Fusion

arXiv:2603.20575v1 Announce Type: cross Abstract: As near-Earth resident space objects proliferate, there is an increasing demand for reliable technologies in applications of on-orbit servicing, debris removal, and orbit modification. Rendezvous and docking are critical mission phases for such applications and can benefit from greater autonomy to reduce operational complexity and human workload. Machine learning-based methods can be integrated within the guidance, navigation, and control (GNC) architecture to design a robust rendezvous and docking framework. In this work, the Multi-Agent Multi-View Experimental and Digital Twin Rendezvous (MMEDR-Autonomous) is introduced as a unified framework comprising a learning-based optical navigation network, a reinforcement learning-based guidance approach under ongoing development, and a hardware-in-the-loop testbed. Navigation employs a lightweight monocular pose estimation network with multi-scale feature fusion, trained on realistic image

arXiv:2603.22209v1 Announce Type: new Abstract: We use the theoretical framework introduced in the companion paper to provide simple formulas as regards the resonance conditions for CBET with smoothed laser beams.Our analytical CBET model with optical smoothing shows that these fusion-critical lasers produce a significantly broader resonance than conventional plane wave models predict. In particular, temporal smoothing, as used in many high energy laser facilities, and flow components normal to the CBET ion acoustic waves, significantly modify the power transfer between smoothed beams. Our model predicts that the energy transfer rate out of resonance is substantially higher with optical smoothing than without, a result that has profound implications for optimizing predicting and interpreting future fusion experiments. We provide a simple criterion which pinpoints the laser and plasma parameters for which laser smoothing impacts CBET. These findings pave the way for experimental

arXiv:2603.22180v1 Announce Type: new Abstract: In the context of inertial confinement fusion (ICF) experiments, spatial and temporal laser beam smoothing techniques are used to control the beams propagation in hohlraum plasmas. Currently, spatial and temporal smoothing are either neglected or not properly taken into account in the inline cross beam energy transfer (CBET) models included in the hydrodynamic codes dedicated to the design of these experiments. In some cases, which we will highlight in this study, this simplification leads to important errors in the power transfer of importance for the implosion symmetry of the capsule, either in the direct or indirect drive ICF configurations. In a recent study [A. Oudin et \textit{al}., Phys. Plasmas \textbf{32}, 042706 (2025)], we demonstrated the necessity of accounting for spatial smoothing when modeling CBET, provided that the beams do not have the same wavelength. This work presents a linear kinetic model compared with Hera

arXiv:2603.20485v1 Announce Type: new Abstract: A disrupting plasma in a high-performance tokamak such as ITER or SPARC may generate large runaway electron currents that, upon impact with the tokamak wall, can cause serious damage to the device. To quickly identify regions of safe operation in parameter space, it is useful to develop reduced models and analytical criteria that predict when a significant fraction of the Ohmic current is converted into a current of runaway electrons. In deuterium-tritium plasmas, the seed runaway current may have a significant contribution from - or may even be dominated by - tritium beta decay and Compton scattering. In this work, a criterion for significant runaway electron generation that includes tritium beta decay and Compton scattering sources is developed. The avalanche gain factor includes the effects of partial screening of injected noble gases. The result is an analytical model that can predict significant runaway electron generation in the

arXiv:2603.21820v1 Announce Type: new Abstract: Infrared and visible image fusion(IVIF) combines complementary modalities while preserving natural textures and salient thermal signatures. Existing solutions predominantly rely on extensive sets of rigidly aligned image pairs for training. However, acquiring such data is often impractical due to the costly and labour-intensive alignment process. Besides, maintaining a rigid pairing setting during training restricts the volume of cross-modal relationships, thereby limiting generalisation performance. To this end, this work challenges the necessity of Strictly Paired Training Paradigm (SPTP) by systematically investigating UnPaired and Arbitrarily Paired Training Paradigms (UPTP and APTP) for high-performance IVIF. We establish a theoretical objective of APTP, reflecting the complementary nature between UPTP and SPTP. More importantly, we develop a practical framework capable of significantly enriching cross-modal relationships even with

arXiv:2603.21768v1 Announce Type: new Abstract: Precipitation nowcasting is critical for disaster mitigation and aviation safety. However, radar-only models frequently suffer from a lack of large-scale atmospheric context, leading to performance degradation at longer lead times. While integrating meteorological variables predicted by weather foundation models offers a potential remedy, existing architectures fail to reconcile the profound representational heterogeneities between radar imagery and meteorological data. To bridge this gap, we propose PW-FouCast, a novel frequency-domain fusion framework that leverages Pangu-Weather forecasts as spectral priors within a Fourier-based backbone. Our architecture introduces three key innovations: (i) Pangu-Weather-guided Frequency Modulation to align spectral magnitudes and phases with meteorological priors; (ii) Frequency Memory to correct phase discrepancies and preserve temporal evolution; and (iii) Inverted Frequency Attention to

arXiv:2603.21160v1 Announce Type: new Abstract: Epistemic intelligence requires machine learning systems to recognise the limits of their own knowledge and act safely under uncertainty, especially when faced with unknown unknowns. Existing uncertainty quantification methods rely on a single signal such as confidence or density and fail to detect diverse structural anomalies. We introduce SPECTRE-G2, a multi-signal anomaly detector that combines eight complementary signals from a dual-backbone neural network. The architecture includes a spectral normalised Gaussianization encoder, a plain MLP preserving feature geometry, and an ensemble of five models. These produce density, geometry, uncertainty, discriminative, and causal signals. Each signal is normalised using validation statistics and calibrated with synthetic out-of-distribution data. An adaptive top-k fusion selects the most informative signals and averages their scores. Experiments on synthetic, Adult, CIFAR-10, and Gridworld

arXiv:2603.20920v1 Announce Type: new Abstract: The exponential growth in data has intensified the demand for computational power to train large-scale deep learning models. However, the rapid growth in model size and complexity raises concerns about equal and fair access to computational resources, particularly under increasing energy and infrastructure constraints. GPUs have emerged as essential for accelerating such workloads. This study benchmarks four deep learning models (Conv6, VGG16, ResNet18, CycleGAN) using TensorFlow and PyTorch on Intel Xeon CPUs and NVIDIA Tesla T4 GPUs. Our experiments demonstrate that, on average, GPU training achieves speedups ranging from 11x to 246x depending on model complexity, with lightweight models (Conv6) showing the highest acceleration (246x), mid-sized models (VGG16, ResNet18) achieving 51-116x speedups, and complex generative models (CycleGAN) reaching 11x improvements compared to CPU training. Additionally, in our PyTorch vs. TensorFlow

arXiv:2603.20575v1 Announce Type: new Abstract: As near-Earth resident space objects proliferate, there is an increasing demand for reliable technologies in applications of on-orbit servicing, debris removal, and orbit modification. Rendezvous and docking are critical mission phases for such applications and can benefit from greater autonomy to reduce operational complexity and human workload. Machine learning-based methods can be integrated within the guidance, navigation, and control (GNC) architecture to design a robust rendezvous and docking framework. In this work, the Multi-Agent Multi-View Experimental and Digital Twin Rendezvous (MMEDR-Autonomous) is introduced as a unified framework comprising a learning-based optical navigation network, a reinforcement learning-based guidance approach under ongoing development, and a hardware-in-the-loop testbed. Navigation employs a lightweight monocular pose estimation network with multi-scale feature fusion, trained on realistic image

arXiv:2603.20488v1 Announce Type: new Abstract: Electricity theft, or non-technical loss (NTL), presents a persistent threat to global power systems, driving significant financial deficits and compromising grid stability. Conventional detection methodologies, predominantly reactive and meter-centric, often fail to capture the complex spatio-temporal dynamics and behavioral patterns associated with fraudulent consumption. This study introduces a novel AI-driven Grid Intelligence Framework that fuses Time-Series Anomaly Detection, Supervised Machine Learning, and Graph Neural Networks (GNN) to identify theft with high precision in imbalanced datasets. Leveraging an enriched feature set, including rolling averages, voltage drop estimates, and a critical Grid Imbalance Index, the methodology employs a Long Short-Term Memory (LSTM) autoencoder for temporal anomaly scoring, a Random Forest classifier for tabular feature discrimination, and a GNN to model spatial dependencies across the

arXiv:2603.20236v1 Announce Type: new Abstract: Recent advances in unimanual manipulation policies have achieved remarkable success across diverse robotic tasks through abundant training data and well-established model architectures. However, extending these capabilities to bimanual manipulation remains challenging due to the lack of bimanual demonstration data and the complexity of coordinating dual-arm actions. Existing approaches either rely on extensive bimanual datasets or fail to effectively leverage pre-trained unimanual policies. To address this limitation, we propose \textbf{EnergyAction}, a novel framework that compositionally transfers unimanual manipulation policies to bimanual tasks through the Energy-Based Models (EBMs). Specifically, our method incorporates three key innovations. First, we model individual unimanual policies as EBMs and leverage their compositional properties to compose left and right arm actions, enabling the fusion of unimanual policies into a

arXiv:2603.20232v1 Announce Type: new Abstract: Autonomous driving testing increasingly relies on mining safety critical scenarios from large scale naturalistic driving data, yet existing screening pipelines still depend on manual risk annotation and expensive frame by frame risk evaluation, resulting in low efficiency and weakly grounded risk quantification. To address this issue, we propose a driver risk fusion based hazardous scenario screening method for autonomous driving. During training, the method combines an improved Driver Risk Field with a dynamic cost model to generate high quality risk supervision signals, while during inference it directly predicts scenario level risk scores through fast forward passes, avoiding per frame risk computation and enabling efficient large scale ranking and retrieval. The improved Driver Risk Field introduces a new risk height function and a speed adaptive look ahead mechanism, and the dynamic cost model integrates kinetic energy, oriented

arXiv:2603.20125v1 Announce Type: new Abstract: Quasi-axisymmetric stellarators are the stellarator analogue of the axisymmetric tokamak, retaining many of its favorable confinement properties, its compacity and its relative coil simplicity, while avoiding its principal limitation, the need for an inductively driven plasma current. Despite these attractive physics properties, the development of quasi-axisymmetric configurations has been severely constrained by the absence of an experimentally validated divertor concept compatible with their large bootstrap current. In this Letter, approximately quasi-axisymmetric fields, complemented with piecewise omnigenous perturbations, are proposed as the basis for a new strategy towards a stellarator reactor that simultaneously achieves simple coil geometries, tokamak-like confinement properties and, through tailoring of the bootstrap current, compatibility with an island divertor. Implications for attaining a high bootstrap current fraction in

arXiv:2603.19606v1 Announce Type: new Abstract: Existing paradigms for remote sensing change detection are caught in a trade-off: CNNs excel at efficiency but lack global context, while Transformers capture long-range dependencies at a prohibitive computational cost. This paper introduces ChangeRWKV, a new architecture that reconciles this conflict. By building upon the Receptance Weighted Key Value (RWKV) framework, our ChangeRWKV uniquely combines the parallelizable training of Transformers with the linear-time inference of RNNs. Our approach core features two key innovations: a hierarchical RWKV encoder that builds multi-resolution feature representation, and a novel Spatial-Temporal Fusion Module (STFM) engineered to resolve spatial misalignments across scales while distilling fine-grained temporal discrepancies. ChangeRWKV not only achieves state-of-the-art performance on the LEVIR-CD benchmark, with an 85.46% IoU and 92.16% F1 score, but does so while drastically reducing

arXiv:2603.19596v1 Announce Type: new Abstract: The integration of Large Language Models (LLMs) and Graph Neural Networks (GNNs) promises to unify semantic understanding with structural reasoning, yet existing methods typically rely on static, unidirectional pipelines. These approaches suffer from fundamental limitations: (1) Bidirectional Error Propagation, where semantic hallucinations in LLMs or structural noise in GNNs permanently poison the downstream modality without opportunity for recourse; (2) Semantic-Structural Dissonance, particularly in heterophilous settings where textual similarity contradicts topological reality; (3) a Blind Leading the Blind phenomenon, where indiscriminate alignment forces models to mirror each other's mistakes regardless of uncertainty. To address these challenges, we propose CO-EVOLVE, a dual-view co-evolution framework that treats graph topology and semantic embeddings as dynamic, mutually reinforcing latent variables. By employing a Gauss-Seidel

arXiv:2603.19504v1 Announce Type: new Abstract: More scientists are now using AI, but prior studies have examined only how they use it 'at the desk' for computer-based work. However, given that scientific work often happens 'beyond the desk' at lab and field sites, we conducted the first study of how scientific practitioners use AI for embodied physical tasks. We interviewed 12 scientific practitioners doing hands-on lab and fieldwork in domains like nuclear fusion, primate cognition, and biochemistry, and found three barriers to AI adoption in these settings: 1) experimental setups are too high-stakes to risk AI errors, 2) constrained environments make it hard to use AI, and 3) AI cannot match the tacit knowledge of humans. Participants then developed speculative designs for future AI assistants to 1) monitor task status, 2) organize lab-wide knowledge, 3) monitor scientists' health, 4) do field scouting, 5) do hands-on chores. Our findings point toward AI as background

arXiv:2603.18267v1 Announce Type: new Abstract: Resonant drive in tokamaks is routinely quantified using a variety of different metrics that target different aspects of a resonant response to an external perturbation. Two of the most direct metrics, $\Delta_{mn}$ and $b_{pen}$, are widely used but their relative behavior was previously uncharacterized. This work examines how these metrics representing the shielding current and penetrated field relate in resistive perturbed tokamak equilibria using asymptotically matched solutions with a resistive MHD inner layer model in GPEC. $b_{pen}$ scales with Lundquist number as $S^{-2/3}$ until saturation at low $S$, and $\Delta_{mn}$ remains consistent with its ideal definition but is affected by global kink structure. Both metrics are shown to yield closely similar dominant coupling modes within the same resistive model. However, the resistive physics shifts this dominant mode spectrum to lower poloidal mode numbers $m$ in a low-rotation ITER

arXiv:2603.18265v1 Announce Type: new Abstract: The non-resonant divertor (NRD) offers a promising exhaust solution for stellarators, combining topological simplicity with resilience to magnetic field perturbations. To experimentally validate the robustness of non-resonant divertors in a quasi-axisymmetric (QA) configuration, we introduce STAR_Lite, a new stellarator experiment at Hampton University. This paper details the design and analysis of the first STAR_Lite coil configuration, STAR_Lite-A. The two field-period configuration manifests an NRD through X-points with zero rotational transform, at the top and bottom of the device. The divertor legs extruding from the X-points are topologically similar to the poloidal divertors of tokamaks. To expand the experimental range, STAR_Lite-A is optimized for experimental flexibility, producing a wide range of distinct QA configurations by only varying the currents in the modular coils. The NRDs not only persist across these configurations,

arXiv:2603.18871v1 Announce Type: new Abstract: Vehicular Ad-hoc Networks (VANETs) are the digital cornerstone of autonomous driving, yet they suffer from severe network fragmentation in urban environments due to physical obstructions. Unmanned Aerial Vehicles (UAVs), with their high mobility, have emerged as a vital solution to bridge these connectivity gaps. However, traditional Deep Reinforcement Learning (DRL)-based UAV deployment strategies lack semantic understanding of road topology, often resulting in blind exploration and sample inefficiency. By contrast, Large Language Models (LLMs) possess powerful reasoning capabilities capable of identifying topological importance, though applying them to control tasks remains challenging. To address this, we propose the Semantic-Augmented DRL (SA-DRL) framework. Firstly, we propose a fragmentation quantification method based on Road Topology Graphs (RTG) and Dual Connected Graphs (DCG). Subsequently, we design a four-stage pipeline to

arXiv:2603.18712v1 Announce Type: new Abstract: The task of multi-channel time series forecasting is ubiquitous in numerous fields such as finance, supply chain management, and energy planning. It is critical to effectively capture complex dynamic dependencies within and between channels for accurate predictions. However, traditional method paid few attentions on learning the interaction among channels. This paper proposes Linear-Network (Li-Net), a novel architecture designed for multi-channel time series forecasting that captures the linear and non-linear dependencies among channels. Li-Net dynamically compresses representations across sequence and channel dimensions, processes the information through a configurable non-linear module and subsequently reconstructs the forecasts. Moreover, Li-Net integrates a sparse Top-K Softmax attention mechanism within a multi-scale projection framework to address these challenges. A core innovation is its ability to seamlessly incorporate and

arXiv:2603.18344v1 Announce Type: new Abstract: Maintaining situational awareness (SA) is critical in human-robot teams. Yet, under high workload and dynamic conditions, operators often experience SA gaps. Automated detection of SA gaps could provide timely assistance for operators. However, conventional SA measures either disrupt task flow or cannot capture real-time fluctuations, limiting their operational utility. To the best of our knowledge, no publicly available dataset currently supports the systematic evaluation of online human SA assessment in human-robot teaming. To advance the development of online SA assessment tools, we introduce HRI-SA, a multimodal dataset from 30 participants in a realistic search-and-rescue human-robot teaming context, incorporating eye movements, pupil diameter, biosignals, user interactions, and robot data. The experimental protocol included predefined events requiring timely operator assistance, with ground truth SA latency of two types (perceptual

arXiv:2603.18062v1 Announce Type: new Abstract: Skeleton-based action recognition is crucial for multimedia applications but heavily relies on power-hungry Artificial Neural Networks (ANNs), limiting their deployment on resource-constrained edge devices. Spiking Neural Networks (SNNs) provide an energy-efficient alternative; however, existing spiking models for skeleton data often compromise the intrinsic sparsity of SNNs by resorting to dense matrix aggregations, heavy multimodal fusion modules, or non-sparse frequency domain transformations. Furthermore, they severely suffer from the short-term amnesia of spiking neurons. In this paper, we propose the Spiking State-Space Topology Transformer (S3T-Former), which, to the best of our knowledge, is the first purely spike-driven Transformer architecture specifically designed for energy-efficient skeleton action recognition. Rather than relying on heavy fusion overhead, we formulate a Multi-Stream Anatomical Spiking Embedding (M-ASE) that

arXiv:2603.17493v1 Announce Type: new Abstract: Dusty plasmas are ubiquitous throughout the universe, spanning laboratory and industrial plasmas, fusion devices, planetary environments, cometary comae, and interstellar media. Despite decades of research, many aspects of their behavior remain poorly understood within a unified framework. While numerous theoretical and numerical models describe specific phenomena, such as dust charging, transport, waves, and self-organization, fully predictive models across the wide range of spatial and temporal scales in both laboratory and natural systems remain elusive. Conventional plasma descriptions rely on coupled differential equations for particle densities, momenta, and energies, but their solutions are often limited by computational cost, numerical uncertainties, and incomplete knowledge of boundary conditions and transport processes. Recent advances in machine learning (ML), particularly deep neural networks, offer new opportunities to

arXiv:2603.17366v1 Announce Type: new Abstract: The optimized stellarator is an attractive concept for which the averaged particle radial drift is zero, and the single particle loss can be significantly reduced. But for the reactor design, global physics such as turbulent transport also need to be optimized besides the confined single particle orbit, or properties estimated using local estimations and heuristic formulations. The first-principle global transport code is too computationally expensive to integrate into the optimization process. The fast surrogate global transport model based on machine learning is a good alternative choice, but the amount of data required to train the surrogate model is numerous due to the high degree-of-freedom of the stellarator design. The work shows that the stellarator design with quasi-helically(QH) symmetric geometry is approximately distributed in a low dimensional latent space, which can be explicitly found by deep learning. This discovery makes

arXiv:2603.17069v1 Announce Type: new Abstract: Falls in wet bathroom environments are a major safety risk for seniors living alone. Recent work has shown that mmWave-only, vibration-only, and existing multimodal schemes, such as vibration-triggered radar activation, early feature concatenation, and decision-level score fusion, can support privacy-preserving, non-intrusive fall detection. However, these designs still treat motion and impact as loosely coupled streams, depending on coarse temporal alignment and amplitude thresholds, and do not explicitly encode the causal link between radar-observed collapse and floor impact or address timing drift, object drop confounders, and latency and energy constraints on low-power edge devices. To this end, we propose a two-stream architecture that encodes radar signals with a Motion--Mamba branch for long-range motion patterns and processes floor vibration with an Impact--Griffin branch that emphasizes impact transients and cross-axis coupling.

Brown dwarfs get a bad rap in the stellar world, often labeled as "failed stars" for their inability to sustain nuclear fusion at their cores. The mass of these objects falls between planets and stars, ranging from 13 to 80 times the mass of Jupiter. Because they aren't massive enough to sustain fusion, they are far fainter and cooler than their stellar comrades.

For decades, astronomers have struggled to differentiate giant planets from brown dwarfs, a class of objects more massive than planets but too small to ignite nuclear fusion like true stars. Through a telescope, these cosmic lookalikes can have overlapping brightness, temperatures, and even atmospheric fingerprints. The striking similarity leaves astronomers unsure if they have observed an oversized planet or an undersized star. Now, a Northwestern University-led team has uncovered a crucial clue that separates the two: how fast they spin.

arXiv:2603.16178v1 Announce Type: new Abstract: Early identification of individuals at risk of stroke remains a major clinical challenge, as prodromal motor im- pairments are often subtle and transient. In this pilot study, a wearable sensor-based framework is proposed for early pre- stroke risk screening using a single inertial measurement unit mounted on the sacral region to capture pelvic motion during gait and standing tasks. The pelvis is treated as a biomechanical proxy for global motor control, enabling the quantification of gait variability and postural drift as digital biomarkers of neurological instability. Raw inertial signals are processed using a sensor fusion pipeline to estimate pelvic kinematics, from which variability and nonlinear dynamic features are extracted. These features are subsequently used to train a machine learning model for risk stratification across control, pre-stroke, and stroke groups. Progressive increases in pelvic angular variability and postural

arXiv:2603.15939v1 Announce Type: new Abstract: Applying machine learning to sensitive time-series data is often bottlenecked by the iteration loop: Performance depends strongly on preprocessing and architecture, yet training often has to run on-premise under strict data-local constraints. This is a common problem in healthcare and other privacy-constrained domains (e.g., a hospital developing deep learning models on patient EEG). This bottleneck is particularly challenging in multimodal fusion, where sensor modalities must be individually preprocessed and then combined. LLM-guided neural architecture search (NAS) can automate this exploration, but most existing workflows assume cloud execution or access to data-derived artifacts that cannot be exposed. We present a novel data-local, LLM-guided search framework that handles candidate pipelines remotely while executing all training and evaluation locally under a fixed protocol. The controller observes only trial-level summaries, such

arXiv:2603.15920v1 Announce Type: new Abstract: Differentiable programming has emerged as a structural prerequisite for gradient-based inverse problems and end-to-end hybrid physics--machine learning in computational fluid dynamics. However, existing differentiable CFD platforms are confined to structured Cartesian grids, excluding the geometrically complex domains where body-conforming unstructured discretizations are indispensable. We present DiFVM, the first GPU-accelerated, end-to-end differentiable finite-volume CFD solver operating natively on unstructured polyhedral meshes. The key enabling insight is a structural isomorphism between finite-volume discretization and graph neural network message-passing: by reformulating all FVM operators as static scatter/gather primitives on the mesh connectivity graph, DiFVM transforms irregular unstructured connectivity into a first-class GPU data structure. All operations are implemented in JAX/XLA, providing just-in-time compilation,

arXiv:2603.15708v1 Announce Type: new Abstract: Imbalanced data distribution remains a critical challenge in sequential learning, leading models to easily recognize frequent categories while failing to detect minority classes adequately. The Mixture-of-Experts model offers a scalable solution, yet its application is often hindered by parameter inefficiency, poor expert specialization, and difficulty in resolving prediction conflicts. To Master the Minority classes effectively, we propose the Uncertainty-based Multi-Expert fusion network (UME) framework. UME is designed with three core innovations: First, we employ Ensemble LoRA for parameter-efficient modeling, significantly reducing the trainable parameter count. Second, we introduce Sequential Specialization guided by Dempster-Shafer Theory (DST), which ensures effective specialization on the challenging-tailed classes. Finally, an Uncertainty-Guided Fusion mechanism uses DST's certainty measures to dynamically weigh expert

arXiv:2603.13681v1 Announce Type: new Abstract: Structural assumptions are central to the causal inference literature. In practice, it is often crucial to assess their validity or to test implications that follow from them. In many settings, such tests can be framed as evaluating whether a function-valued parameter equals zero. In this paper, we propose a class of generalized projection tests based on series estimators for function-valued parameters. We establish conditions under which the proposed tests are valid and illustrate their applicability through examples from the data fusion and instrumental variables literature. Our approach accommodates flexible machine learning methods for estimating nuisance parameters. In contrast to many existing approaches, the limiting distribution of the proposed test statistics is straightforward to compute under the null hypothesis. We apply our method to test the equality of conditional COVID-19 risk across vaccine arms in the COVID-19 Variant

arXiv:2603.13681v1 Announce Type: cross Abstract: Structural assumptions are central to the causal inference literature. In practice, it is often crucial to assess their validity or to test implications that follow from them. In many settings, such tests can be framed as evaluating whether a function-valued parameter equals zero. In this paper, we propose a class of generalized projection tests based on series estimators for function-valued parameters. We establish conditions under which the proposed tests are valid and illustrate their applicability through examples from the data fusion and instrumental variables literature. Our approach accommodates flexible machine learning methods for estimating nuisance parameters. In contrast to many existing approaches, the limiting distribution of the proposed test statistics is straightforward to compute under the null hypothesis. We apply our method to test the equality of conditional COVID-19 risk across vaccine arms in the COVID-19 Variant

arXiv:2603.15244v1 Announce Type: cross Abstract: This work was conducted within the framework of the exploratory French project PEPR SupraFusion, which aims to advance the field of fusion energy by developing High-Temperature Superconductor (HTS)-based demonstrators capable of storing significant energy while operating under high magnetic fields and currents. Ensuring a reliable protection during a quench in Insulated REBCO conductors is challenging\,: slow normal-zone propagation and validation delays allow the hotspot's temperature to reach damaging levels. We compare (i) conductor protection via copper-stabilizer optimization and (ii) a co-wound, REBCO superconducting quench detector (SQD) that is electrically isolated yet thermally coupled and intentionally deoxygenated to lower Tc and Ic for an earlier transition. Onedimensional THEA modeling shows that a good choice of stabilizer cross-section makes the protection possible during quench events by keeping the temperature of the

arXiv:2603.15240v1 Announce Type: new Abstract: Finding feasible coils for stellarator fusion devices is a critical challenge of realizing this concept for future power plants. Years of research work can be put into the design of even a single reactor-scale stellarator design. To rapidly speed up and automate the workflow of designing stellarator coils, we have designed an end-to-end ``runner'' for performing stellarator coil optimization. The entirety of pre and post-processing steps have been automated; the user specifies only a few basic input parameters, and final coil solutions are updated on an open-source leaderboard. Two policies are available for performing non-stop automated coil optimizations through a genetic algorithm or a context-aware LLM. Lastly, we construct a novel in-the-loop optimization of Von Mises stresses in the coils, opening up important future capabilities for in-the-loop finite-element calculations.

arXiv:2603.13838v1 Announce Type: new Abstract: High-quality proton beams generated by laser-plasma interactions are of significant interest for applications ranging from tumor therapy to fast ignition in inertial confinement fusion. However, simultaneously achieving high energy coupling efficiency and beam collimation remains a challenge. In this work, we investigate the enhancement of proton acceleration via geometric confinement in Near-Critical Density (NCD) plasma-filled micro-structured targets using two-dimensional particle-in-cell (PIC) simulations. To optimize laser-to-particle energy transfer, we systematically compared various target configurations, such as rectangular tubes, hybrid funnels, and straight cones. Our results reveals that increasing geometric complexity does not necessarily translate to superior acceleration performance. Instead, the relatively simple NCD-filled straight-cone target outperforms more complex hybrid geometries, achieving a maximum proton cutoff

arXiv:2603.15240v1 Announce Type: cross Abstract: Finding feasible coils for stellarator fusion devices is a critical challenge of realizing this concept for future power plants. Years of research work can be put into the design of even a single reactor-scale stellarator design. To rapidly speed up and automate the workflow of designing stellarator coils, we have designed an end-to-end ``runner'' for performing stellarator coil optimization. The entirety of pre and post-processing steps have been automated; the user specifies only a few basic input parameters, and final coil solutions are updated on an open-source leaderboard. Two policies are available for performing non-stop automated coil optimizations through a genetic algorithm or a context-aware LLM. Lastly, we construct a novel in-the-loop optimization of Von Mises stresses in the coils, opening up important future capabilities for in-the-loop finite-element calculations.

arXiv:2603.15281v1 Announce Type: new Abstract: Inertial navigation using low-cost MEMS sensors is plagued by rapid drift due to sensor noise and bias instability. While recent data-driven approaches have made significant strides, they often struggle with micro-drifts during stationarity and mode fusion during complex motion transitions due to their reliance on fixed-window regression. In this work, we introduce Gated Neural Inertial Odometry (GNIO), a novel learning-based framework that explicitly models motion validity and context. We propose two key architectural innovations: \ding{182} a learnable Motion Bank that queries a global dictionary of motion patterns to provide semantic context beyond the local receptive field, and \ding{183} a Gated Prediction Head that decomposes displacement into magnitude and direction. This gating mechanism acts as a soft, differentiable Zero-Velocity Update (ZUPT), dynamically suppressing sensor noise during stationary periods while scaling

arXiv:2603.15194v1 Announce Type: new Abstract: A comprehensive understanding of heat transport is essential for optimizing various mechanical and engineering applications, including 3D printing. Recent advances in machine learning, combined with physics-based models, have enabled a powerful fusion of numerical methods and data-driven algorithms. This progress is driven by the availability of limited sensor data in various engineering and scientific domains, where the cost of data collection and the inaccessibility of certain measurements are high. To this end, we present PiGRAND, a Physics-informed graph neural diffusion framework. In order to reduce the computational complexity of graph learning, an efficient graph construction procedure was developed. Our approach is inspired by the explicit Euler and implicit Crank-Nicolson methods for modeling continuous heat transport, leveraging sub-learning models to secure the accurate diffusion across graph nodes. To enhance computational

arXiv:2603.15004v1 Announce Type: new Abstract: Code clone detection (CCD) supports software maintenance, refactoring, and security analysis. Although pre-trained models capture code semantics, most work reduces CCD to binary classification, overlooking the heterogeneity of clone types and the seven fine-grained categories in BigCloneBench. We present Full Model, a multimodal fusion framework that jointly integrates heuristic similarity priors from classical machine learning, structural signals from abstract syntax trees (ASTs), and deep semantic embeddings from CodeBERT into a single predictor. By fusing structural, statistical, and semantic representations, Full Model improves discrimination among fine-grained clone types while keeping inference cost practical. On the seven-class BigCloneBench benchmark, Full Model raises Macro-F1 from 0.695 to 0.875. Ablation studies show that using the primary model's probability distribution as a prior to guide selective arbitration by a large

arXiv:2603.14956v1 Announce Type: new Abstract: Spiking Federated Learning (SFL) has been widely studied with the energy efficiency of Spiking Neural Networks (SNNs). However, existing SFL methods require model homogeneity and assume all clients have sufficient computational resources, resulting in the exclusion of some resource-constrained clients. To address the prevalent system heterogeneity in real-world scenarios, enabling heterogeneous SFL systems that allow clients to adaptively deploy models of different scales based on their local resources is crucial. To this end, we introduce SFedHIFI, a novel Spiking Federated Learning framework with Fire Rate-Based Heterogeneous Information Fusion. Specifically, SFedHIFI employs channel-wise matrix decomposition to deploy SNN models of adaptive complexity on clients with heterogeneous resources. Building on this, the proposed heterogeneous information fusion module enables cross-scale aggregation among models of different widths, thereby

arXiv:2603.14845v1 Announce Type: new Abstract: Accurate day-ahead solar irradiance forecasting is essential for integrating solar energy into the power grid. However, it remains challenging due to the pronounced diurnal cycle and inherently complex cloud dynamics. Current methods either lack fine-scale resolution (e.g., numerical weather prediction, weather foundation models) or degrade at longer lead times (e.g., satellite extrapolation). We propose Baguan-solar, a two-stage multimodal framework that fuses forecasts from Baguan, a global weather foundation model, with high-resolution geostationary satellite imagery to produce 24- hour irradiance forecasts at kilometer scale. Its decoupled two-stage design first forecasts day-night continuous intermediates (e.g., cloud cover) and then infers irradiance, while its modality fusion jointly preserves fine-scale cloud structures from satellite and large-scale constraints from Baguan forecasts. Evaluated over East Asia using CLDAS as

arXiv:2603.14803v1 Announce Type: new Abstract: Target speech extraction (TSE) aims to recover a target speaker's voice from a mixture. While recent text-prompted approaches have shown promise, most approaches assume fully overlapped mixtures, limiting insight into behavior across realistic overlap ratios. We introduce VorTEX (Various overlap ratio for Target speech EXtraction), a text-prompted TSE architecture with a Decoupled Adaptive Multi-branch (DAM) Fusion block that separates primary extraction from auxiliary regularization pathways. To enable controlled analysis, we construct PORTE, a two-speaker dataset spanning overlap ratios from 0% to 100%. We further propose Suppression Ratio on Energy (SuRE), a diagnostic metric that detects suppression behavior not captured by conventional measures. Experiments show that existing models exhibit suppression or residual interference under overlap, whereas VorTEX achieves the highest separation fidelity across 20-100% overlap (e.g., 5.50

arXiv:2603.14789v1 Announce Type: new Abstract: Achieving accurate garment grasping under dynamically changing illumination is crucial for all-day operation of service robots.However, the reduced illumination in low-light scenes severely degrades garment structural features, leading to a significant drop in grasping robustness.Existing methods typically enhance RGB features by exploiting the illumination-invariant properties of non-RGB modalities, yet they overlook the varying dependence on non-RGB features under varying lighting conditions, which can introduce misaligned non-RGB cues and thereby weaken the model's adaptability to illumination changes when utilizing multimodal information.To address this problem, we propose GraspALL, an illumination-structure interactive compensation model.The innovation of GraspALL lies in encoding continuous illumination changes into quantitative references to guide adaptive feature fusion between RGB and non-RGB modalities according to varying

arXiv:2603.14648v1 Announce Type: new Abstract: Nuclear power plant operators face significant challenges due to unpredictable deviations between offline and online thermal limits, a phenomenon known as thermal limit bias, which leads to conservative design margins, increased fuel costs, and operational inefficiencies. This work presents a deep learning based methodology to predict and correct this bias for Boiling Water Reactors (BWRs), focusing on the Maximum Fraction of Limiting Power Density (MFLPD) metric used to track the Linear Heat Generation Rate (LHGR) limit. The proposed model employs a fully convolutional encoder decoder architecture, incorporating a feature fusion network to predict corrected MFLPD values closer to online measurements. Evaluated across five independent fuel cycles, the model reduces the mean nodal array error by 74 percent, the mean absolute deviation in limiting values by 72 percent, and the maximum bias by 52 percent compared to offline methods. These

arXiv:2603.14498v1 Announce Type: new Abstract: Embodied manipulation requires accurate 3D understanding of objects and their spatial relations to plan and execute contact-rich actions. While large-scale 3D vision models provide strong priors, their computational cost incurs prohibitive latency for real-time control. We propose Real-time 3D-aware Policy (R3DP), which integrates powerful 3D priors into manipulation policies without sacrificing real-time performance. A core innovation of R3DP is the asynchronous fast-slow collaboration module, which seamlessly integrates large-scale 3D priors into the policy without compromising real-time performance. The system maintains real-time efficiency by querying the pre-trained slow system (VGGT) only on sparse key frames, while simultaneously employing a lightweight Temporal Feature Prediction Network (TFPNet) to predict features for all intermediate frames. By leveraging historical data to exploit temporal correlations, TFPNet explicitly

arXiv:2603.14361v1 Announce Type: new Abstract: Recognizing complex behavioral states such as Ambivalence and Hesitancy (A/H) in naturalistic video settings remains a significant challenge in affective computing. Unlike basic facial expressions, A/H manifests as subtle, multimodal conflicts that require deep contextual and temporal understanding. In this paper, we propose a highly regularized, multimodal fusion pipeline to predict A/H at the video level. We extract robust unimodal features from visual, acoustic, and linguistic data, introducing a specialized statistical text modality explicitly designed to capture temporal speech variations and behavioral cues. To identify the most effective representations, we evaluate 15 distinct modality combinations across a committee of machine learning classifiers (MLP, Random Forest, and GBDT), selecting the most well-calibrated models based on validation Binary Cross-Entropy (BCE) loss. Furthermore, to optimally fuse these heterogeneous models

arXiv:2603.14002v1 Announce Type: new Abstract: A promising pathway for restoring communication in patients with dysarthria and anarthria is speech neuroprostheses, which directly decode speech from cortical neural activity. Two benchmarks, Brain-to-Text '24 and '25, released intracranial recordings from patients with dysarthria along with a baseline algorithm trained with Connectionist Temporal Classification (CTC). Despite significant innovation on these benchmarks, all leading published prior work relies on a WFST-based CTC decoder that requires ${\sim}$320 GB of RAM. These memory requirements limit accessibility for both patients and researchers. Here, we propose LightBeam, a non-WFST based CTC decoder that requires only ${\sim}$10 GB of RAM and achieves state-of-the-art performance on both benchmarks. LightBeam achieves this by integrating an LLM into the beam-search process via delayed fusion, obviating the prior need for using a large N-gram LM. LightBeam is implemented in

arXiv:2603.13749v1 Announce Type: new Abstract: Detecting subtle deviations in noisy acoustic environments is central to anomalous sound detection (ASD). A common training-free ASD pipeline temporally pools frame-level representations into a band-preserving feature vector and scores anomalies using a single nearest-neighbor match. However, this global matching can inflate normal-score variance through two effects. First, when normal sounds exhibit band-wise variability, a single global neighbor forces all bands to share the same reference, increasing band-level mismatch. Second, cosine-based matching is energy-coupled, allowing a few high-energy bands to dominate score computation under normal energy fluctuations and further increase variance. We propose BEAM, which stores temporally pooled sub-band vectors in a memory bank, retrieves neighbors per sub-band, and uniformly aggregates scores to reduce normal-score variability and improve discriminability. We further introduce a

arXiv:2603.12635v1 Announce Type: cross Abstract: High-fidelity numerical simulations of chaotic, high dimensional nonlinear dynamical systems are computationally expensive, necessitating the development of efficient surrogate models. Most surrogate models for such systems are deterministic, for example when neural operators are involved. However, deterministic models often fail to capture the intrinsic distributional uncertainty of chaotic systems. This work presents a surrogate modeling formulation that leverages generative machine learning, where a deep learning diffusion model is used to probabilistically forecast turbulent flows over long horizons. We introduce a multi-step autoregressive diffusion objective that significantly enhances long-rollout stability compared to standard single-step training. To handle complex, unstructured geometries, we utilize a multi-scale graph transformer architecture incorporating diffusion preconditioning and voxel-grid pooling. More importantly,

arXiv:2603.12972v1 Announce Type: new Abstract: The RAPTOR code is used to model how the time required for controlled termination of Ohmic plasmas scales from present tokamaks like TCV and JET, to reactor-grade tokamaks like ITER and DEMO. We show that ramping the plasma current $I_p$ down to 20% of the flat-top value over a time $\Delta t_{ramp-down}=\tau_{LR}=L_i/R$, with internal inductance $L_i$ and resistance $R$ evaluated at flat-top conditions, results in an approximately self-similar peaking of the current density for these four tokamaks, indicating the adequacy of $\tau_{LR}$ as a relevant timescale for cross-machine comparison, yielding $\tau_{LR} =$ 0.033s (TCV), 2.87s (JET), 63.2s (ITER) and 166.9s (DEMO). Note that $\tau_{LR}$ is easy to evaluate, both in systems codes and on a real-time control system. For the simulated ramp-downs with $\Delta t_{ramp-down}=\tau_{LR}$, the end-of-ramp-down normalized internal inductance $\ell_{i3}$ is limited below 2. An $I_p$ ramp-down

arXiv:2603.12377v1 Announce Type: new Abstract: Due to their simultaneous optimization for radial and parallel neoclassical transport, quasi-isodynamic fields have been the main choice of stellarator magnetic configuration for most fusion reactor candidates in recent years. However, achieving a high degree of quasi-isodynamicity often comes at the cost of a strong shaping of the flux surfaces of the stellarator and complex coil geometries. In this work, the concepts of quasi-isodynamicity and piecewise omnigenity are combined to form QI-pwO fields. These fields are quasi-isodynamic in the low-field region of the magnetic surface, whereas they significantly depart from quasi-isodynamicity in the high-field region without sacrificing the neoclassical transport properties of quasi-isodynamic fields. This departure could make it easier to integrate the optimization of neoclassical transport with other physical and technological aspects of a stellarator reactor.

arXiv:2603.12291v1 Announce Type: new Abstract: We introduce a reduced set of thermodynamic-topological observables for ordered multiscale dissipative systems. An interface-local quadratic reduction produces bounded integrity and residual channels, a flux-force stability channel, a weighted path-graph bottleneck channel, and a coarse-graining drift indicator. The goal is practical rather than universal: a compact and interpretable layer of observables that can be computed repeatedly and compared across regimes. The main case study is a fusion-relevant MHD Sabra shell model. Across 400 synthetic anomalous-dissipation probes, the local Prigogine-style channel detects 400/400 events, while a composite alarm detects 399/400 with lower latency. When an OPCR trigger and an energy-collapse proxy are both observed within the same event, the earliest OPCR trigger leads the proxy by 11.29+/-13.49 model-time units on average (median 6.15, IQR [1.23, 17.22]; 255/313 early cases). A scan over 5000

arXiv:2603.12721v1 Announce Type: new Abstract: Robust point cloud registration is a fundamental task in 3D computer vision and geometric deep learning, essential for applications such as large-scale 3D reconstruction, augmented reality, and scene understanding. However, the performance of established learning-based methods often degrades in complex, real world scenarios characterized by incomplete data, sensor noise, and low overlap regions. To address these limitations, we propose CMHANet, a novel Cross-Modal Hybrid Attention Network. Our method integrates the fusion of rich contextual information from 2D images with the geometric detail of 3D point clouds, yielding a comprehensive and resilient feature representation. Furthermore, we introduce an innovative optimization function based on contrastive learning, which enforces geometric consistency and significantly improves the model's robustness to noise and partial observations. We evaluated CMHANet on the 3DMatch and the

arXiv:2603.12635v1 Announce Type: new Abstract: High-fidelity numerical simulations of chaotic, high dimensional nonlinear dynamical systems are computationally expensive, necessitating the development of efficient surrogate models. Most surrogate models for such systems are deterministic, for example when neural operators are involved. However, deterministic models often fail to capture the intrinsic distributional uncertainty of chaotic systems. This work presents a surrogate modeling formulation that leverages generative machine learning, where a deep learning diffusion model is used to probabilistically forecast turbulent flows over long horizons. We introduce a multi-step autoregressive diffusion objective that significantly enhances long-rollout stability compared to standard single-step training. To handle complex, unstructured geometries, we utilize a multi-scale graph transformer architecture incorporating diffusion preconditioning and voxel-grid pooling. More importantly,

arXiv:2603.12588v1 Announce Type: new Abstract: Cross-modal ship re-identification (ReID) between optical and synthetic aperture radar (SAR) imagery is fundamentally challenged by the severe radiometric discrepancy between passive optical imaging and coherent active radar sensing. While existing approaches primarily rely on statistical distribution alignment or semantic matching, they often overlook a critical physical prior: ships are rigid objects whose geometric structures remain stable across sensing modalities, whereas texture appearance is highly modality-dependent. In this work, we propose SDF-Net, a Structure-Aware Disentangled Feature Learning Network that systematically incorporates geometric consistency into optical--SAR ship ReID. Built upon a ViT backbone, SDF-Net introduces a structure consistency constraint that extracts scale-invariant gradient energy statistics from intermediate layers to robustly anchor representations against radiometric variations. At the terminal

arXiv:2603.12139v1 Announce Type: new Abstract: We present a method for optimizing stellarator configurations that combine omnigenity and piecewise omnigenity (pwO). Within the \texttt{OOPS} optimization framework [Liu \textit{et al.}, arXiv:2502.09350 (2025)], we introduce a mapping technique that can ``squeeze'' general omnigenous fields to approximate pwO in the high-field side. Using this approach, we obtain a range of optimized configurations that combine poloidal omnigenity (PO) and pwO, spanning different field periods and aspect ratios. We further show that these configurations are compatible with a magnetic well. The resulting configurations exhibit favorable neoclassical transport and bootstrap current properties while partially relaxing the strict constraints of omnigenity. These results suggest that such configurations are promising candidates for future stellarator reactors.

arXiv:2603.11912v1 Announce Type: new Abstract: In the Tokamak \`a Configuration Variable (TCV), infrared thermography (IR) is currently composed of the horizontal, vertical, and tangential infrared systems (HIR, VIR, TIR), which all use Equus 81k M cameras. The IR diagnostics obtain the surface temperature of TCV's graphite tiles for post-discharge analysis. Target heat flux profiles are inferred from the tile temperature with the THEODOR (Thermal Energy Onto Divertor) code. Fast transient analysis is possible in reduced frame mode, with acquisition frequencies above 10kHz. The main views are the lower inner wall for HIR, the floor for VIR, and the lower outer wall for TIR. The HIR camera can also be moved to view the midplane inner wall, while TIR can be moved to see the midplane inner wall and the upper outer wall, mainly to measure synchrotron radiation and heat deposition due to runaway electrons. Recent developments in TCV's IR systems include (i) tile diffusivity and

arXiv:2603.11856v1 Announce Type: new Abstract: Radiated power information is crucial to diagnose and optimize the performance of fusion plasmas. Traditionally, at the TCV tokamak, radiated power analysis has only ever been possible following plasma discharge termination. However, recently, TCV bolometer data have become available in real-time. This offers the opportunity of integrating the radiated power information into the TCV plasma control system. In this work, we propose a novel real-time tomography-based Bayesian technique allowing estimation of the power radiated from user-defined regions of interest in the plasma. The real-time estimates are obtained as computationally cheap linear combinations of bolometer measurements, using pre-computed coefficients that are optimized for the specific discharge planned. This method is not, thus, trained on a set of synthetic or tomographically reconstructed emissivity profiles. We detail the derivation of the technique and show its

arXiv:2603.11737v1 Announce Type: new Abstract: In this paper, the design of the the plasma equilibrium and superconducting coils for the Electrons and Positrons in an Optimized Stellarator EPOS experiment is presented. With newly developed stellarator optimization tools, including single-stage and stochastic optimization, as well as HTS strain, this work demonstrates that it is possible to achieve key metrics for the buildability and confinement properties of the device. In particular, satisfactory quality of quasisymmetry and stellarator robustness is designed, and engineering requirements are met for eight different candidates. A feasibility study is presented that optimizes multiple candidates for different plasma major radii and coil currents, as well as the best EPOS candidate to date.

arXiv:2603.11699v1 Announce Type: new Abstract: In this paper, single-stage stellarator optimization is combined with stochastic coil optimization to improve the robustness of the stellarator as compared to deterministic methods. The plasma boundary, solved with an MHD solver in fixed-boundary mode, is linked to a set of randomly perturbed coils via the squared flux. The optimizer avoids sharp local minima and can reach improved configurations. Two different configurations obtained with our method, one quasi-axisymmetric and one quasi-helically symmetric, are compared against both the standard stochastic stage II method and the single-stage method. The new configurations shown here yield improved squared flux, quasisymmetry, and particle loss following a posteriori perturbation of the coils.

arXiv:2603.11549v1 Announce Type: new Abstract: PRAGYA is India's first privately developed low aspect ratio tokamak designed by Pranos Fusion Energy. The device is designed for a plasma major radius (R0) of about 0.4 m, a plasma minor radius (a) greater than 0.18 m, a plasma current (Ip) of up to 25 kA, and a toroidal magnetic field (B_T) of 0.1 T. The PRAGYA vacuum vessel incorporates several distinctive features, including a toroidal electrical break to minimize induced eddy currents and a double O-ring arrangement to reduce vacuum leakage. This paper presents the final design of the PRAGYA vacuum vessel and a comprehensive three-dimensional (3D) finite element model (FEM) assessment of its structural performance. The analysis evaluates the effects of self-weight, atmospheric pressure loading, and thermal stress arising from in-situ baking. The results confirm that the design satisfies the required safety margins under these combined loading conditions, providing a robust

arXiv:2603.11515v1 Announce Type: new Abstract: Today's scientific challenges, from climate modeling to Inertial Confinement Fusion design to novel material design, require exploring huge design spaces. In order to enable high-impact scientific discovery, we need to scale up our ability to test hypotheses, generate results, and learn from them rapidly. We present MADA (Multi-Agent Design Assistant), a Large Language Model (LLM) powered multi-agent framework that coordinates specialized agents for complex design workflows. A Job Management Agent (JMA) launches and manages ensemble simulations on HPC systems, a Geometry Agent (GA) generates meshes, and an Inverse Design Agent (IDA) proposes new designs informed by simulation outcomes. While general purpose, we focus development and validation on Richtmyer--Meshkov Instability (RMI) suppression, a critical challenge in Inertial Confinement Fusion. We evaluate on two complementary settings: running a hydrodynamics simulations on HPC

arXiv:2603.11392v1 Announce Type: new Abstract: Millimeter-wave or terahertz communications can meet demands of low-altitude economy networks for high-throughput sensing and real-time decision making. However, high-frequency characteristics of wireless channels result in severe propagation loss and strong beam directivity, which make beam prediction challenging in highly mobile uncrewed aerial vehicles (UAV) scenarios. In this paper, we employ agentic AI to enable the transformation of mmWave base stations toward embodied intelligence. We innovatively design a multi-agent collaborative reasoning architecture for UAV-to-ground mmWave communications and propose a hybrid beam prediction model system based on bimodal data. The multi-agent architecture is designed to overcome the limited context window and weak controllability of large language model (LLM)-based reasoning by decomposing beam prediction into task analysis, solution planning, and completeness assessment. To align with the

arXiv:2603.10838v1 Announce Type: cross Abstract: Understanding ion-matter interactions at the atomistic level is key to advancing materials for the semiconductor industry, space systems, and nuclear fusion technologies. However, most atomistic frameworks still rely on simplified descriptions of how ions transfer energy to the electronic subsystem, overlooking the sensitivity of this process to the actual ion path. Existing electron-ion interaction models, such as the tensorial unified two-temperature model, were developed to study self-irradiation scenarios, but their suitability for light-ion irradiation remains unexplored. Here, we propose that for light projectiles, stepping back from the tensorial formulation toward a simpler, local model of electronic stopping provides a more efficient and physically transparent trajectory-dependent description. We parameterize and validate both models for hydrogen and helium in tungsten using ab initio electronic stopping data and large-scale

arXiv:2603.10916v1 Announce Type: new Abstract: Machine learning models have demonstrated remarkable success in sports prediction in the past years, often treating sports prediction as a classification task within the field. This paper introduces new perspectives for analyzing sports data to predict outcomes more accurately. We leverage rankings to generate team rankings for the 2024 dataset using Combinatorial Fusion Analysis (CFA), a new paradigm for combining multiple scoring systems through the rank-score characteristic (RSC) function and cognitive diversity (CD). Our result based on rank combination with respect to team ranking has an accuracy rate of $74.60\%$, which is higher than the best of the ten popular public ranking systems ($73.02\%$). This exhibits the efficacy of CFA in enhancing the precision of sports prediction through different lens.

arXiv:2603.10678v1 Announce Type: new Abstract: Magnetohydrodynamic (MHD) effects play a key role in the design and operation of nuclear fusion systems, where electrically conducting fluids (such as liquid metals or molten salts in reactor blankets) interact with magnetic fields of varying intensity and orientation, which affect the resulting flow. The numerical resolution of MHD models involves highly nonlinear multiphysics systems of equations and can become computationally expensive, particularly in multi-query, parametric, or real-time contexts. This work investigates a fully data-driven framework for MHD state reconstruction that combines dimensionality reduction via Singular Value Decomposition (SVD) with the SHallow REcurrent Decoder (SHRED), a neural network architecture designed to recover the full spatio-temporal state from sparse time-series measurements of a limited number of observables. The methodology is applied to a parametric MHD test case involving compressible

arXiv:2603.10049v1 Announce Type: new Abstract: Ensemble learning is a well established body of methods for machine learning to enhance predictive performance by combining multiple algorithms/models. Combinatorial Fusion Analysis (CFA) has provided method and practice for combining multiple scoring systems, using rank-score characteristic (RSC) function and cognitive diversity (CD), including ensemble method and model fusion. However, there is no general-purpose Python tool available that incorporate these techniques. In this paper we introduce \texttt{InFusionLayer}, a machine learning architecture inspired by CFA at the system fusion level that uses a moderate set of base models to optimize unsupervised and supervised learning multiclassification problems. We demonstrate \texttt{InFusionLayer}'s ease of use for PyTorch, TensorFlow, and Scikit-learn workflows by validating its performance on various computer vision datasets. Our results highlight the practical advantages of

arXiv:2603.10007v1 Announce Type: new Abstract: We present our approach to the AbjadGenEval shared task on detecting AI-generated Arabic text. We fine-tuned the multilingual E5-large encoder for binary classification, and we explored several pooling strategies to pool token representations, including weighted layer pooling, multi-head attention pooling, and gated fusion. Interestingly, none of these outperformed simple mean pooling, which achieved an F1 of 0.75 on the test set. We believe this is because complex pooling methods introduce additional parameters that need more data to train properly, whereas mean pooling offers a stable baseline that generalizes well even with limited examples. We also observe a clear pattern in the data: human-written texts tend to be significantly longer than machine-generated ones.

arXiv:2603.09689v1 Announce Type: new Abstract: Visual Question Answering (VQA) is a fundamental multimodal task that requires models to jointly understand visual and textual information. Early VQA systems relied heavily on language biases, motivating subsequent work to emphasize visual grounding and balanced datasets. With the success of large-scale pre-trained transformers for both text and vision domains -- such as PhoBERT for Vietnamese language understanding and Vision Transformers (ViT) for image representation learning -- multimodal fusion has achieved remarkable progress. For Vietnamese VQA, several datasets have been introduced to promote research in low-resource multimodal learning, including ViVQA, OpenViVQA, and the recently proposed ViTextVQA. These resources enable benchmarking of models that integrate linguistic and visual features in the Vietnamese context. Evaluation of VQA systems often employs automatic metrics originally designed for image captioning or machine

arXiv:2603.09685v1 Announce Type: new Abstract: To overcome the limitations of manual administrative coding in geriatric Cardiovascular Risk Management, this study introduces an automated classification framework leveraging unstructured Electronic Health Records (EHRs). Using a dataset of 3,482 patients, we benchmarked three distinct modeling paradigms on longitudinal Dutch clinical narratives: classical machine learning baselines, specialized deep learning architectures optimized for large-context sequences, and general-purpose generative Large Language Models (LLMs) in a zero-shot setting. Additionally, we evaluated a late fusion strategy to integrate unstructured text with structured medication embeddings and anthropometric data. Our analysis reveals that the custom Transformer architecture outperforms both traditional methods and generative \acs{llm}s, achieving the highest F1-scores and Matthews Correlation Coefficients. These findings underscore the critical role of specialized

arXiv:2603.09333v1 Announce Type: new Abstract: Low-cost embedded processors such as the ESP32 (Xtensa LX6, 32-bit dual-core, 240 MHz) are increasingly used in edge computing applications that require real-time physical simulation, sensor fusion, and control systems. Although the ESP32 integrates a single-precision IEEE 754 floating-point unit, floating-point operations introduce pipeline overhead and higher energy consumption compared to integer arithmetic, limiting throughput for floating-point intensive workloads. This paper presents the design, formal specification, and empirical evaluation of a Dynamic Precision Math Engine for the ESP32. The system integrates three main components: a Q16.16 fixed-point arithmetic core that maps mathematical operations onto the integer pipeline of the Xtensa LX6, a 16-iteration CORDIC trigonometric module that computes sine and cosine using only additions and bit shifts, and a cache-aware tiled matrix multiplication kernel with deferred

arXiv:2603.09266v1 Announce Type: new Abstract: Current text-to-3D generation methods excel in natural scenes but struggle with industrial applications due to two critical limitations: domain adaptation challenges where conventional LoRA fusion causes knowledge interference across categories, and geometric reasoning deficiencies where pairwise consistency constraints fail to capture higher-order structural dependencies essential for precision manufacturing. We propose a novel framework named ForgeDreamer addressing both challenges through two key innovations. First, we introduce a Multi-Expert LoRA Ensemble mechanism that consolidates multiple category-specific LoRA models into a unified representation, achieving superior cross-category generalization while eliminating knowledge interference. Second, building on enhanced semantic understanding, we develop a Cross-View Hypergraph Geometric Enhancement approach that captures structural dependencies spanning multiple viewpoints

arXiv:2603.09038v1 Announce Type: new Abstract: Finite element simulations play a critical role in a wide range of applications, from automotive design to tsunami modeling and computational electromagnetics. Performing these simulations efficiently at the high resolutions needed for practical applications and scientific insights necessitates the use of high-order methods and large-scale supercomputing. While much progress has been made in porting finite element codes to GPU systems in recent years, additional improvements in the efficiency and computational speed of GPU-accelerated high-order finite element simulations are in constant demand. In this paper, we demonstrate that the FP64 tensor cores on NVIDIA GPUs can be used to further accelerate such simulations, achieving significant speedups in key kernels of MFEM, a scalable open-source finite element library widely used in HPC applications. By integrating FP64 tensor cores with kernel fusion optimizations, we were able to achieve

Researchers at the Department of Energy’s Oak Ridge National Laboratory are helping to enable the next generation of

arXiv:2603.06733v1 Announce Type: new Abstract: Credit risk scoring must support high-stakes lending decisions where data distributions change over time, probability estimates must be reliable, and group-level fairness is required. While modern machine learning models improve default prediction accuracy, they often produce poorly calibrated scores under distribution shift and may create unfair outcomes when trained without explicit constraints. This paper proposes Calibrated Credit Intelligence (CCI), a deployment-oriented framework that combines (i) a Bayesian neural risk scorer to capture epistemic uncertainty and reduce overconfident errors, (ii) a fairnessconstrained gradient boosting model to control group disparities while preserving strong tabular performance, and (iii) a shiftaware fusion strategy followed by post-hoc probability calibration to stabilize decision thresholds in later time periods. We evaluate CCI on the Home Credit Credit Risk Model Stability benchmark using a

arXiv:2603.06733v1 Announce Type: cross Abstract: Credit risk scoring must support high-stakes lending decisions where data distributions change over time, probability estimates must be reliable, and group-level fairness is required. While modern machine learning models improve default prediction accuracy, they often produce poorly calibrated scores under distribution shift and may create unfair outcomes when trained without explicit constraints. This paper proposes Calibrated Credit Intelligence (CCI), a deployment-oriented framework that combines (i) a Bayesian neural risk scorer to capture epistemic uncertainty and reduce overconfident errors, (ii) a fairnessconstrained gradient boosting model to control group disparities while preserving strong tabular performance, and (iii) a shiftaware fusion strategy followed by post-hoc probability calibration to stabilize decision thresholds in later time periods. We evaluate CCI on the Home Credit Credit Risk Model Stability benchmark using

arXiv:2603.08459v1 Announce Type: new Abstract: Safe predictions are a crucial requirement for integrating predictive models into clinical decision support systems. One approach for ensuring trustworthiness is to enable models' ability to express their uncertainty about individual predictions. However, current machine learning models frequently lack reliable uncertainty estimation, hindering real-world deployment. This is further observed in multimodal settings, where the goal is to enable effective information fusion. In this work, we propose $\texttt{MedCertAIn}$, a predictive uncertainty framework that leverages multimodal clinical data for in-hospital risk prediction to improve model performance and reliability. We design data-driven priors over neural network parameters using a hybrid strategy that considers cross-modal similarity in self-supervised latent representations and modality-specific data corruptions. We train and evaluate the models with such priors using clinical

arXiv:2603.08240v1 Announce Type: new Abstract: Collaborative perception integrates multi-agent perspectives to enhance the sensing range and overcome occlusion issues. While existing multimodal approaches leverage complementary sensors to improve performance, they are highly prone to failure--especially when a key sensor like LiDAR is unavailable. The root cause is that feature fusion leads to semantic mismatches between single-modality features and the downstream modules. This paper addresses this challenge for the first time in the field of collaborative perception, introducing Single-Modality-Operable Multimodal Collaborative Perception (SiMO). By adopting the proposed Length-Adaptive Multi-Modal Fusion (LAMMA), SiMO can adaptively handle remaining modal features during modal failures while maintaining consistency of the semantic space. Additionally, leveraging the innovative "Pretrain-Align-Fuse-RD" training strategy, SiMO addresses the issue of modality competition--generally

arXiv:2603.07630v1 Announce Type: new Abstract: Nasotracheal intubation (NTI) is a vital procedure in emergency airway management, where rapid and accurate glottis detection is essential to ensure patient safety. However, existing machine assisted visual detection systems often rely on high performance computational resources and suffer from significant inference delays, which limits their applicability in time critical and resource constrained scenarios. To overcome these limitations, we propose Mobile GlottisNet, a lightweight and efficient glottis detection framework designed for real time inference on embedded and edge devices. The model incorporates structural awareness and spatial alignment mechanisms, enabling robust glottis localization under complex anatomical and visual conditions. We implement a hierarchical dynamic thresholding strategy to enhance sample assignment, and introduce an adaptive feature decoupling module based on deformable convolution to support dynamic

arXiv:2603.07249v1 Announce Type: new Abstract: Second primary cancer (SPC), a new cancer in patients different from previously diagnosed, is a growing concern due to improved cancer survival rates. Early prediction of SPC is essential to enable timely clinical interventions. This study focuses on lung cancer survivors treated in Taiwanese hospitals, where the limited size and geographic scope of local datasets restrict the effectiveness and generalizability of traditional machine learning approaches. To address this, we incorporate external data from the publicly available US-based Surveillance, Epidemiology, and End Results (SEER) program, significantly increasing data diversity and scale. However, the integration of multi-source datasets presents challenges such as feature inconsistency and privacy constraints. Rather than naively merging data, we proposed a loss fusion horizontal federated learning (LF2L) framework that can enable effective cross-institutional collaboration while

arXiv:2603.06925v1 Announce Type: new Abstract: Targets in remote sensing images are usually small, weakly textured, and easily disturbed by complex backgrounds, challenging high-precision detection with general algorithms. Building on our earlier ESM-YOLO, this work presents ESM-YOLO+ as a lightweight visible infrared fusion network. To enhance detection, ESM-YOLO+ includes two key innovations. (1) A Mask-Enhanced Attention Fusion (MEAF) module fuses features at the pixel level via learnable spatial masks and spatial attention, effectively aligning RGB and infrared features, enhancing small-target representation, and alleviating cross-modal misalignment and scale heterogeneity. (2) Training-time Structural Representation (SR) enhancement provides auxiliary supervision to preserve fine-grained spatial structures during training, boosting feature discriminability without extra inference cost. Extensive experiments on the VEDAI and DroneVehicle datasets validate ESM-YOLO+'s superiority.

arXiv:2507.19229v2 Announce Type: replace-cross Abstract: The modeling of genomic sequences presents unique challenges due to their length and structural complexity. Traditional sequence models struggle to capture long-range dependencies and biological features inherent in DNA. In this work, we propose TrinityDNA, a novel DNA foundational model designed to address these challenges. The model integrates biologically informed components, including Groove Fusion for capturing DNA's structural features and Gated Reverse Complement (GRC) to handle the inherent symmetry of DNA sequences. Additionally, we introduce a multi-scale attention mechanism that allows the model to attend to varying levels of sequence dependencies, and an evolutionary training strategy that progressively adapts the model to both prokaryotic and eukaryotic genomes. TrinityDNA provides a more accurate and efficient approach to genomic sequence modeling, offering significant improvements in gene function prediction,

arXiv:2603.06298v1 Announce Type: new Abstract: Axial plugging is a critical challenge for fusion in open-ended magnetic confinement systems. Multi-mirror systems, consisting of a series of axially aligned magnetic mirrors, aim to enhance axial confinement by increasing the effective diffusion coefficient; however, additional plugging is required to meet the Lawson criterion. In [T. Miller et al., Phys. Plasmas 30, 072510 (2023)], it was found that applying a traveling and rotating electric field in multi-mirror machines can significantly suppress axial loss due to a selectivity effect induced by the Doppler shift of the ion cyclotron resonance. However, this method is energetically expensive and vulnerable to plasma screening effects. Here, we propose using a traveling, rotating magnetic field that can achieve comparable plugging effectiveness while offering better penetration and lower energy costs. Two limiting scenarios, with and without an induced electric field, were considered.

arXiv:2603.06234v1 Announce Type: new Abstract: Plasmoid growth is considered to enhance the rate of magnetic reconnection and is frequently used to explain fast mag netic reconnection in highly conductive (collisionless) plasmas. In strongly magnetized plasmas, the long wavelength dimension parallel to the magnetic field can be separated from the small wavelength perpendicular plane, justifying an isolated 2D approach. While 2D systems have been simulated using delta F gyrofluids, a novel Full-F gyrofluid model with arbitrary wavelength polarization is used to simulate 2D Harris-sheet magnetic reconnection with domain aspect ratios Ly/Lx

arXiv:2603.05718v1 Announce Type: new Abstract: Activation foils are used to independently measure the time integrated neutron yield and total fusion energy produced in both inertial and magnetic confinement fusion, making them crucial in the neutron diagnostic suite. The activated foils must be remotely transported from the neutron source to the detector inside of a small capsule, which can impact both the foil irradiation and the associated activation measurement. The aim of this paper is to evaluate the performance of various activation foils and to characterize the effects of different capsule materials to inform the design choices for future systems, such as the SPARC tokamak. Through a combination of FISPACT simulations and irradiation experiments with a deuterium-tritium neutron generator, we tested several different material choices for foils, capsules, and gamma-ray spectrometers. Aluminum and copper foils are found to be suitable for a multi-foil irradiation configuration.

arXiv:2603.06250v1 Announce Type: new Abstract: Generalised 3D Referring Expression Segmentation (3D-GRES) localizes objects in 3D scenes based on natural language, even when descriptions match multiple or zero targets. Existing methods rely solely on sparse point clouds, lacking rich visual semantics for fine-grained descriptions. We propose HCF-RES, a multi-modal framework with two key innovations. First, Hierarchical Visual Semantic Decomposition leverages SAM instance masks to guide CLIP encoding at dual granularities -- pixel-level and instance-level features -- preserving object boundaries during 2D-to-3D projection. Second, Progressive Multi-level Fusion integrates representations through intra-modal collaboration, cross-modal adaptive weighting between 2D semantic and 3D geometric features, and language-guided refinement. HCF-RES achieves state-of-the-art results on both ScanRefer and Multi3DRefer.

arXiv:2603.05807v1 Announce Type: new Abstract: Dynamic vision sensors, also known as event cameras, are rapidly rising in popularity for robotic and computer vision tasks due to their sparse activation and high-temporal resolution. Event cameras have been used in robotic navigation and localization tasks where accurate positioning needs to occur on small and frequent time scales, or when energy concerns are paramount. In this work, we present EventGeM, a state-of-the-art global to local feature fusion pipeline for event-based Visual Place Recognition. We use a pre-trained vision transformer (ViT-S/16) backbone to obtain global feature patch for initial match predictions embeddings from event histogram images. Local feature keypoints were then detected using a pre-trained MaxViT backbone for 2D-homography based re-ranking with RANSAC. For additional re-ranking refinement, we subsequently used a pre-trained vision foundation model for depth estimation to compare structural similarity

arXiv:2603.04748v1 Announce Type: new Abstract: Motivation: Receptor-binding proteins (RBPs) initiate viral infection and determine host specificity, serving as key targets for phage engineering and therapy. However, the identification of RBPs is complicated by their extreme sequence divergence, which often renders traditional homology-based alignment methods ineffective. While machine learning offers a promising alternative, such approaches struggle with severe class imbalance and the difficulty of selecting informative negative samples from heterogeneous tail proteins. Existing methods often fail to balance learning from these ``hard negatives'' while maintaining generalization. Results: We present SeekRBP, a sequence--structure framework that models negative sampling as a sequential decision-making problem. By employing a multi-armed bandit strategy, SeekRBP dynamically prioritizes informative non-RBP sequences based on real-time training feedback, complemented by a multimodal

arXiv:2603.04457v1 Announce Type: cross Abstract: The fundamental topology of manufacturing has not undergone a paradigm-level transformation since Henry Ford's moving assembly line in 1913. Every major innovation of the past century, from the Toyota Production System to Industry 4.0, has optimized within the Fordist paradigm without altering its structural logic: centralized mega-factories, located near labor pools, producing at scale. We argue that embodied intelligence is poised to break this century-long stasis, not by making existing factories more efficient, but by triggering phase transitions in manufacturing economic geography itself. When embodied AI capabilities cross critical thresholds in dexterity, generalization, reliability, and tactile-vision fusion, the consequences extend far beyond cost reduction: they restructure where factories are built, how supply chains are organized, and what constitutes viable production scale. We formalize this by defining a Capability Space

arXiv:2603.05003v1 Announce Type: new Abstract: Turbulent transport near the X-point of a large tokamak is examined using local, gradient-driven simulations that determine the saturated plasma profiles. The distribution of a representative set of particle tracers evolving within these profiles is then analyzed. The study demonstrates that the resulting transport is diffusive, characterized by a coefficient that depends on the spectral properties of the turbulent energy and attains anomalous high values under broad conditions. These findings suggest that anomalous transport is an inherent outcome of the fundamental non-linear drift dynamics of plasmas. The scaling of transport with turbulent energy is also addressed, with implications for future progress toward a mean-field framework for turbulent transport.

arXiv:2603.04709v1 Announce Type: new Abstract: This work investigates toroidal momentum transport in type-I ELMy H-mode plasmas in the ASDEX Upgrade tokamak, focusing on the formation of hollow rotation profiles under strong electron cyclotron resonance heating (ECRH). Applying the established momentum transport analysis framework to a neutral beam injection (NBI) modulation experiment, momentum transport coefficients were inferred self-consistently. This was done for phases with dominant NBI heating and with additional strong ECRH, during which the rotation profile severely collapsed without significant changes in the externally applied torque. The experimental rotation profiles were accurately reproduced, confirming the robustness of the inferred diffusive, convective, and residual-stress contributions. While the Prandtl number and inward Coriolis pinch remained comparable between phases, the NBI+ECRH phase exhibited a strong counter-current intrinsic torque. Linear gyrokinetic

arXiv:2603.05255v1 Announce Type: new Abstract: Cooperative perception significantly enhances scene understanding by integrating complementary information from diverse agents. However, existing research often overlooks critical challenges inherent in real-world multi-source data integration, specifically high temporal latency and multi-source noise. To address these practical limitations, we propose Collaborative Alignment and Transformation Network (CATNet), an adaptive compensation framework that resolves temporal latency and noise interference in multi-agent systems. Our key innovations can be summarized in three aspects. First, we introduce a Spatio-Temporal Recurrent Synchronization (STSync) that aligns asynchronous feature streams via adjacent-frame differential modeling, establishing a temporal-spatially unified representation space. Second, we design a Dual-Branch Wavelet Enhanced Denoiser (WTDen) that suppresses global noise and reconstructs localized feature distortions

arXiv:2603.05114v1 Announce Type: new Abstract: Pedestrian Attribute Recognition is a foundational computer vision task that provides essential support for downstream applications, including person retrieval in video surveillance and intelligent retail analytics. However, existing research is frequently constrained by the ``one-model-per-dataset" paradigm and struggles to handle significant discrepancies across domains in terms of modalities, attribute definitions, and environmental scenarios. To address these challenges, we propose UniPAR, a unified Transformer-based framework for PAR. By incorporating a unified data scheduling strategy and a dynamic classification head, UniPAR enables a single model to simultaneously process diverse datasets from heterogeneous modalities, including RGB images, video sequences, and event streams. We also introduce an innovative phased fusion encoder that explicitly aligns visual features with textual attribute queries through a late deep fusion

arXiv:2603.04989v1 Announce Type: new Abstract: Tracking any point (TAP) is a fundamental yet challenging task in computer vision, requiring high precision and long-term motion reasoning. Recent attempts to combine RGB frames and event streams have shown promise, yet they typically rely on synchronous or non-adaptive fusion, leading to temporal misalignment and severe degradation when one modality fails. We introduce TAPFormer, a transformer-based framework that performs asynchronous temporal-consistent fusion of frames and events for robust and high-frequency arbitrary point tracking. Our key innovation is a Transient Asynchronous Fusion (TAF) mechanism, which explicitly models the temporal evolution between discrete frames through continuous event updates, bridging the gap between low-rate frames and high-rate events. In addition, a Cross-modal Locally Weighted Fusion (CLWF) module adaptively adjusts spatial attention according to modality reliability, yielding stable and

arXiv:2603.04595v1 Announce Type: new Abstract: Duplicate records pose significant challenges in customer relationship management (CRM)and healthcare, often leading to inaccuracies in analytics, impaired user experiences, and compliance risks. Traditional deduplication methods rely heavily on direct identifiers such as names, emails, or Social Security Numbers (SSNs), making them ineffective under strict privacy regulations like GDPR and HIPAA, where such personally identifiable information (PII) is restricted or masked. In this research, I propose a novel, scalable, multimodal AI framework for detecting duplicates without depending on sensitive information. This system leverages three distinct modalities: semantic embeddings derived from textual fields (names, cities) using pre-trained DistilBERT models, behavioral patterns extracted from user login timestamps, and device metadata encoded through categorical embeddings. These heterogeneous modalities are combined using a late fusion

arXiv:2603.04457v1 Announce Type: new Abstract: The fundamental topology of manufacturing has not undergone a paradigm-level transformation since Henry Ford's moving assembly line in 1913. Every major innovation of the past century, from the Toyota Production System to Industry 4.0, has optimized within the Fordist paradigm without altering its structural logic: centralized mega-factories, located near labor pools, producing at scale. We argue that embodied intelligence is poised to break this century-long stasis, not by making existing factories more efficient, but by triggering phase transitions in manufacturing economic geography itself. When embodied AI capabilities cross critical thresholds in dexterity, generalization, reliability, and tactile-vision fusion, the consequences extend far beyond cost reduction: they restructure where factories are built, how supply chains are organized, and what constitutes viable production scale. We formalize this by defining a Capability Space C

arXiv:2603.04085v1 Announce Type: cross Abstract: In stellarator design, the coil winding surfaces $\Sigma\subset\mathbb R^3$ support current distributions $j$ that shape the magnetic field. This work provides a theoretical framework explaining the emergence of centre and saddle point regions, a key feature in coil optimisation. For coil winding surfaces with a toroidal shape, we prove a dichotomy principle: the current distribution has both centre and saddle point regions or is no-where vanishing. For coil winding surfaces that consist of piecewise cylinders, we show that if $j$ is oppositely oriented on the two boundary circles, centre and saddle points appear, and all but finitely many field lines of $j$ are periodic. When $j$ admits a harmonic potential, all field lines are closed poloidal orbits. These results offer insights into current patterns on winding surfaces, with implications for coil design strategies and their simplification.

arXiv:2603.04085v1 Announce Type: new Abstract: In stellarator design, the coil winding surfaces $\Sigma\subset\mathbb R^3$ support current distributions $j$ that shape the magnetic field. This work provides a theoretical framework explaining the emergence of centre and saddle point regions, a key feature in coil optimisation. For coil winding surfaces with a toroidal shape, we prove a dichotomy principle: the current distribution has both centre and saddle point regions or is no-where vanishing. For coil winding surfaces that consist of piecewise cylinders, we show that if $j$ is oppositely oriented on the two boundary circles, centre and saddle points appear, and all but finitely many field lines of $j$ are periodic. When $j$ admits a harmonic potential, all field lines are closed poloidal orbits. These results offer insights into current patterns on winding surfaces, with implications for coil design strategies and their simplification.

arXiv:2603.03952v1 Announce Type: new Abstract: Effects of neoclassical toroidal viscosity (NTV) on plasma flow evolution in the presence of resonant magnetic perturbation (RMP) in a tokamak have been evaluated using a cylindrical theory model. Calculations show that the introduction of NTV has almost no effect on the flow on the resonant surface, so the locked or unlocked state on the resonant surface remains unchanged, but it impacts the rotation profile in the core region. The toroidal, poloidal, and parallel flows in the core region are slightly reduced with uniform pressure. For non-uniform pressure profiles, elevated $\beta$ enhances the global amplitude of NTV torque but suppresses that of electromagnetic (EM) torque. These two driving terms collectively maintain the locked mode state.