Papers

Data-driven surrogates reduce the nonlinear solves required for probabilistic power flow (PPF), but models tied to a fixed bus–branch graph often generalize poorly when outages change electrical coupling. This paper proposes a topology-decomposition-driven hierarchical bidirectional graph neural network (TDD–HBGNN) for PPF under unseen topologies. For each realized topology, the network is decomposed into fundamental topological units (FTUs), dynamically homogeneous voltage clusters (DHVCs), and a core backbone sub-grid (CBSG), with equivalent-admittance representations preserving electrical ports between units. A sensitivity-based linear prior captures the first-order voltage-magnitude and phase-angle response, while the HBGNN learns topology-dependent nonlinear residuals. Impedance- and uncertainty-aware bottom-up aggregation condenses local perturbations, while mode-modulated top-down broadcasting restores system context at buses. We evaluate our approach on the IEEE 118-bus and Polish 2746-bus systems under unseen N–1 and N–2 contingencies against six baseline methods. Compared with these methods, TDD–HBGNN achieves 2.47–4.20-fold lower all-bus phase-angle errors and lower phase-angle errors in every held-out event. Its voltage-magnitude advantage is concentrated near the outaged branches, where paired error ratios reach 1.75–2.58 on the IEEE system and 1.87–1.97 in Polish N–2 events. All system–contingency–baseline comparisons favor TDD–HBGNN in joint voltage–angle error. Moreover, shared hierarchical operators keep the model size at 0.658 million parameters across both grids and yield 0.161–1.793~ms forward latency per probabilistic case. These results demonstrate system-wide angular accuracy, outage-local voltage accuracy, and parameter-efficient scaling across unseen topologies.

Nonlinear model predictive control (NMPC) can be applied to large nonlinear systems with complex objectives while enforcing constraints, but the cost of solving the underlying optimization problems limits real-time application for systems with fast dynamics or many states. Neural networks can reduce this online cost by approximating the control law, but may violate constraints and typically require many precomputed optimizer solutions for training. The learning-based iterative solver for constrained optimization (LISCO) instead learns to predict primal-dual steps toward satisfaction of the Karush-Kuhn-Tucker (KKT) conditions through self-supervised training without precomputed solutions, but its dense, monolithic architecture grows with the prediction horizon. We remove this horizon dependence by exploiting the temporal structure of the problem in the solver architecture, which makes the network size independent of the horizon while leaving the training loss and the solver iteration unchanged. A closed-loop resampling strategy further aligns the training distribution with closed-loop operation at no additional sampling cost. On an economic NMPC benchmark of a CSTR cascade, the proposed solver scales to larger problems than monolithic LISCO and achieves median speedups of up to a factor of four over IPOPT. In closed loop, it achieves comparable control performance to IPOPT with no observed constraint violations.

Incentives can align private interests with collective goals, but their design becomes intertwined with learning when agents' preferences are unknown. This paper develops an adaptive incentive-design framework for multi-agent linear-quadratic dynamic games, where a system planner can simultaneously learn agents' private preferences and steer their strategic behavior towards the social optimum. To this end, we first characterize the complete family of affine incentives that implement the social optimum as the unique open-loop Nash equilibrium while ensuring exact budget balance. When agents' preferences are unknown, we develop an online algorithm to estimate their preference parameters and establish strong consistency of the estimates under a weak trajectory-dependent excitation condition, without requiring persistent excitation or i.i.d. trajectories. We then integrate this estimator into an active-learning incentive mechanism that combines certainty-equivalent coordination with decaying random probing. We prove that the probing generates sufficient information through the agents' equilibrium responses and the system dynamics, ensuring consistent preference learning from the resulting closed-loop trajectories. We show that the induced trajectories converge almost surely to their socially optimal counterparts, while cumulative social-cost regret relative to the full-information oracle and cumulative net-budget imbalance grow sublinearly. Simulations of a transboundary pollution game demonstrate the effectiveness of the proposed framework.

Speech recognition on low-cost microcontrollers is, in practice, limited to keyword spotting or a fixed list of commands. We present Oído, an inference engine that runs open-vocabulary English speech recognition entirely on an ESP32-S3 (240 MHz dual-core Xtensa LX7, 512 KB of SRAM, 8 MB of PSRAM, 16 MB of flash, no neural accelerator). Oído executes a 13M-parameter Conformer-CTC model with int8 weights, per-frame dynamic int8 activations and int8 attention on the chip's SIMD instructions, with a schedule designed around flash and PSRAM bandwidth. On LibriSpeech test-clean/test-other the on-device arithmetic reaches 3.70%/8.23% word error rate (WER), within 0.12 points of full precision, and 3.33%/7.15% with an on-chip 1.3M-parameter language model; a transducer head gives 2.99%/6.67% and an int4 variant fits in 8.3 MB. This is 2.3-2.8x fewer errors than the vendor's recognizer on the same chip, and fewer than Whisper tiny.en and Moonshine tiny in full precision on a laptop, including in real noise and reverberation. Measured on a physical board, the firmware runs at 1.97x the duration of the audio on one core, with transcripts identical to an instruction-exact emulator: accurate, but not yet real time (a 4 s command is transcribed about 8.7 s after it ends). The two-core schedule gives wrong results on silicon and is disabled. A streaming variant shows partial text while the speaker talks and delivers the final text 4.1-11.1 s after clips of 3.6-10.6 s (7.2-22.4 s in utterance mode) at 4.86%/11.03% WER; a Spanish model makes 34-60% fewer errors than multilingual Whisper tiny. Code and models are released.

Adaptive notch filters (ANFs) based on the normalized least mean squares (NLMS) algorithm are an established technique for suppressing narrowband interference in GNSS receivers. The traditional solution for multiple interferers is simply cascading the NLMS-ANF into multiple stages but this solution degrades performance when multiple chirp jammers of similar power are present. We propose a low-complexity modification to the cascaded NLMS-ANF update rule to resolve this: a local gradient (LG) formulation: LG-ANF. The LG-ANF forms a locally bandlimited gradient signal by translating the input to baseband around the current frequency estimate, low-pass filtering it, and translating it back, thereby attenuating competing interferers before adaptation. The modification generalizes the traditional NLMS-ANF to handle the similar-power multiple interferer problem, using just one additional parameter. We characterize the gradient-averaging failure analytically and visually, demonstrate independent chirp tracking in the worst-case two-chirp scenarios, and evaluate statistical suppression performance by sweeping amplitude ratio against a standard NLMS-ANF, a frequency-locked-loop (FLL)-based ANF and a per-sample oracle upper bound. We further characterize the LG advantage over various ANF parameters and chirp conditions. Simulations show that the LG-ANF reduces to the standard NLMS-ANF outside the pathological multi-chirp regime, while recovering up to 4dB of in-band suppression around pathological amplitude ratios under per-condition optimization of the local bandwidth, and up to 13dB for selected edge-case parameter combinations. These optimized results represent an upper bound on the achievable LG gain. In a hardware experiment using a fixed alpha_lp=0.5, the LG-ANF extends position-lock survival by 24s relative to the standard ANF.

Recent accent classification systems can achieve high accuracy in standard in-corpus evaluations. However, when speakers overlap between the training and test sets, such results may overestimate model robustness compared with real-world deployment scenarios. This study revisits the robustness of DistilHuBERT-based accent classification by using L2-ARCTIC and multiple external English accent datasets. The comparison covers file-level and speaker-disjoint splits, along with balanced external-data training and teacher-student pseudo-labeling. The large performance gap between in-corpus and external evaluations suggests that file-level evaluation substantially overestimates performance, while speaker-disjoint evaluation can provide a more realistic estimate of the generalization to unseen speakers. In this study, models trained on external data alone do not reliably generalize to L2-ARCTIC. Among the tested configurations, balanced external-data training offers the most promising supervised setting. The pseudo-labeling experiments further indicate that confidence-based filtering alone is insufficient under domain mismatch. Overall, this study suggests that robust accent classification evaluation should simultaneously include speaker-disjoint evaluation and cross-corpus testing.

Building upon our preliminary study, this paper proposes a standalone beam-management framework for mobile MIMO-OFDM systems that integrates Blind Initial Beam Acquisition (BIBA) with Clock-Assisted Re-Acquisition (CARA). BIBA is a blind and asynchronous protocol that requires neither prior synchronization nor an out-of-band feedback channel, so the Base Station (BS) and User Equipment (UE) can independently identify the beam pair maximizing received signal strength. To handle subsequent link drops, the newly introduced CARA protocol exploits the clock synchronization established after BIBA to reduce the re-acquisition time by approximately 55% compared with BIBA. The operational reliability of both protocols is bounded by a probabilistic feasibility analysis under random UE mobility trajectories. Protocols are validated on a 28 GHz mmWave testbed under practical hardware constraints, including beam-switching and Time Division Duplex (TDD) latencies, and measured Software-Defined Radio (SDR) software processing delays. Finally, we formulate a unified Finite State Machine (FSM) model for both BIBA and CARA to ensure robust integration into practical wireless systems

This paper addresses the grid quantization error of deep-learning (DL)-based direction-of-arrival (DOA) estimation. Classification- based DOA methods discretize the continuous angular space into a grid, so that with a peak-based read-out, the accuracy and resolution are limited by the grid step and by the width of the label peak. To address this issue, we propose a sigmoid label that encodes the sources' DOAs as the sum of angle-shifted monotonic sigmoid functions. Differentiating the output of a CNN-Transformer network trained with the sigmoid label produces sharp spectral peaks at the source DOAs. Combined with a moderately refined grid step of 0.25 degrees, the CNN-Transformer with the proposed sigmoid label attains sub-degree accuracy and a 1-degree resolution, breaking the limit of the grid quantization error. Under the unified CNN-Transformer backbone, we compare the proposed Sigmoid label with the Gaussian label, the 0-1 hard label, and the binary cross-entropy (BCE) baseline. Numerical results demonstrate that the Sigmoid label attains the highest accuracy and superior resolution compared to other labels.

Earnings call transcription presents a realistic continual learning scenario where Automatic Speech Recognition (ASR) systems are updated with new recordings but risk catastrophic forgetting of previously learned financial speech patterns. Adapting Whisper-small from a general-domain speech corpus to the Earnings22 benchmark reveals the opposite: all evaluated methods exhibit improvement on held-out financial speech from the same domain during adaptation rather than forgetting. This task complementarity makes regularization counterproductive. Nevertheless, when regularization is warranted, a layer-wise mini-Kronecker-Factored Approximate Curvature (mini-KFAC) approximation seems to achieve lower word error rates (WERs) and more stable optimization than standard diagonal Elastic Weight Consolidation (EWC). The results suggest that task complementarity should be assessed before applying continual learning constraints, and prefer mini-KFAC EWC over diagonal EWC when needed.

Time series causal discovery — the process of detecting causation between time series — has its roots in signal processing research from over a half century ago. It has since received active and widespread interest, evolving into different forms across domains such as statistics, dynamical systems and machine learning. Entering this fragmented and constantly evolving research landscape can be daunting. This tutorial provides a unified overview of time series causal discovery across research communities. We introduce the key principles invoked by modern discovery methods, and we discuss their limitations and assumptions. We also discuss how discovery methods are evaluated, benchmarked and fine-tuned. We further discuss how causal discovery for time series is becoming crucial for downstream tasks in agriculture and Earth science, in healthcare, and in artificial intelligence systems. We believe that this tutorial will serve as a valuable resource in navigating the deluge of recent causality research, as well as an invitation to signal processing audiences to contribute to causal problems. To further help readers, we host a companion repository to link to useful code packages and resources at https://github.com/CHAI-UK/Awesome-Time-Series-Causal-Discovery.

The open-loop minimum-energy control (MEC) problem asks for control inputs that minimize the control energy while ensuring the terminal state reaches a desired value. In this paper, we study the open-loop MEC problem for stochastic linear time-invariant systems. We consider a variety of stochastic constraints that enforce the terminal error to be small, and show that Stochastic MEC (SMEC) problem can be formulated as a QCQP. We obtain the optimal solution for SMEC and show that it generalizes the well-known Deterministic MEC problem by introducing novel stochastic variants of controllability and Gramian matrices. Finally, we show that there exists a fundamental lower bound on the optimal control energy which cannot be overcome by increasing the time horizon of the problem.

Traditional system identification typically involves two steps: initially, training data is processed to estimate a model of the system of interest; then, the model is used to predict outputs for new input queries. In contrast, recent in-context learning approaches learn a meta model that maps directly from training data and query inputs to outputs in a zero-shot fashion, bypassing explicit model estimation. This approach leverages a meta model trained offline, on large sets of synthetically generated data, to emulate the behavior of a system identification algorithm. In this work, we advance this meta-learning framework for system identification in three directions. First, we develop a unified probabilistic meta-identification architecture that combines a Transformer-based encoder-decoder with tailored mechanisms for handling long training sequences, enabling calibrated predictions over long horizons. Second, we show how a pre-trained meta model can act as an uncertainty-aware synthetic data generator for classical system identification, providing task-specific data that regularize parametric models when real data are scarce. Third, we systematically evaluate the framework on standard nonlinear system identification benchmarks and on a challenging out-of-distribution scenario, demonstrating competitive accuracy with state-of-the-art methods and strong generalization, including cases where input-only adaptation is sufficient to handle significant distribution shifts.

Forecasting desertification has received far less attention than mapping it. We use a convolutional LSTM (ConvLSTM) trained on ten annual Earth Observation (EO) variables (2000-2023) over the semi-arid state of Ceará, Brazil, on a 256x206 grid at roughly 2.2 km resolution, to measure how much of the year-to-year evolution of a MapBiomas-derived degradation index (IDV) can be predicted from satellite data. The network predicts a bounded change on top of the last observed state, is trained in two stages (single-step, then recursive rollout) and is evaluated on a contiguous spatial and temporal holdout. State-wide mean IDV rose from 0.238 to 0.350 between 2000 and 2023, about 0.005 per year, but cell-level change is dominated by year-to-year jumps, and a gradient-boosting model given all channels explains only 4.7% of its variance. Against this ceiling, the ConvLSTM lowers RMSE relative to persistence by 1.6% one year ahead and 4.4% four years ahead in a held-out southern band (p < 0.001), beats persistence with a constant drift and reproduces 53-69% of the regional drift. Drift-aware baselines only help from the third year onward, by up to 4.4% in RMSE at six years. An earlier formulation that redrew the whole field each year predicted changes unrelated to the observed ones (correlation -0.06) and produced a spurious decline when run recursively. Channel knockouts suggest that elevation and EVI degrade the forecast, while slope and aspect have no measurable effect. Monte Carlo dropout and stochastic scenario maps convey the uncertainty that the expected-value forecast hides. Under climatological forcing, mean IDV rises from 0.350 in 2023 to 0.370 in 2030. EO-based desertification forecasting is therefore reliable for the regional trend, not for individual cells.

We propose a coupling-aware tri-hybrid precoding architecture for simultaneous wireless information and power transfer (SWIPT) using reconfigurable parasitic antennas. By integrating multiport network theory and information theory, we develop a third precoding layer that explicitly accounts for mutual coupling and reactive energy storage within the antenna domain. To efficiently optimize this architecture, we propose a low-complexity tri-hybrid solution comprising coordinate descent (CD) for parasitic reactive load optimization, a singular value decomposition (SVD)-based analog precoder, and a protective partial zero-forcing (PPZF) digital precoder. Numerical results demonstrate that the transmit array introduces impedance mismatch that reduces the real power accepted by the antenna array. Moreover, the proposed tri-hybrid architecture significantly outperforms hybrid precoding by constructively exploiting the coupling effects, yielding substantial gains in spectral efficiency (SE) and sum non-linear harvested energy (NLHE) at a fraction of the hardware cost.

Thyroid nodule malignancy assessment relies on ultrasound imaging, yet current risk stratification systems suffer from substantial interobserver variability, contributing to a high rate of unnecessary biopsies. We propose a radiomics-based framework combining quantitative feature extraction via PyRadiomics with a Generalized Additive Model (GAM) classifier, selected among six candidates for its favorable trade-off between discriminative performance and interpretability. To complement point predictions with uncertainty estimates, Cross-Conformal Prediction with label-conditional guarantees is applied, flagging ambiguous cases for clinical review rather than forcing a binary decision. Experiments are conducted on TN5000, a publicly available ultrasound dataset of thyroid nodules. After Spearman correlation filtering, 36 features are retained and the GAM achieves an AUC of $0.89$, a sensitivity of $0.83$, and a specificity of $0.82$ on the validation set. At a significance level $\epsilon=0.09$, selected to ensure empirical coverage guarantees on both classes, 63% of nodules on the test set receive a certain (i.e., single-label) prediction, with conditional precision of $0.94$ for certain malignant and conditional Negative Predictive Value (NPV) of $0.73$ for certain benign cases, while the remaining 37% are referred to the clinician.

Physical layer (PHY) information hiding refers to the concealment of secrets within PHY waveforms without compromising the integrity of the primary communication link, thereby enabling critical applications such as covert communication, PHY steganography, and watermarking. The widespread deployment of Wi-Fi has motivated substantial research interest in hiding secrets within Wi-Fi waveforms. However, existing Wi-Fi information hiding approaches suffer from two limitations: restricted steganographic capacity and insufficient robustness against environmental variations, both of which severely constrain practical deployment. This work proposes to embed secrets within the Wi-Fi legacy short training field (L-STF). The designed method preserves the autocorrelation properties of Wi-Fi L-STF, ensuring that packet detection and coarse frequency offset compensation functionalities remain uncompromised. The secret embedding and extraction processes employ a pair of neural networks designed to learn optimal embedding rules and maximize steganographic capacity. We developed a prototype system and conducted comprehensive experiments across both indoor and outdoor environments. Experimental results show that our method can reliably embed 46 secret bits within Wi-Fi packets across diverse channel conditions, achieving significantly enhanced steganographic capacity and environmental robustness. The hardware prototype study validates the feasibility of the proposed Wi-Fi information hiding scheme.

Secure transmission of electrocardiogram (ECG) signals remains a major challenge in telecardiology. While traditional mathematical watermarking can embed patient metadata directly into ECG signals, it is highly fragile against temporal desynchronization such as circular phase shifts. In contrast, existing deep watermarking approaches for temporal signals, such as audio, typically rely on lossy quantization, which compromises the high-frequency micro-morphological details critical for clinical diagnosis. To bridge this gap, we propose ECGWmark, a fully convolutional and quantization-free deep watermarking framework for medical time-series. ECGWmark operates in a continuous, full-resolution latent space and utilizes a localized frame-level mechanism to improve the robustness to severe desynchronization. Crucially, we introduce a differentiable Physiological Loss guided by a clinical classifier to preserve diagnostic semantics. Experiments on the benchmark MIT-BIH database show that ECGWmark maintains channel-bit extraction accuracy above 90% under the evaluated circular temporal shifts. Furthermore, under a rigorous inter-patient diagnostic evaluation, Class-V sensitivity changes from 71.97% to 76.50%, whereas Macro-F1 changes from 35.25% to 34.71% and Class-S sensitivity from 8.89% to 1.39%; these results indicate mixed class-level effects.

Multimodal emotion preference learning uses pairwise comparisons to identify which open-ended emotion descriptions align more closely with human judgments. However, existing methods primarily focus on the final preference decision and provide little explicit supervision for localizing and correcting erroneous content. To address this limitation, we propose EmoGavel, a Grounding-Aware reinforcement learning framework with Verifiable rewards for multimodal Emotion preference Learning. We first introduce grounded supervised fine-tuning (Grounded SFT), which exploits controlled-error construction information to jointly learn preference prediction, error localization, repair, and missing-evidence recovery. We then optimize these verifiable outcomes with grounding-aware group relative policy optimization (GA-GRPO), combining preference-gated grounding rewards with dynamic zero-variance rescue to prioritize preference correctness and recover learning signals from all-wrong groups. Experiments on EmoPrefer and controlled-error evaluation sets show that EmoGavel improves both natural preference prediction and controlled-error discrimination while maintaining strong error-grounding performance. Our code and model checkpoints will be released once the paper is accepted.

Large-scale renewable energy profile clustering is a computationally challenging NP-hard problem. Recent quantum computing techniques offer a potential remedy, yet their scalability is severely constrained by the limited qubit count of current quantum hardware. To address this issue, we first cast the problem into a QUBO form and then we propose a consensus-based variable reduction (CBVR) method to drastically cut qubit usage. More specifically, we tailor the one-hot constraint to eliminate assignment variables and leverage consensus analysis to convert stable clustering relations from initial anneals into equality constraints for further reduction. Experiments on real quantum hardware confirm that CBVR compresses oversized QUBO models into feasible sizes while retaining high clustering quality and efficiency.

In this paper, we propose a signal processing concept, signal processing in networks (SPIN), that executes signal processing tasks directly on programmable switches. Traditionally, signal processing tasks, such as denoising, decomposition, and dimensionality reduction, are performed on edge or cloud servers, i.e., raw streams are transmitted from sensors. This imposes a heavy processing load on edge/cloud servers and requires substantial bandwidth. SPIN addresses this problem by performing signal processing on the forwarding path. As an initial realization of SPIN, we implement a DCT-based block transform on a programmable switch. We redesign its computation using a lifting-based lattice structure, multiplierless approximation, and pipeline partitioning across stages in a programmable switch. We verify at the compiler level that the proposed implementation satisfies the strict hardware constraints of the Intel Tofino2–specifically regarding pipeline stages, packet header vector, and ternary content-addressable memory usage–and therefore enables line-rate processing. We further quantify the per-packet latency of the ingress match-action pipeline. The SPIN-oriented preprocessing adds 46 cycles (approximately 31 ns at 1.5 GHz). The total ingress match-action unit latency is approximately 87 ns, which remains consistent with the low-latency operation targeted by SPIN. Furthermore, we implement the proposed block transform on a BMv2 switch using P4 and evaluate it as preprocessing for in-network machine learning. Experimental results demonstrate that the proposed method improves classification performance over unfiltered baselines.

This paper considers channel state information (CSI) prediction using neural networks for time-varying multi-antenna channels from noisy historical observations. The main challenge is to exploit temporal correlation while distinguishing channel evolution from observation noise. We propose JEPA-kMMVAE, which combines a joint-embedding predictive architecture with a k-memory Markov variational autoencoder to predict future CSI through temporally structured latent representations. The framework supports direct prediction and residual corrections relative to the latest latent state and noisy CSI observation. Simulation results show that the residual configuration outperforms the evaluated baselines at moderate and high input CSI signal-to-noise ratios, while sequence baselines remain more accurate under stronger noise.

Speech anonymization is evaluated with automatic speaker verification attackers whose performance depends on the attribute distribution of both their training and evaluation data, which may bias privacy estimates for some speaker groups. To alleviate this issue, we introduce VoxAnon, a dataset built from VoxCeleb1 and 2 with training and evaluation splits balanced across gender, age, accent, and pitch. We extract and cross-check attributes using two complementary methods, balance trial lists per subgroup, and draw test/enrollment utterances from distinct video sources in order to limit linguistic and session bias. Comparing VoxAnon with unbalanced training and test data across four attackers and three anonymizers, we show that attribute imbalance changes which subgroups appear hardest to re-identify, in directions that depend on the attribute. In the balanced setting, residual disparities persist, mostly tied to accent and pitch, and depend on the anonymizer/attacker pair. Code and trial lists are publicly available.

This paper describes our systems for the Track 1 time-stamped speaker-attributed ASR task of the SmartGlasses Challenge. To handle overlapping speech, background noise and far-field conditions in two-person conversations, we employ SoulX-Transcriber, a multimodal large language model that takes full-length audio as input and generates structured transcriptions. In addition, we investigate the effect of speech frontend processing, including a rule-based Guided Source Separation method and a data-driven speech separation model, both of which improve performance on the validation set. Considering inference stability, our final submission uses the fine-tuned SoulX-Transcriber system, achieving a tcpCER of 6.22 on the test set.

Neuronal excitability is commonly modeled using different principles at the cellular and population scales, making it difficult to study how microscopic mechanisms shape macroscopic activity. We introduce a spatiotemporal excitatory-inhibitory (EI) model that unifies these scales through a common mixed-feedback architecture. The model retains the conductance-based circuit structure of Hodgkin-Huxley models and describes intrinsic and synaptic currents through memductances whose temporal and spatial voltage dependence encodes dynamics at the relevant scale. In its minimal form, temporal excitability arises from fast excitatory feedback opposed by slower inhibitory feedback, while spatial excitability arises from short-range excitation opposed by longer-range inhibition, recovering the core lateral-inhibition motif of Amari neural fields. We show that combining these mechanisms gives rise to population excitability in strongly heterogeneous populations. In particular, the model generates population spikes that are localized in both space and time and possess population-level scales distinct from those of individual neuronal spikes, despite substantial cell-to-cell parameter variability. We then show that both intrinsic temporal excitability and spatial synaptic feedback are required for coherent mean-field events and robust spatiotemporal population rhythms. Finally, selective modulation of memductances in a subset of the population tunes the frequency of the population rhythm, illustrating how changes at the cellular scale can shape macroscopic dynamics. By separating circuit architecture from modeling scale, the proposed framework provides a physically interpretable approach to analyzing and designing neuronal excitability across scales.

Open audio-language models are used as judges of generated music: given a clip and its text prompt, the model rates the clip, and the judge is validated by how well its ratings correlate with expert ratings. The correlation does not show what the judge responds to: the music, the text prompt, or the way the question is asked. We propose four probes that test each of these, three of them read against a control computed from the human ratings: removing the audio, replacing the clip with another model's clip for the same prompt, swapping the order of two clips in pairwise judging, and rewording the instruction. We apply them to five open judges on all 2,748 clips of MusicEval, and the last two to the SongEval songs. Without audio, no judge exceeds what the prompt alone predicts from the human ratings, and for prompt alignment every judge falls below it. With audio, every judge follows the clip it hears when rating prompt alignment, two do when rating musical quality, and the strongest judge's quality score tracks the experts' alignment rating more than their quality rating. The pairwise verdicts of the four smaller judges follow the position of a clip in the pair (its slot) rather than its quality on both rating sets. For musical quality, rewording the instruction moves the correlation of most judges as much as changing the judge. MusicEval's prompt assignment bounds any prompt-only judge at the model level. We recommend reporting the four probes with every music judge.

Joint-embedding predictive architectures (JEPAs) learn by predicting teacher representations of masked regions from visible context. Point prediction summarizes the conditional target distribution and cannot reproduce variation that the context does not resolve. We call this constraint predictability bias. In computed tomography (CT), recurrent anatomy is predictable from context, whereas the sporadic, localized pathology that determines clinical interpretation often is not. We introduce TAP-JEPA, a ViT-B/3D trained on 113,611 CT volumes that is competitive with state-of-the-art encoders (COLIPRI, Merlin, VoxelFM) in scan-level classification. We then audit the learned predictive pathway by replaying the exact masking, teacher, and predictor computations used during training. Prediction lowers the mean Skill (prevalence-normalized average precision) from 0.388 to 0.271 for pathology, but only from 0.921 to 0.873 for anatomy. Along the teacher direction, separating pathology from matched control regions in the same organ, prediction moves pathology 34.7% toward the controls and controls 13.6% toward pathology, with larger shifts farther from visible context. Replacing visible matched-control representations increases surrounding-target prediction loss more than fully replacing pathology representations across four pathology categories. Downstream, removing all final-layer pathology tokens reduces the decision margin by at most 0.45% in four of seven scan-level probes. Together, these results show that contextual predictability is not a sufficient proxy for clinical relevance in medical representation learning. Training objectives should therefore retain diagnostic features regardless of their predictability from surrounding anatomy.

We study query-conditioned multi-instance open-vocabulary sound event localization (QC-MI-OVSEL), where a system must recover all active physical sources matching a natural-language category query, including when the queried category is absent. We introduce the QC-MI-OVSEL benchmark, a 40,000-scene first-order Ambisonics benchmark with 80,000 balanced present/absent query views over 83 classes, including 17 novel classes excluded from training and development. Each frame is annotated with the activity and 3-D direction of every query-matching source, with up to three simultaneous targets. We further propose Q-RSet, which uses query-conditioned source slots and responsibility-based evidence aggregation for source-set prediction. Experiments show improved source recovery, localization recall, cardinality estimation, exact-set recovery, and absent-query suppression over complementary baselines. Analysis indicates that novel categories, paraphrased queries, and higher source multiplicity mainly impair source discovery and set completion, while directional accuracy remains comparatively stable for successfully recovered sources.

Recent advances in speech synthesis technology have enabled the generation of speech with diverse characteristics. One such characteristic is a foreign accent. In our preliminary experiments, we found that UTMOS, a widely used automatic speech quality assessment system, tended to assign lower scores to speech with stronger foreign accents, even though it is produced by a human speaker. This observation raised a question: How do humans evaluate the naturalness of foreign-accented speech? We conducted a large-scale subjective evaluation involving both native and non-native listeners under different evaluation instructions. The results showed that humans also considered accentedness when evaluating naturalness, but that defining naturalness in terms of human-likeness and selecting raters who have greater exposure to the target accent may enable more appropriate evaluation. This study aims to provide a starting point for future discussions on fairer and more appropriate evaluation of synthetic speech with a foreign accent.

We compare the Nordic Grid Development Perspective 2025 (NGDP) with a least-cost transition pathway for the Nordic power system obtained with PyPSA-Eur for the period 2025-2050 under a net-zero CO$_2$ constraint. The model co-optimises generation, storage, and transmission expansion with hourly system operation while constraining the Nordic generation mix to remain consistent with NGDP milestone ranges. Overall, the optimised capacity pathways reproduce the broad NGDP expansion of wind and solar power, with Denmark emerging as the main locus of offshore wind and solar deployment, Finland as a major site for onshore wind expansion, and Norway and Sweden relying more strongly on existing hydropower and nuclear capacity. The main discrepancy between the two pathways does not lie in the broad build-out of variable renewables, but in the timing and role of hydrogen. Whereas the NGDP treats hydrogen as an early and steadily increasing component of industrial demand and system development already from 2030, PyPSA-Eur delays substantial hydrogen deployment until 2045-2050, when it becomes cost-effective primarily as a flexibility and seasonal storage option in a strongly variable-renewable-energy-dominated system. This contrast suggests that early hydrogen build-out in the Nordics is not a robust outcome of cost optimisation alone, but depends strongly on assumptions regarding industrial demand growth, sector coupling, and anticipatory infrastructure planning. We therefore argue that hydrogen remains the principal strategic uncertainty in the Nordic transition, with the NGDP representing a more anticipatory planning perspective and PyPSA-Eur a more conservative least-cost timing of deployment.

Belief propagation (BP) is widely used for inference on probabilistic graphical models but often suffers from non-convergence and poor accuracy on loopy factor graphs, yielding inexact variance estimates in the Gaussian case. Generalized BP (GBP) improves accuracy but relies on manual design or application-specific heuristics for region-graph construction. We propose extended neighborhood message passing (ENMP), which automatically constructs extended neighborhoods of variable nodes from the bipartite factor graph topology, capturing loop-induced correlations without manual region design. The framework applies to general distributions on factor graphs with a tunable accuracy-complexity trade-off. In the exact regime, message updates converge to the unique fixed point of the ENMP equations in finitely many iterations, yielding exact marginals; in the approximate regime, ENMP provides approximate marginals. For Gaussian factor graphs, we derive closed-form update rules enabling exact variance estimation in the exact regime, addressing the long-standing inaccurate variance estimates of loopy Gaussian BP. We evaluate ENMP on densely loopy Gaussian factor graphs from power-grid benchmarks and on a discrete Ising factor graph. Results show accuracy improves with neighborhood size, with exact marginals achieved on several systems in both settings. On larger Gaussian systems, ENMP outperforms regularized Gaussian BP (sGaBP-m) in both mean and variance estimation accuracy.