification is achievable without excessive computational resources, with important implications for AI-assisted diagnosis in both resource-rich and resource-constrained healthcare settings.

Deep learning has achieved transformative performance across diverse domains, largely driven by large-scale and high-quality training data. In contrast, the development of brain-computer interfaces (BCIs) is fundamentally constrained by limited, heterogeneous, and privacy-sensitive neural recordings. Generating synthetic yet physiologically plausible brain signals has therefore emerged as a promising strategy to mitigate data scarcity, improve model generalization, and support data-efficient BCIs. This survey provides a comprehensive review of synthetic brain data generation for BCIs, covering methodological taxonomies, benchmark experiments, evaluation metrics, key applications, and future directions. We systematically categorize existing generation approaches into four types: signal-transformation-based, feature-based, model-based, and translation-based generation, and discuss their characteristics, advantages, and limitations. Furthermore, we benchmark representative brain signal generation approaches across four BCI paradigms, including motor imagery, epileptic seizure detection, steady-state visually evoked potentials, and auditory attention detection, to provide an objective comparison of their downstream utility. We also summarize evaluation principles for generated brain signals from multiple perspectives, including signal realism, physiological plausibility, downstream utility, and privacy preservation. Finally, we discuss the potential and challenges of current generation approaches and outline future research directions toward accurate, data-efficient, generalizable, and privacy-aware BCI systems. The benchmark codebase is available at https://github.com/wzwvv/DG4BCI.

We investigate continued pretraining (CPT) for adapting wav2vec2-bert-2.0 to Swahili automatic speech recognition (ASR). Our approach combines unlabeled audio with limited labeled data through pseudo-labeled CPT followed by supervised finetuning. With 20,000 labeled samples, we achieve 3.24% WER on Common Voice Swahili-an 82% relative improvement over the baseline. This result surpasses the best previously reported academic system (8.3% WER from XLS-R) by 61% relative improvement. We provide concrete data requirements and a replicable methodology applicable to other low-resource languages.

This paper investigates the susceptibility of Integrated Sensing and Communication (ISAC) systems to hostile jamming, focusing on an aerial Reconfigurable Holographic Surface (RHS)-aided unmanned aerial vehicle (UAV). The proposed framework, termed RHOSI, enhances ISAC's resilience by dynamically shaping the wireless propagation environment. Specifically, RHOSI introduces a strategy to improve jamming resistance by jointly optimizing transmit beamforming at the hybrid base station, RHS phase shift configuration, and UAV spatial deployment, while ensuring the required echo signal-to-interference-plus-noise ratios for reliable sensing. The resulting non-linear optimization problem features highly coupled variables, which are decomposed into sub-problems and solved using an alternating optimization (AO) approach. Simulation results confirm the practicality and effectiveness of RHOSI in significantly improving the throughput and robustness of ISAC under adversarial jamming.

Voice biometric systems can exhibit sex-related performance gaps even when overall verification accuracy is strong. We attribute these gaps to two practical mechanisms: (i) demographic shortcut learning, where speaker classification training exploits spurious correlations between sex and speaker identity, and (ii) feature entanglement, where sex-linked acoustic variation overlaps with identity cues and cannot be removed without degrading speaker discrimination. We propose Fair-Gate, a fairness-aware and interpretable risk-gating framework that addresses both mechanisms in a single pipeline. Fair-Gate applies risk extrapolation to reduce variation in speaker-classification risk across proxy sex groups, and introduces a local complementary gate that routes intermediate features into an identity branch and a sex branch. The gate provides interpretability by producing an explicit routing mask that can be inspected to understand which features are allocated to identity versus sex-related pathways. Experiments on VoxCeleb1 show that Fair-Gate improves the utility–fairness trade-off, yielding more sex-fair ASV performance under challenging evaluation conditions.

Many control, optimization, and learning algorithms rely on discretizations of continuous-time contracting systems, where preservation of contractivity under numerical integration is key for stability, robustness, and reliable fixed-point computation. In this paper, we establish conditions under which multi-stage Runge-Kutta methods preserve strong contractivity when discretizing infinitesimally contractive continuous-time systems. For explicit Runge-Kutta methods, preservation conditions are derived by bounding Lipschitz constants of the associated composite stage mappings, leading to coefficient-dependent criteria. For implicit methods, the algebraic structure of the stage equations enables explicit conditions on the Runge-Kutta coefficients that guarantee preservation of strong contractivity. In the implicit case, these results extend classical guarantees, typically limited to weak contractivity in the Euclidean metric, to strong contractivity with respect to the $\ell_1$-, $\ell_2$-, and $\ell_\infty$-norms. In addition, we study well-definedness of implicit methods through an auxiliary continuous-time system associated with the stage equations. We show that strong infinitesimal contractivity of this auxiliary system is sufficient to guarantee unique solvability of the stage equations. This analysis generalizes standard well-definedness conditions and provides a dynamic implementation approach that avoids direct solution of the implicit algebraic equations.

Threshold-free cluster enhancement (TFCE) integrates cluster extent across thresholds to improve voxel-wise neuroimaging inference, but permutation testing makes it prohibitively slow for large datasets. Probabilistic TFCE (pTFCE) uses analytical Gaussian random field (GRF) p-values but discretises the threshold grid. Exact TFCE (eTFCE) eliminates discretisation via a union-find data structure but still requires permutations. We combine eTFCE's union-find for exact cluster-size retrieval with pTFCE's analytical GRF inference. The union-find builds the cluster hierarchy in one pass over sorted voxels and enables exact size queries at any threshold; GRF theory then converts these sizes to analytical p-values without permutations. Validation on synthetic phantoms (64^3, 80 subjects): FWER controlled at nominal level (0/200 null rejections, 95% CI [0.0%, 1.9%]); power matches baseline pTFCE (Dice >= 0.999); smoothness error below 1%; concordance r > 0.99. On UK Biobank (N=500) and IXI (N=563), significance maps form strict subsets of reference R pTFCE, which supports conservative error control. Implemented in pytfce (pip install pytfce): baseline completes whole-brain VBM in ~5s (75x faster than R pTFCE), hybrid in ~85s (4.6x faster) with exact cluster sizes; both >1000x faster than permutation TFCE.

This paper presents an implementation and evaluation of a Distributed Kalman–Consensus Filter (DKCF) for Multi-Object Tracking (MOT) in mobile robot networks operating under partial observability and heterogeneous localization uncertainty. A key challenge in such systems is the fusion of information from agents with differing localization quality, where frame misalignment can lead to inconsistent estimates, track duplication, and ghost tracks. To address this issue, we build upon the MOTLEE framework and retain its frame-alignment methodology, which uses consistently tracked dynamic objects as transient landmarks to improve relative pose estimates between robots. On top of this framework, we propose an uncertainty-aware adaptive consensus weighting mechanism that dynamically adjusts the influence of neighbor information based on the covariance of the transmitted estimates, thereby reducing the impact of unreliable data during distributed fusion. Local tracking is performed using a Kalman Filter (KF) with a Constant Velocity Model (CVM) and Global Nearest Neighbor (GNN) data association. simulation results demonstrate that adaptive weighting effectively protects local estimates from inconsistent data, yielding a MOTA improvement of 0.09 for agents suffering from localization drift, although system performance remains constrained by communication latency.

Anisotropic image analysis is ubiquitous in medical and scientific imaging, and while the literature on the subject is extensive, the robustness to numerical rotations of numerous methods remains to be studied. Indeed, the principal directions and angular profile of a rotated image are often expected to rotate accordingly. In this work, we propose a new spectral method for the anisotropic analysis of images (EquivAnIA) using two established directional filters, namely cake wavelets, and ridge filters. We show that it is robust to numerical rotations throughout extensive experiments on synthetic and real-world images containing geometric structures or textures, and we also apply it successfully for a task of angular image registration. The code is available at https://github.com/jscanvic/Anisotropic-Analysis

Low Earth orbit (LEO) inter-satellite links (ISLs) must achieve joint synchronization and ranging under severe hardware impairments, namely oscillator phase noise, clock drift, and measurement outliers, exacerbated by rapid relative dynamics exceeding 7~km/s. In coherent Doppler processing, the frequency observable depends on the difference between consecutive carrier phase states, creating a cross-epoch coupling structure that fundamentally affects estimation-theoretic performance limits. This paper makes three contributions. First, we prove analytically that this cross-epoch Doppler coupling is necessary to avoid unbounded carrier phase uncertainty: without it, phase variance grows linearly without bound. Second, we derive a posterior Cramér-Rao bound (PCRB) via the Tichavský recursion that explicitly incorporates the resulting 10$\times$10 block information structure. Third, we propose a hybrid robust filtering framework combining hard gating for impulsive cycle-slip outliers with Huber M-estimation for heavy-tail contamination, using TASD-aware innovation covariance to account for cross-epoch uncertainty in residual normalization. Monte Carlo simulations at Ka-band confirm that the PCRB accurately lower-bounds estimator performance under nominal conditions, while the hybrid method reduces 95th-percentile phase error by 27–93% compared to standard extended Kalman filtering across different outlier regimes.

This work extends previous 1D irreversible port-Hamiltonian system (IPHS) formulations to boundary-controlled ND distributed parameter systems describing conduction-diffusion fluid phenomena. Within a unified and thermodynamically consistent framework, we show that conduction and diffusion can be represented through a single coherent structure that preserves global energy balance and ensures a correct characterization of entropy production. The resulting formulation provides a foundation for the systematic modeling and control of complex multi-physical processes governed by coupled transport mechanisms in N dimensions. In the longer term, this framework opens the door to structure-preserving numerical schemes capable of enforcing thermodynamic principles directly at the discretized level.

Coverage control is essential for the optimal deployment of agents to monitor or cover areas with sensory demands. While traditional coverage involves single-task robots, increasing autonomy now enables multitask operations. This paper introduces a novel multitask coverage problem and addresses it for both the cases of known and unknown sensory demands. For known demands, we design a federated multitask coverage algorithm and establish its convergence properties. For unknown demands, we employ a multitask Gaussian Process (GP) framework to learn sensory demand functions and integrate it with the multitask coverage algorithm to develop an adaptive algorithm. We introduce a novel notion of multitask coverage regret that compares the performance of the adaptive algorithm against an oracle with prior knowledge of the demand functions. We establish that our algorithm achieves sublinear cumulative regret, and numerically illustrate its performance.

The Irreversible Port-Hamiltonian Systems (IPHS) framework is extended to the modelling of non-isentropic fluids with viscous dissipation in the Eulerian description. Building on earlier IPHS formulations for diffusion-driven and non-convective distributed systems, it is shown that convective transport can be consistently encompassed by the framework by modifying the underlying differential operators. After revisiting the constitutive relations of non-isentropic fluids in both Eulerian and Lagrangian coordinates, it is demonstrate how these systems fit within an extended IPHS formulation. Furthermore, an extended parametrisation of the boundary port variables which ensures that the first and second laws of Thermodynamics are fulfilled allows to define a general class of boundary controlled IPHS.

We propose self-speculative decoding for speech-aware LLMs by using the CTC encoder as a draft model to accelerate auto-regressive (AR) inference and improve ASR accuracy. Our three-step procedure works as follows: (1) if the frame entropies of the CTC output distributions are below a threshold, the greedy CTC hypothesis is accepted as final; (2) otherwise, the CTC hypothesis is verified in a single LLM forward pass using a relaxed acceptance criterion based on token likelihoods; (3) if verification fails, AR decoding resumes from the accepted CTC prefix. Experiments on nine corpora and five languages show that this approach can simultaneously accelerate decoding and reduce WER. On the HuggingFace Open ASR benchmark with a 1B parameter LLM and 440M parameter CTC encoder, we achieve a record 5.58% WER and improve the inverse real time factor by a factor of 4.4 with only a 12% relative WER increase over AR search. Code and model weights are publicly available under a permissive license.

The automatic identification of cough segments in audio through the determination of start and end points is pivotal to building scalable screening tools in health technologies for pulmonary related diseases. We propose the application of two current pre-trained architectures to the task of cough activity detection. A dataset of recordings containing cough from patients symptomatic for tuberculosis (TB) who self-present at community-level care centres in South Africa and Uganda is employed. When automatic start and end points are determined using XLS-R, an average precision of 0.96 and an area under the receiver-operating characteristic of 0.99 are achieved for the test set. We show that best average precision is achieved by utilising only the first three layers of the network, which has the dual benefits of reduced computational and memory requirements, pivotal for smartphone-based applications. This XLS-R configuration is shown to outperform an audio spectrogram transformer (AST) as well as a logistic regression baseline by 9% and 27% absolute in test set average precision respectively. Furthermore, a downstream TB classification model trained using the coughs automatically isolated by XLS-R comfortably outperforms a model trained on the coughs isolated by AST, and is only narrowly outperformed by a classifier trained on the ground truth coughs. We conclude that the application of large pre-trained transformer models is an effective approach to identifying cough end-points and that the integration of such a model into a screening tool is feasible.

We present an intelligent wearable system to monitor and predict mood states of elderly people during their daily life activities. Our system is composed of a wristband to record different physiological activities together with a mobile app for ecological momentary assessment (EMA). Machine learning is used to train a classifier to automatically predict different mood states based on the smart band only. Our approach shows promising results on mood accuracy and provides results comparable with the state of the art in the specific detection of happiness and activeness.

Large language models (LLMs), trained on large-scale text, have recently attracted significant attention for their strong performance across many tasks. Motivated by this, we investigate whether a text-trained LLM can help localize fake words in partially fake speech, where only specific words within a speech are edited. We build a speech LLM to perform fake word localization via next token prediction. Experiments and analyses on AV-Deepfake1M and PartialEdit indicates that the model frequently leverages editing-style pattern learned from the training data, particularly word-level polarity substitutions for those two databases we discussed, as cues for localizing fake words. Although such particular patterns provide useful information in an in-domain scenario, how to avoid over-reliance on such particular pattern and improve generalization to unseen editing styles remains an open question.

Audio-visual emotion recognition (AVER) methods typically fuse utterance-level features, and even frame-level attention models seldom address the frame-rate mismatch across modalities. In this paper, we propose a Transformer-based framework focusing on the temporal alignment of multimodal features. Our design employs a multimodal self-attention encoder that simultaneously captures intra- and inter-modal dependencies within a shared feature space. To address heterogeneous sampling rates, we incorporate Temporally-aligned Rotary Position Embeddings (TaRoPE), which implicitly synchronize audio and video tokens. Furthermore, we introduce a Cross-Temporal Matching (CTM) loss that enforces consistency among temporally proximate pairs, guiding the encoder toward better alignment. Experiments on CREMA-D and RAVDESS datasets demonstrate consistent improvements over recent baselines, suggesting that explicitly addressing frame-rate mismatch helps preserve temporal cues and enhances cross-modal fusion.

Metaheuristic algorithms are currently widely used to solve a variety of optimization problems across various industries. This article discusses the application of a metaheuristic algorithm to optimize the hierarchical architecture of an industrial distributed control system. The success of the algorithm depends largely on the choice of starting conditions and algorithm parameters. We examine the impact of parameter selection on the convergence of a modified ant colony algorithm and provide recommendations for tuning the algorithm to achieve optimal results for a specific industrial problem. The findings presented in this article can also be applied to other combinatorial optimization problems.