Papers

Analytical solutions to the three-dimensional ideal magnetohydrodynamic equilibrium equations $(\nabla \times \vec{B}) \times \vec{B} = \nabla p$ and $\nabla \cdot \vec{B} = 0$ provide benchmarks for stellarator equilibrium codes. Two families of non-axisymmetric solutions were recently discovered by Landreman [arXiv:2609.26742] as counterexamples to Grad's conjecture, but both satisfy stellarator symmetry. Here, we generalize both families by introducing a symmetry-breaking parameter into each of the solutions, resulting in new non-axisymmetric solutions that break stellarator symmetry and reduce to the stellarator symmetric solutions when the newly introduced parameter is set to zero.

The complex of injective words is a cell complex that arises in a number of different areas. It has applications to proving homological stability and to the study of group cohomology, and it is closely related to the random-to-random Markov chain. This complex was first studied by Farmer, who proved it has the homology of a wedge of top-dimensional spheres. Later, Björner and Wachs established its shellability, and Reiner and Webb uncovered its $S_n$-module structure, observing in the process that the rank of its top homology group is the $n$th derangement number. We introduce natural subcomplexes of the complex of injective words by fixing a permutation pattern $\sigma$ and considering only those injective words in the alphabet $\{1,2,\dots,n\} $ that avoid $\sigma$. We prove that such pattern-avoiding complexes are shellable if $\sigma$ begins or ends with its largest or smallest letter, and we construct homology bases for the complexes avoiding such patterns. For patterns of length 3, all of which have this property, we show that the rank of the top homology of the resulting complex is a Riordan number. All but four patterns of length 4 also have this property, and for two of the remaining four patterns, we establish shellability using a different method. We also introduce a technique to use enumerative combinatorics to prove shellability, and we apply it to the complex of separable injective words, thereby deducing shellability in this case. Along the way, we give a combinatorial formula for all of the $h$-numbers in the full complex of injective words as well as for each of the subcomplexes which we prove are shellable. Going in the other direction, we use shellability of complexes of pattern-avoiding injective words to deduce new refined counting formulas for pattern-avoiding permutations.

We prove McLeod's (1959) conjecture that $(h_{\ell+r}/h_\ell)^{1/r}$ is subadditive for all $\ell\ge0$ and $r\ge1$, where $h_k$ is the complete homogeneous symmetric polynomial of degree $k$ (J. B. McLeod, "On four inequalities in symmetric functions", Proc. Edinburgh Math. Soc., 1959). We first prove that the ratio $h_k/h_{k-1}$ is convex, and our proof takes an unexpected detour through queuing theory. The geometric mean of $r$ consecutive such ratios equals McLeod's conjectured function, but geometric means of convex functions need not be convex. Our second and decisive step is a tensorization trick that couples neighboring ratios tightly enough to make their geometric mean convex.

Sparse attention mechanisms have become increasingly important for transformer models processing long input sequences due to their lower computational and memory complexity compared to full self-attention. Longformer achieves this through a sliding-window attention mechanism that produces a structured banded sparse attention matrix. However, existing sparse transformer accelerators primarily target attention generation or unstructured sparsity, leaving sparse matrix–dense matrix multiplication (SpMM) for structured sparse attention largely unexplored. This paper presents a pipelined FPGA architecture for accelerating banded SpMM in Longformer. The proposed design exploits the predictable sparsity pattern of Longformer's attention matrix through a custom row-wise storage scheme with implicit indexing, eliminating the overhead of conventional sparse matrix formats while enabling regular memory accesses. The architecture employs parallel processing elements, pipelined adder trees, and a dual-path computation strategy to maximize throughput and hardware utilization. Implemented in Verilog and evaluated on an RFSoC platform using Vivado 2024.2, the accelerator sustains one complete dot-product result per clock cycle after an initial latency of 11 cycles while maintaining power consumption below 2.9 W. Operating at 100 MHz, the design achieves over 100 million dot-product outputs per second, demonstrating the effectiveness of directly exploiting structured sparsity for sparse transformer acceleration.

This paper introduces Bayesian Continuation Model Predictive Control (BC-MPC), an adaptive method that replans near-term antenna pointings while continuing a feasible baseline sweep, for finding and establishing contact with spacecraft immediately after launch when their true positions are most uncertain. After separation, launch and deployment errors can leave the spacecraft sufficiently far from its predicted position that operators must actively search for it during ground contacts to establish communications. Each missed detection provides information about which orbital hypotheses remain plausible, but searching one part of the uncertainty region can consume the time needed to reach another. BC-MPC updates a particle-based orbital belief after missed detections and replans antenna pointing guidance while retaining feasibility. Its continuation helps preserve future search coverage that greedy and short-horizon methods can sacrifice for immediate acquisition gains. We compare BC-MPC against the operationally prevalent along-track sweep and additional search algorithms under common search-time budgets, initial uncertainty, and antenna constraints. Across 96 simulated deployments with matched uncertainty, BC-MPC improves acquisition probability over the along-track sweep and reduces restricted mean acquisition time. We additionally evaluate the method using pre-launch, post-launch, and initial acquisition ephemerides for 83 spacecraft from the Transporter-16 rideshare launch. Using either pre-deployment or post-deployment two-line element sets (TLEs) for planning, BC-MPC improves modeled acquisition probability over the along-track sweep by approximately 12 percentage points under the commonly assumed uncertainty model.

In this paper, we present a model-based Koopman-Galerkin framework for output-feedback control of nonlinear systems with known dynamics. With a suitable actuator augmentation, the finite-dimensional realization preserves the actuator dynamics and the constant input matrix. The approach is based on an exact decomposition of the lifted dynamics into an LTI system and a finite-order closure residual. It also applies to non-control-affine systems and allows dictionaries that do not span an invariant subspace. Regional bounds are derived for the closure residual, actuator-realization defect, and output-reconstruction error. We use these bounds to formulate a standard $\mathcal{H}_\infty$ output-feedback problem. The controller is linear and dynamic and is implemented using only the measured tracking error. Neither full-state measurements nor a separate state observer are required. Lifting is performed offline. With the state coordinates retained in the dictionary, we obtain explicit transient bounds on the original state and tracking error. A sufficient condition for regional containment is also given. Under this condition, the nonlinear closed loop is forward complete, and the state and tracking error are uniformly ultimately bounded. The cart-pendulum example illustrates the nonlinear closed-loop guarantees. The proposed output feedback gives lower RMSE and peak tracking error on average than Jacobian-based $\mathcal{H}_\infty$ feedback, full-state backstepping, and Koopman-based hybrid MPC.

Cyber threat analysis increasingly depends on evidence distributed across vendor advisories, vulnerability databases, and threat intelligence sources. Turning these fragmented observations into timely decisions requires models to connect technical severity with evolving exploitation evidence and available defensive actions. We present POLAR, an LLM-powered framework for synthesizing real-world cyber evidence into threat-centric assessments for prioritization and mitigation. POLAR first disentangles overlapping incidents and grounds each threat in source-linked evidence. For prioritization, it infers severity metrics from cyber evidence and combines the resulting assessment with temporally ordered exploitation signals to estimate near-term exploitation likelihood. For mitigation, it links the synthesized threat data to authoritative remediation knowledge and organizes applicable actions according to threat urgency and operational constraints. We evaluate POLAR on real-world vulnerability evidence collected from public resources and compare it with multiple baselines. Across heterogeneous incidents and zero-day settings, POLAR improves threat ranking and mitigation retrieval while producing evidence-linked intermediate assessments that support analyst inspection. The results establish evidence synthesis as a practical foundation for LLM-based cyber decision support across related security tasks.

Information sharing, threat perception, and collective evasion highlight the advantages of swarm intelligence. Here, we study the behavior of flocks of cognitive agents, based on numerical simulations of the augmented inertial spin model, with an explicit predator. Each agent on the predator-facing side of the flock perceives the threat, triggering collective turns through local interactions. We show that the augmented inertial spin model sustains nearly linear information propagation in the under-damped regime, even with realistic flocking interactions such as local collision avoidance and swarm cohesion in addition to alignment. The turning distance of the flock to a stationary predator scales with the ratio $\Delta_p/(v_0\sqrt{\eta})$, identifying an optimal balance between response strength $\Delta_p$, flock speed $v_0$, and damping $\eta$ of information propagation. Fragmentation arises when different parts of the flock turn in different directions, however, fast information propagation can synchronize these responses and maintain flock cohesion while under pursuit. For two predators, a combination of stability analysis and simulations reveals a stark contrast between flock and single-particle reactions, with scattering angle shaped by flock geometry, predator position, and the heterogeneous responses of the detecting front. Together, these results show how information propagation, flock geometry, and dynamics together shapes collective escape, providing general principles relevant to both biological flocks and robotic swarms.

E-textile design involves complex decisions across materials, sensor and actuator structures, fabrication, garment integration, and data processing. It typically requires iterative prototyping and testing, which are time- and labour-intensive, while few practitioners possess cross-disciplinary expertise across all relevant domains. To identify current bottlenecks and explore how Generative AI (GenAI) might support the design process, we conducted two half-day co-design workshops, one in Shanghai and one in Winchester, with practitioners from materials science, electronics, garment design, human-computer interaction, and manufacturing. Twenty practitioners participated in the Shanghai workshop; ten of them had prior experience in e-textiles and form the contributing sample analysed here. A further ten practitioners participated in the Winchester workshop. Participants mapped their own design pipelines, annotated bottlenecks, and proposed where GenAI could provide support. Rather than presenting a ranked list of opportunities, we report a process map that indexes each proposed GenAI role to the pipeline stage at which practitioners located it, together with the conditions on which they stated its usefulness would depend. Across both sites, practitioners consistently identified domain-specific operational barriers, including data scarcity, the disconnect between prototyping and manufacturing, and trade-offs in material-hardware integration. They also emphasized that the primary barrier to GenAI-driven e-textile design is not general model capability, but the lack of standardized, machine-readable representations of e-textile designs. Based on these findings, we identify four classes of domain-tailored AI tools that could support future e-textile design processes.

We consider the problem of learning nonlinear operators $G_0: \mathcal X\to \mathcal Y$ from finitely many noisy observations. While finite-regularity classes such as Lipschitz- or $C^k$-operators suffer from the curse of sample complexity (i.e. algebraic sample complexity is impossible), algebraic convergence rates can be obtained for holomorphic operators. In this paper we study information-theoretic minimax rates for estimating holomorphic operators. We provide algebraic upper and lower bounds of the metric entropy of holomorphy classes (matching up to an arbitrarily small loss in the exponent) as well as statistical minimax lower bounds in both the random-design and a Gaussian white noise model. For the mean squared error, our novel lower bounds match our upper bounds (up to an arbitrarily small loss in the exponent) and coincide with the convergence rates obtained previously for related holomorphy classes (arXiv:2412.17582). Finally, we apply the theory to the parameter-to-solution map and the parameter-to-quantity-of-interest map arising from the elliptic Darcy flow equation.

Maintaining a stable thermal environment is critical for spacecraft carrying precision optics, as temperature gradients and thermal expansion can alter the geometric alignment of optical components, compromise wavefront sensing, and degrade image quality even when active thermal systems provide real-time control. In particular, the temperature of the optical payload must be controlled to sub-Kelvin tolerances about the desired setpoint, a requirement that is often difficult to achieve with conventional reactive control methods. To this end, this paper develops a novel nonlinear thermal model for the optical system of a spacecraft, accounting for conduction, radiation, and the changing radiative environment induced by the spacecraft's orbit. Using this model, we develop a model predictive control (MPC) methodology that enables closed-loop thermal regulation of the optical compartment under realistic orbital conditions. Since only partial, noise-corrupted temperature measurements are available, an Unscented Kalman Filter is employed to estimate the full thermal state, yielding a certainty-equivalence MPC controller obtained via recursive online optimization. The thermal model is validated against a high-fidelity thermal simulator, and the proposed approach is compared against conventional PID control, the prevailing method for on-orbit thermal regulation.

We construct a deterministic Farey-boundary algorithm converting a cyclic prefix table for an element of $\mathsf T$ into a word in addressed boundary flips. The reduction uses rational boundary pairs and cyclic order, with a marked edge determining the residual modular correction. Its unmarked stage uses at most twice the minimum number of diagonal exchanges. For input size $s$, the complete output has $O(s)$ factors, encoded length $O(s^2)$, and running time $O(s^3)$. Adjoining shuffles extends the construction to $\mathsf V$ with polynomial bounds. We also give explicit boundary generating sets and their local flip–shuffle relations. Changes of generators in the Lochak–Schneps and Bleak–Quick presentations yield a two-flip presentation of $\mathsf T$ and a presentation of $\mathsf V$ with two flips, one shuffle, and nine relations.

Missing covariates are frequently encountered in supervised learning problems, and classical methods for estimation using such data use carefully chosen imputation schemes for missing data, or likelihood approximations that lead to nonconvex $M$-estimation problems. These methods and their relatives are suitable for scenarios in which the covariate distribution is known, and more broadly, have enjoyed tremendous success in linear models. But even in basic nonlinear problems such as logistic regression in moderate dimensions, such methods can experience drastic failure modes when the covariate distribution is unknown. Motivated by the need for reliable alternatives, we consider the problem of parameter estimation in logistic regression with missing covariates. Crucially, we operate in the assumption-lean setting where the covariate distribution is unknown (but bounded). We design a stochastic approximation method that is based on $Z$-estimation with a novel monotone operator, and establish that our algorithm is computationally efficient and achieves provable signal recovery at parametric rates under the hypothesis that covariates are missing completely at random. Our theory sharply characterizes the $\ell_2^2$ risk of the estimator in terms of the missingness profile, accommodating heterogeneous observation probabilities. Importantly, it shows that our method always outperforms the de facto “complete-case” estimator that ignores observations with any missing data. Even in the setting with homogeneous missingness (in which each covariate is observed independently with probability $q$), our bounds exhibit intricate and nonstandard dependence on $q$ that can yield significant improvements over using only complete cases. We complement our upper bounds with new information-theoretic lower bounds that show that this intricate dependence on $q$ is fundamental in a minimax sense.

Chiral four-spin exchange generates a momentum-dependent magnon spin in a compensated honeycomb antiferromagnet at zero magnetic field. For a Néel vector tilted by an angle $\theta_0$ from the layer normal, the interaction contributes to intersublattice magnon pairing already at harmonic order, with an amplitude proportional to $\cos\theta_0\sin^2\theta_0$. Biaxial anisotropy mixes the two longitudinal-spin sectors, producing nonquantized mode spins that are odd in momentum, while a sublattice-exchange relation keeps the magnon energies reciprocal. A temperature gradient then induces a spin density along the Néel axis through a redistribution of magnon occupations. Reversing the four-spin coupling reverses this spin density and the mode spins without changing the magnon energies. MnPS$_3$ motivates the magnetic geometry; realizing the interaction requires a structural modification compatible with the symmetry assumptions of the model.

Recent advances in penalty-based methods for stochastic bilevel optimization (SBO) have eliminated the need for second-order derivative oracles. However, for stochastic nonconvex-strongly convex bilevel problems, existing first-order methods typically rely on nested loops and/or large batch sizes for attaining $O(\epsilon^{-6})$ or $O(\epsilon^{-4})$ sample complexity under standard bounded-variance assumption or mean-square smoothness assumption. Achieving these rates with a single-loop penalty method and a constant batch size remains challenging due to a large penalty value needed for an accurate approximation. To address this challenge, we develop a stochastic SIngle-loop COnstant-Batch first-order penalty method (SICO) that combines two complementary ingredients. First, it performs one stochastic-gradient update per-iteration for both the original lower-level and penalized problems, with a projection that controls the separation between their iterates. Second, it applies an exponential moving average to stabilize the upper-level gradient estimator. We show that this combination achieves $ O(\epsilon^{-6}) $ sample complexity using only $O(1)$ stochastic-gradient samples per iteration under unbiased, bounded-variance stochastic gradients. Under the additional mean-square smoothness assumption on the lower-level stochastic gradients, the same algorithm improves the complexity to $O(\epsilon^{-4})$ also with $O(1)$ batch size. To the best of our knowledge, this is the first work to match the best-known convergence rate for fully first-order SBO methods using a single loop and a constant batch size. This result addresses an open problem posed in the literature.

Manual repair of program failures is time-consuming and disruptive for software developers, particularly during the pre-submit phase where test failures occur within continuous integration systems. While Automated Program Repair has seen significant advancement through Large Language Models, existing state-of-the-art techniques primarily focus on post-submit workflows, operating offline without the low-latency requirements necessary to assist developers in real-time within their flow before they switch context. In this paper, we introduce FlowAgent, an AI agent deployed at Google to automatically repair test failures in the pre-submit outer-loop workflow inside continuous integration systems. Integrated into Google's internal developer tools, Critique and Cider,FlowAgent utilizes a ReAct-style generate-and-validate loop, as well as rigorous pre-execution and post-execution abstention filters to ensure high-quality suggestions under strict latency constraints. Based on our case studies, FlowAgent is highly effective. First, a manual evaluation conducted on 195 real-world test failures demonstrated 67.18% accuracy in suggesting correct fixes. Following its Google-wide deployment, FlowAgent suggested fixes on 295,508changes, of which developers previewed 65,069 and applied 28,554. Developer feedback from interviews indicate that the agent is useful in suggesting correct fixes, integration of autonomous repair agents into industrial software engineering workflows is received well, while interesting challenges and opportunities still remain.

Context: This paper presents a study of the X-ray emission for a sample of radio-detected (RD-) broad- and narrow-line Seyfert 1 galaxies (NLS1s and BLS1s). Aims: To investigate the X-ray emission properties of early-stage accretion activity or early jets in NLS1s. Methods: By fitting XMM-Newton X-ray spectra and combining with multi-band data from SDSS and FIRST, we conducted a statistical comparative study on the X-ray emission and multi-band correlations of NLS1s and BLS1s. Results: NLS1s and BLS1s show no significant difference in their soft X-ray blackbody temperatures, but NLS1s exhibit steeper hard X-ray photon indices and optical-to-X-ray slopes ($\alpha_{\rm ox}$). Moreover, correlation analyses of various parameters reveal intrinsic differences in the X-ray emission between the two samples. For instance, at comparable optical luminosities, NLS1s exhibit a higher $ \alpha_{\rm ox} $ (i.e., weaker X-ray emission), marginally distinct dependencies between radio luminosity and black hole mass in the fundamental plane, and different correlations between $\alpha_{\rm ox} $ and radio loudness. Based on the relationship between UV luminosity and $\alpha_{\rm ox} $ in RQ-AGN, the estimated X-ray excess ($\Delta \alpha_{\rm ox} $) in both samples is positively correlated with radio loudness. RQ-BLS1s generally display the X-ray excess, whereas RQ-NLS1s tend to show an X-ray deficiency. The Eddington ratio of RQ-NLS1s derived from X-ray luminosity is lower than that inferred from optical luminosity, further supporting their intrinsically weaker X-ray emission. Conclusions: These differences in X-ray emission between NLS1s and BLS1s may ultimately stem from the earlier evolutionary stage of NLS1s. Multivariate linear fits among radio luminosity, X-ray luminosity, and radio loudness for NLS1s and BLS1s support a multi-wavelength radiative connection associated with jets.

We study embeddings of factorizations of the $\lambda$-fold complete $h$-uniform hypergraph $\lambda K_m^h$ in factorizations of $\lambda K_n^h$, where $m<n$ and both the source and target degrees may vary by color. From every embedding one can obtain another that minimizes the number of components in every target color without increasing its multiplicity spread. This change also does not increase the sum of any convex function of the edge multiplicities, simultaneously in all colors. For $h=2,3$ we give exact existence criteria for every $m<n$. For $h\ge4$, the necessary divisibility and degree-sum conditions are sufficient once $n\ge(h-1)m$, improving the previously known threshold $n\ge hm$ for color-dependent degrees. In this range we also give exact criteria for connected, simple, and equimultiple target factors. The proofs use a system of integer counts recording the number of added edges of each type. Every system satisfying these equations gives an embedding with the minimum multiplicity spread allowed by

Model optimizations help improve inference performance and accuracy of ML workflows. However, relying on a single model to perform inference across all data batches often fails to maximize accuracy and thus overall performance. In many cases, alternate models could perform better on specific subsets of data where a primary model underperforms. Our experiments with real ML workflows indeed show that switching models improves workflow accuracy by up to 23%. Yet, current systems lack the ability to adaptively switch between models based on performance, forcing users to manually test models in sequence. We present FlexiFlow, a dataflow system that dynamically switches between alternate models when the current model exhibits low accuracy. FlexiFlow learns to rank models using a novel multi-armed bandit approach that accounts for model runtimes, probability of passing user-defined assertions, and the computational structure of the ML workflow. We show that the standard Thompson sampling approach is insufficient for switching models in ML workflows. In contrast, our proposed approaches are effective and scales to complex real-world ML workflows. Experiments show that switching models at runtime while reusing intermediate results provides higher accuracy, but also 48% efficiency gain compared to sequential workflow runs.

We study an online trading problem where a trader, given a sequence of i.i.d. prices drawn from a known distribution $F$ on $[0,1]$, must make irrevocable buy, sell, or hold decisions subject to storage constraints. We investigate achievable algorithmic performance measured in terms of regret, the difference between the expected profit of the hindsight optimal policy that knows the entire price sequence and an online algorithm. We analyze finite atomic and continuous distributions characterized by their local behavior around the median which we capture using a parameter $\beta$. The parameter $\beta$ quantifies how the mass of prices accumulates around the distribution median. We identify a new driver of algorithmic performance, demonstrating that median gaps coupled with an initial inventory level of zero can force regret scaling of $\Omega(T^{(\beta + 1)/(2\beta+4)})$ — establishing a novel spectrum of fundamental limits on algorithmic performance. We then study STARS, short for Storage Trading by Averaging Repeatedly across multiple Scenarios, which simulates possible future price scenarios to approximate the value-to-go function and make buy/sell/hold decisions. We show that STARS obtain near-optimal algorithmic performance (upto poly-logarithmic factors) across a broad range of distributions. In particular, it achieves $O(\log T)$ regret for finite atomic prices, $\widetilde{O}(T^{\beta/(2\beta+2)})$ for continuous distributions without a median gap and $\widetilde{O}(T^{(\beta+1)/(2\beta+4)})$ for continuous distributions with a median gap for $\beta \geq 0$.

Spectral methods are central to high-dimensional statistics and machine learning, underlying procedures for covariance estimation, matrix denoising, representation learning, clustering, and latent variable modeling. In this work, we focus primarily on spectral estimators for high-dimensional factor models, where leading singular vectors of the data matrix are used to estimate latent structure and covariance parameters. In many modern applications, however, data are collected across related but heterogeneous tasks, studies, domains, or populations. Applying such estimators separately to each task can lead to unstable estimates when sample sizes are limited relative to dimension, while completely pooling the data can obscure meaningful task-specific structure. We introduce Hierarchical Spectral Shrinkage (HSS), a scalable empirical-Bayes framework for partially pooling task-specific spectral estimators. The method regularizes the leading spectral directions of each task toward a data-adaptivelylearned common basis, while allowing the amount of shrinkage to vary across tasks and spectral components. The proposed estimator arises as the posterior mean in a surrogate Bayesian regression formulation of the empirical singular vectors. For factor models, HSS yields partially pooled estimators of the group-specific loading spaces and covariance matrices and can be combined with different spectral estimators by replacing their empirical singular vectors with hierarchically regularized counterparts. More broadly, the same construction provides a mechanism for extending spectral estimators to collections of related but heterogeneous datasets. We demonstrate substantial improvements in estimation accuracy and out-of-sample performance over separate, fully pooled, and shared-subspace approaches in synthetic experiments and multi-study gene-expression data.

Analog hardware platforms offer the potential to reduce energy consumption over digital architectures, but in order to succeed, large-scale analog systems must also be able to operate with or recover from the variability of their components. Towards this goal, we derive and demonstrate the lock-in equilibrium propagation (LIEP) training method. LIEP provides local gradient information for each component in an oscillatory network without separate forward and backward sweeps, potentially allowing for in-situ learning capabilities on analog oscillatory hardware platforms. We demonstrate that LIEP can be used both for ab-initio training as well as recovering performance when pre-trained parameters are perturbed. We show that LIEP can be formulated as a three-factor update rule, and suggest that although the method is currently only validated on shallow networks, alternate architectures may allow it to extend to deep and large-scale networks addressing complex tasks.

Protecting critical infrastructure increasingly depends on continuously verifying large IoT fleets against formal security specifications at runtime. Yet the runtime-verification (RV) pipelines proposed for this task are typically single-host prototypes whose monitors read a shared log file, with no resilience to the failures such deployments incur: a crash or overload silently drops events, clock skew corrupts the ordering metric monitors require, a time-triggered "node has gone silent" property cannot fire when the network itself falls silent, and one slow consumer stalls the pipeline. Each failure is silent: the monitor keeps emitting verdicts over a corrupted view. We present RV-Fabric, a resilient delivery layer that carries the hierarchy over two brokers (MQTT for device ingest, a durable stream broker for backend delivery) and re-establishes five continuity guarantees: durable delivery under crashes, a trusted event order, progress under total silence, consumer isolation and flow control under bounded overload, each an invariant conditioned on broker durability. Above the transport, RV-Fabric makes evidence completeness part of runtime-verification semantics: every verdict carries a status (sound, degraded, incomplete or unavailable) derived from delivery gaps, retention pressure and liveness, so an incomplete stream cannot yield an unqualified all-clear. Under controlled fault injection on a containerised testbed, measured against a fault-free oracle using the real MonPoly engine, the shared-log baseline misses six of seven injected incidents, reporting each as an unqualified all-clear, whereas RV-Fabric preserves all seven; removing a delivery mechanism reintroduces silent loss, removing isolation costs only timeliness. Two published critical-infrastructure datasets, water-SCADA and IoT/IIoT, replay end-to-end.

We introduce a stationary, field-free billiard that reproduces the adiabatic confinement mechanism of a magnetic mirror through boundary geometry alone. Rapid transverse reflections generate an approximately conserved action that plays the role of the magnetic moment, producing an effective longitudinal barrier and a nominal loss cone for escape. For the half-width profile considered here, confinement is controlled by a competition between mirror strength and adiabaticity: shortening the profile increases the mirror ratio but also reduces the accuracy of the adiabatic approximation. As a result, the escaping fraction is nonmonotonic and confinement is strongest at an intermediate profile length. We characterize the associated phase-space structure, escape-time statistics, and escape basins, revealing long-lived transient motion and a strong sensitivity of the final escape direction to the initial condition that persists as the uncertainty in the initial condition is reduced.

Let $W$ be an irreducible complex reflection group with reflection representation $V$. A $W$-stable, faithful homogeneous system of parameters $\Theta \subseteq \mathrm{Sym}(V^*)$ of common positive degree $p$ is called a posh hsop; $\Theta$ carries a $W$-representation $U$ if $\Theta \simeq U$ as ungraded $W$-modules. We classify the posh pairs $(p,U)$ for which a posh hsop of degree $p$ carrying $U$ exists. Every such $U$ is a Galois twist of $V^*$. Extending work of Ito and Okada, we deduce that the quotient $S/(\Theta)$ is a permutation module for $W$ if and only if $U \simeq V^*$.

There have been no robust identifications of fully deuterated methanol, CD3OD, in the interstellar medium (ISM) to date. Our goal is continue the search for CD3OD in star-forming regions to determine the deuteration pattern of methanol and set constraints on the deuteration process. We used a sensitive imaging spectral survey of the protostellar system IRAS4A in the NGC1333 molecular cloud performed as part of the large program carried out with the Atacama Large Millimeter/submillimeter Array (ALMA) called Complex Organic Molecules in Protostars with ALMA Spectral Surveys (COMPASS). The column densities of methanol and several of its less abundant isotopologs, including deuterated ones, were derived under the assumption of local thermodynamic equilibrium (LTE). Along with the detection of several methanol isotopologs toward the binary component IRAS4A2, we report the first interstellar detection of CD3OD. The deuteration pattern of methanol in IRAS4A2 is similar to the pattern measured in other low-mass protostars, with an increase in the apparent D/H ratio from the singly to multiply deuterated isotopologs. Chemical models of prestellar and protostellar cores focused on deuteration do not reproduce this pattern quantitatively. This discrepancy is even more exacerbated by the detection of CD3OD. Our LTE modeling of the spectra yields low column density ratios of methanol to its 13C and 18O isotopologs while the 13C to 18O and 18O to 17O ratios are consistent with the isotopic ratios expected in the local ISM. We suspect that the column density of methanol is underestimated. One possible explanation is that the emission is clumpy rather than monolithic. The first detection of a fully deuterated COM highlights the power of deuterium fractionation as a diagnostic tool for investigating the formation and inheritance of organic molecules during the process of star formation.

Compact extraplanetary rovers and micro aerial vehicles require robust state estimation frameworks designed to operate under strict computational constraints in unforgiving environments. Typical solutions involving vision- or LiDAR-based sensing are computationally expensive and vulnerable to environments with perceptual degradation, making them poorly suited for resource-constrained platforms and austere conditions. Alternatively, Frequency Modulated Continuous Wave (FMCW) radar offers both robustness and computational efficiency by directly providing velocity measurements coupled with inherent resilience to perceptual degradation. These factors enable robust estimation, thereby reducing reliance on human operators for supervision to ensure platform safety in challenging environments. In this manuscript, we propose an embedded radar-inertial estimator tailored for low-compute platforms. All sensor drivers, data processing, and aided inertial navigation are performed on a single-core microcontroller, demonstrating its computational efficiency. Flight experiments show translational APE of $0.51-0.82\,\mathrm{m}$ and RPE of $<$3\,% against motion capture, alongside closed-loop flight through an unmodified PX4 stack. The firmware and printed circuit board design are openly available on GitHub at https://github.com/ntnu-arl/embedded_rio and https://github.com/ntnu-arl/embedded_rio-pcb respectively.

Scenario-based MPC is an attractive strategy for chance-constrained motion planning that approximates uncertainty via a finite set of sampled scenarios. As a sampling-based method, scenario-based MPC is sensitive to distribution mismatch. We address this problem in the context of Safe-Horizon Model Predictive Control (SH-MPC) with obstacles governed by switching dynamic modes. From finite mode observations, we construct a confidence set for the unknown categorical mode law and derive a multiplicative domination bound that transfers a Safe-Horizon collision-risk certificate from a selected scenario-sampling distribution to every law in the confidence set. Wasserstein geometry is used to regularize probability reallocation among modes according to the similarity of their induced trajectory predictions, while a collision-risk surrogate biases sampling toward dangerous modes. The resulting certificate explicitly quantifies the additional tightening required under distribution mismatch and exposes the multiplicative conservatism that arises when several obstacle-wise transfer factors are combined

Deployment-time safety of language models is commonly implemented through runtime guardrails such as input moderation, routing, retrieval verification, and output filtering. Existing deployment frameworks provide increasingly capable mechanisms for these functions, but offer limited guidance on how the safety decisions they produce should be explicitly organized, coordinated, and audited. We formulate deployment-time safety as a decision-organization problem with two elements: responsibility-oriented decomposition of safety decisions and explicit coordination among them. We instantiate this formulation in SAFESHIELD, a deployment-time safety system for small language models that organizes four recurring decision responsibilities (admission, routing, evidence, and release) and records committed decisions in auditable Decision Traces. We evaluate SAFESHIELD through mechanism-level experiments, aggregate stage ablations, controlled coordination ablations, and a deployment-oriented stress suite. Mechanism-level results show that the instantiated safeguards provide the capabilities required by the decision process, while aggregate ablations show substantial degradation in end-to-end safety as the surrounding safety organization is removed. More importantly, dedicated coordination ablations preserve the participating safeguard mechanisms while selectively severing their dependencies: removing admission gating substantially increases false release, and withholding upstream evidence from the release decision reduces release accuracy from 96.0% to 69.5%. These results provide system-level evidence that deployment-time safety depends not only on the capability of individual guardrails, but also on how their decisions are organized and coordinated.

Deep-water biological observation is essential for understanding marine organisms and their interactions with the environment. However, conventional optical and acoustic approaches can introduce stimuli that alter animal behavior and bias biological observations. This paper proposes a fluorescence-enhanced whisker array sensing system that pinpoints underwater hydrodynamic sources through local optical readout rather than direct source imaging. Five spatially oriented whiskers, fabricated with nitinol cores and fluorescent urethane shells, are integrated with ultraviolet excitation and a monocular camera. Image enhancement and segmentation are applied to track the whisker deformation. A lightweight convolutional neural network captures temporal and cross-whisker features from 2 s sequences to estimate source localization. Pool experiments achieve a mean spatial localization error of 88 mm, with 73.5 mm in radius and $2.5^\circ$ in angle, across a test region of 600 mm with $\pm30^\circ$. Real-time localization of a moving thruster demonstrates the capability of the proposed method in dynamic scenarios, highlighting its potential for integration into underwater robots for hydrodynamic source detection, localization, and tracking in low-light environments.