Papers

Hierarchical data is ubiquitous in the empirical sciences and most commonly analyzed with generalized linear mixed-effects models (GLMMs), which have interpretable parameters. Bayesian inference for GLMMs yields calibrated uncertainty estimates but requires Markov Chain Monte Carlo (MCMC); the gold-standard No-U-Turn Sampler (NUTS) is slow and must restart for every new dataset, model and prior specification. We introduce MetaBeta, a pretrained amortized model for in-context Bayesian inference over GLMMs. The full prior and model specification are supplied as context at test time, and the model returns an asymptotically exact posterior at the speed of a forward pass. Two set transformers and two conditional neural spline flows mirror the GLMM's hierarchical structure and are trained jointly on millions of realistic simulated datasets with continuous, binary and count outcomes. The flow posterior is refined by Independence Metropolis-Hastings against the unnormalized posterior, yielding tuning-free asymptotically correct inference. Alternatively, MetaBeta warm-starts NUTS, yielding nearly identical inference 3.5-13x faster than its cold-started counterpart, while removing over 98% of its divergent transitions. On controlled benchmarks with ground-truth parameters, MetaBeta matches NUTS in parameter recovery, posterior calibration and out-of-sample prediction; on real datasets with unknown parameters, priors and model structure, its posteriors closely agree with those of NUTS; and it remains faithful where amortized models typically struggle: under misspecified likelihoods, misspecified priors, out-of-distribution predictors, collinear designs and data-poor regimes. On real data, a full prior sensitivity analysis including model comparison takes a few seconds with MetaBeta against 14-18 core-hours for NUTS, tracing its posteriors throughout. Our model is open-source and open-weights.

In this paper, we study standard and distributionally robust $\mathcal{H}_2$ synthesis problem of a stabilizing state-feedback controller for discrete-time linear time-invariant systems. Without requiring a nonsingular disturbance controllability Gramian or a positive-definite control penalty, we establish that stationarity of a stabilizing gain in standard $\mathcal{H}_2$ synthesis is equivalent to global optimality and derive an exact Pythagorean identity for the performance difference. We then generalize the $\mathcal{H}_2$ synthesis problem by considering i.i.d. disturbances with zero mean and uniformly bounded second moments, and extend the stationarity-optimality equivalence to the corresponding distributionally robust $\mathcal{H}_2$ synthesis. Moreover, we show that every standard $\mathcal{H}_2$-optimal gain is simultaneously optimal for every such distributionally robust problem. Finally, we specialize the framework to Frobenius, Kullback–Leibler, and Wasserstein-2 ambiguity sets and obtain explicit characterizations of their worst-case covariances, which, in turn, provide performance certificates for a common optimal controller.

Let $f\colon X \to Y$ be a proper and tame morphism of algebraic stacks, where $X$ and $Y$ are locally of finite type over an algebraic stack $S$. We prove that $R f_*$ sends complexes that are pseudo-coherent relative to $S$ to pseudo-coherent complexes relative to $S$. In the scheme case, this resolves a conjecture of Illusie from SGA6. We also prove related and new results in the non-tame setting (e.g., infinite stabilizers). Our methods use derived algebraic geometry and also give new proofs of classical statements for schemes due to Kiehl. Along the way, we extend some foundational results for quasi-coherent sheaves on algebraic stacks to the derived setting.

Current experimental scientists increasingly rely on simulation-based inference (SBI) to invert complex models with intractable likelihoods. A primary goal in these settings is to obtain credible regions with valid coverage. While recent model-agnostic conformal calibration methods have succeeded in constructing credible sets with prescribed local Bayesian coverage, their approximate nature introduces inherent epistemic uncertainty in the calibration process. In this work, we propose a novel tool for diagnosing calibration uncertainty. Our approach is based on a simple three-way hypothesis testing procedure. We demonstrate how this tool can be used to effectively assess necessary simulation budgets for calibration sets and analyze the epistemic uncertainty associated with cutoff estimation.

We prove a recursive formula for construction of generalized spectral basis. Example of application is presented for symbolic computation with non-diagonalizable matrix that describes electromagnetic wave propagation in anisotropic media. The method is general enough and can be applied, for example, to multivectors of the Clifford algebra.

Despite advances in long-context inference, large language models (LLMs) remain fundamentally limited by the key-value (KV) caching mechanisms that are necessary for stable computation. Techniques such as selective token eviction and pruning have vastly mitigated these issues, but often discard core information to manage the growing cache. In this paper, we propose Attention with Routed Memory (ARM) a novel KV caching structure that introduces a fully differentiable, fixed-size memory system organized as a hierarchical router. Via a Gumbel-Softmax, ARM learns to select memory slots and perform sigmoid-gated updates that softly combine new and stored information, avoiding hard eviction and reducing information loss. By further training a policy to dynamically select varying amounts of memory at inference, ARM adapts its accesses for both simple contexts and inputs that require deeper reasoning, enabling more scalable and effective retrieval on both short- and long-contexts. Experimental results on standard commonsense and long-context reasoning benchmarks demonstrate that ARM achieves superior performance and efficiency compared to fixed KV-caching approaches, while remaining efficient and scalable in terms of both memory and generation latency.

We study global strong solutions and their large-time behavior for the one-dimensional outer pressure problem of a viscous, heat-conducting ideal polytropic gas. The viscosity and heat conductivity satisfy $\mu=\tilde\mu\theta^\alpha$ and $\kappa=\tilde\kappa\theta^\beta$. For each prescribed nonnegative pressure and each fixed $\beta\geq0$, we allow large $H^1$ initial data with positive specific volume and temperature, provided that $\alpha\geq0$ is sufficiently small. When the limiting outer pressure is positive, the solution has uniform bounds and converges to a stationary state. When the limiting outer pressure is zero and the stated weighted pressure conditions hold, the normalized solution converges to an expanding state at an algebraic rate in physical time.

We investigate field-induced excitation production from the unit-filled Mott background of a one-dimensional dipole-conserving Bose–Hubbard chain driven by a periodic quadratic potential, realizing a time-dependent rank-two electric field. Unlike protocols that manipulate pre-existing dipolar or fractonic excitations, we address their production from an initially Mott-like state without prepared dipolar or fractonic excitations. A single correlated hop directly nucleates the compact $|030\rangle$ configuration at the atomic energy $3U$, while a lower manifold derived from $2U$ contains spatially resolved dipole–antidipole configurations. Exact diagonalization shows that finite hopping turns this lower manifold into a dispersive band whose finite-chain edge lies substantially below the compact $3U$ scale. Despite this lower energetic edge, its normalized odd-harmonic spectral weight is perturbatively suppressed as $(J/U)^2$, reflecting virtual hybridization of the initial and final eigenstates. A consistent second-order treatment of the $|1111\rangle\leftrightarrow|0220\rangle$ channel includes diagonal self-energy shifts and predicts a weak-field resonance displacement confirmed by exact time evolution. Extended-chain dynamics shows genuine dipole–antidipole separation together with redistribution into higher-energy many-body sectors. The resulting mechanism provides a condensed-matter analogue of multiphoton dynamical Schwinger production in which exact dipole conservation replaces mobile opposite charges by mobile opposite dipoles, while isolated fractonic charges remain immobile.

We study necessary and sufficient conditions on a Banach space $X$ such that $L(Y,X)$ is transfinite (rigid) octahedral for any Banach space $Y$. We present several examples of such $X$. As application of our results, we find the first example in the literature of Banach space which is rigid octahedral but fails any transfinite version of octahedrality. We also improve some previous results about octahedrality in ultrapower spaces.

In this work, we propose a haptic update control law that uses reaction torques to correct axial misalignment during robotic valve manipulation. Unlike vision-based estimates, which can be affected by calibration errors, occlusion, and uncertainty in the contact geometry, reaction torques arise directly from the physical interaction between the gripper and valve. A geometric relationship exists between the error (misalignment) vector and these torques. The primary aim of this work is to propose a stable controller exploiting this geometric property. Our control law is proven to be uniformly asymptotically stable. Simulations are performed for verification. Furthermore, we experimentally test the robustness of our method using a Kinova Gen3 robotic arm for initial misalignments ranging from $-15^\circ$ to $15^\circ$ at 3 different valve positions and report the resulting data distribution. The absolute value of the median misalignment across all 18 test cases is found to be within $2.46^\circ$ and that of reaction torques within $0.23\mathrm{Nm}$.

For a positive integer $n$, let $$S_0(n)=\{1,2,\ldots,\lfloor n/2\rfloor\},\qquad S_{j+1}(n)=\{n\bmod k:k\in S_j(n)\setminus\{0\}\},$$ and put $s_j(n) := |S_j(n)|$. The sets defined above arise naturally in the study of the length of the Pierce series expansion of a rational number. In \cite{Ba-Vu}, Baraskar and Vukusic conjectured that for every fixed $j\geq 2$, the limit $$\lim_{n\to\infty} s_j(n)/n$$ does not exist. In this paper, we prove this conjecture in the affirmative. Moreover, we define a new class of iterated remainder sets $T_j(n)$ that naturally arises from the study of the length of the Engel series expansion of a rational number. We analogously study the asymptotic behavior of $t_j(n) := |T_j(n)|$. We show that $$\lim_{n\to\infty}\frac{t_j(n)}n$$ exists precisely for $j\in\{0,1\}$ and fails to exist for every fixed integer $j\geq2$.

Egocentric video offers a scalable source of physical interaction experience, yet translating it into robot-executable knowledge and enabling continual adaptation remain challenging. We introduce Zeva-Ego, a unified framework that learns physical priors from human experience and evolves through robot interaction. An Action-Centric Encoder (ACE) converts egocentric visual transitions into action-centered supervision for VLA mid-training, while In-Context Causal Learning (ICCL) enables parameter-free adaptation from action-effect feedback at deployment. Scaling Ego data to 10K hours improves RoboTwin success from 63.8% to 75.3%, matching 2K hours of robot demonstrations (74.7%), corresponding to an empirical data ratio of roughly 4-5:1. With accumulated interaction experience, ICCL further improves success from 58% to 89% within four attempts without parameter updates. These results demonstrate a scalable path toward embodied intelligence that learns from human experience and continuously improves through its own interaction.

Speech-to-text engines are extremely needed nowadays for different applications, representing an essential enabler in human-robot interaction. Still, some languages suffer from the lack of labeled speech data, especially in the Arabic dialects or any low-resource languages. The need for a self-supervised training process and self-training using noisy training is proven to be one of the up-and-coming feasible solutions. This article proposes an end-to-end, transformers-based model with a framework for low-resource languages. In addition, the framework incorporates customized audio-to-text processing algorithms to achieve a highly efficient Jordanian Arabic dialect speech-to-text system. The proposed framework enables ingesting data from many sources, making the ground truth from external sources possible by speeding up the manual annotation process. The framework allows the training process using noisy student training and self-supervised learning to utilize the unlabeled data in both pre- and post-training stages and incorporate multiple types of data augmentation. The proposed self-training approach outperforms the fine-tuned Wav2Vec model by 5% in terms of word error rate reduction. The outcome of this work provides the research community with a Jordanian-spoken data set along with an end-to-end approach to deal with low-resource languages. This is done by utilizing the power of the pretraining, post-training, and injecting noisy labeled and augmented data with minimal human intervention. It enables the development of new applications in the field of Arabic language speech-to-text area like the question-answering systems and intelligent control systems, and it will add human-like perception and hearing sensors to intelligent robots.

Multi-task visual grounding requires models to jointly understand linguistic semantics and perform accurate visual localization and segmentation. Despite the success of multimodal large language models, effectively adapting them to multiple grounding objectives remains challenging. Existing methods commonly enforce task cooperation through shared representations, while overlooking the intrinsic conflict between task-oriented feature interests. In this paper, we introduce $\textbf{DeCo}$, an efficient $\textbf{De}$couple-to-$\textbf{Co}$uple learning framework that resolves this dilemma through a two-stage paradigm: task-specific representation decoupling followed by complementary prior coupling. Specifically, we first propose Task-aware Semantic Decoupling (TSD) to route shared visual cues into individual features under salient word-level guidance, alleviating representation interference between localization and segmentation. Furthermore, we observe that segmentation naturally provides informative localization priors due to dense supervision. Based on this insight, we introduce Hybrid Prior Coupling (HPC), which integrates sentence-level semantic prior with mask-derived spatial prior for enhanced grounding. Built upon a frozen multimodal encoder, DeCo requires lightweight trainable parameters while achieving strong generalization across multiple grounding objectives. Extensive experiments on RefCOCO/+, G-Ref, ReferIt, Flickr, DIOR-RSVG, SARVG1.0, RRSIS-D, RIS-LAD, and RefDIOR demonstrate that DeCo achieves state-of-the-art performance on both natural and remote sensing benchmarks. The code and models are available at https://github.com/xiaoqiang-lu/DeCo.

The exact behaviour of the enveloppes of the single-time and two-time correlators is found for the voter model on a fractal substrate, with nearest-neighbour interactions. Herein the geometry of the fractal substrate is described by its non-integer geometric fractal dimension $d_f$, and its topology and diffusive transport by the distinct spectral dimension $d_s$. On the level of the equations of motion of the correlators this can be modelled by considering a space-dependent diffusion constant ${\cal D}(r)\sim r^{-\theta}$ which implies the spectral index $\theta$, itself a function of $d_f$ and $d_s$. With a scaling ansatz, the generic phenomenology of ageing is confirmed and the dynamic exponent ${z}=2+\theta$ and the autocorrelation exponent $\lambda=d_f$ are derived. The explicitly found dynamic scaling functions are shown to depend only on the spectral dimension $d_s$. The decay of the enveloppe of the density of active interfaces with time is described by the exponent $\alpha=1-d_s/2$ for $d_s<2$, confirming the results of preexisting numerical simulations.

AI-generated overviews are becoming an increasingly prominent layer of search interfaces, yet their behavior in Chinese-language search remains underexplored. We conduct a cross-lingual audit of AI overview behavior on Baidu and Google using English queries sampled from MS MARCO and their translated Chinese counterparts. Our analysis examines when overviews are triggered across platform-language settings, which host domains receive visible exposure in Chinese-language overviews, how concentrated that exposure is, and how source overlap varies across settings. We also compare the embedding-based semantic similarity of generated answers for matched query intents. The results reveal substantial differences across platform-language settings in overview availability and visible source exposure. At the aggregate level, the settings exhibit low overlap in visible host-domain inventories, while matched-query answers yield median cosine similarities ranging from 0.701 to 0.813. These findings indicate that answer-level semantic similarity and aggregate source exposure capture distinct dimensions of AI-mediated search. Evaluations of AI search should therefore consider not only the content of generated answers but also how source visibility is distributed across platforms, languages, and information environments.

In this paper, we establish an upper bound for the second eigenvalue of the scalar Jacobi operator of an arbitrary closed two-sided surface immersed in a contracted Berger sphere. The estimate involves the total squared mean curvature, the Euler characteristic, and an explicit nonpositive term determined by the angle between the surface normal and the Hopf direction. The proof combines the realization of a Berger sphere as a geodesic hypersurface of a complex projective plane, the first standard embedding of the latter into a Euclidean sphere, a weighted Hersch–Li–Yau balancing argument, and the conformal invariance of the Willmore functional. No minimality or constant mean curvature assumption is imposed. As an application, if $1/3\leq\alpha\leq1$ and the surface has nonpositive Euler characteristic, then its second Jacobi eigenvalue is nonpositive; it is strictly negative for $\alpha>1/3$. At the critical value $\alpha=1/3$, equality forces the immersed image to be congruent to the minimal Clifford torus; in the embedded category, this yields a complete characterization of the equality case.

Abridged. Mounting observations of transiting dusty debris orbiting WDs display complex signatures that often feature orbital period drifts. The asteroid or planet progenitor of this debris might still be partly intact on a near-circular orbit that is inclined or coplanar, with a period that is similar to that of the dust. Alternatively, the debris could persist concurrently with other intact asteroids or planets that had tidally circularised around the WD at a later time, but still remain directly unobservable. We sought to first improve the identification of this detectability threshold by expanding our previous investigation of the WD-secondary-dust co-orbital circular restricted three-body problem (CRTBP) to higher secondary masses corresponding to Ceres, Mercury, Earth, and Neptune. Next, we continued to explore the link between the analytic and numerical characterisations of the CRTBP. First, we computed families of periodic orbits in the 1/1 mean-motion-resonance (MMR) in both the planar and spatial CRTBP cases, where the dust is on an arbitrarily eccentric orbit. Second, we generated global phase space portraits. Third, we determined observational links with debris located in very close proximity to WDs via N-body simulations. For each method we applied, we report a novel set of results. These include 3 newly discovered bifurcation points, $B_T^4, B_{cs}^4$, and $B_{cs}^5$, 2 branches of asymmetric periodic orbits, called $A_2$ and $A_3$, at high dust eccentricity regime in the 2D-CRTBP, and 2 families of symmetric orbits in the 3D-CRTBP. The maps reveal stability domains as a function of the dust's eccentricity, argument of pericentre, and slight semi-major axis deviations from the 1/1 MMR. The N-body simulations reveal that over 10 yr, mass analogues of Mercury, Earth, and Neptune generate orbital period deviations in the dust of, respectively, 1-10 s, 10-100 s, and over 100 s.

Ahead-of-time (AOT) compilers (e.g., revmc, evmone, and DTVM) for the Ethereum Virtual Machine (EVM) reuse compilation artifacts at contract-code-hash granularity. This granularity is poorly matched to real EVM workloads dominated by contract families: factory-, proxy-, and template-driven deployments that share instruction structure but differ in a small set of embedded constants. Across four EVM chains (Base, Ethereum, BSC, and Arbitrum), we find that 23.1–47.6% of unique compilable bytecodes map to shared family skeletons within 10K-block windows. Per-hash AOT therefore redundantly recompiles structurally equivalent code, inflating compile time and artifact footprint while reducing workload coverage under finite compile budgets. We present SkelOT, an AOT framework that lifts the unit of compilation reuse from code hash to family skeleton. SkelOT compiles one native artifact per family, bakes invariant constants into the artifact, and reads variant constants from a per-contract runtime table. Built on revmc/LLVM and evaluated on a 10K-block Base mainnet corpus (3.52M transactions), SkelOT reduces compilation units by 47.5%, artifact footprint by 57.4%, and compile time by $2.19\times$, while preserving byte-identical execution outcomes versus per-hash AOT. At runtime, SkelOT delivers a $1.31\times$ median per-contract speedup across family members. Under a compile budget targeting 75% execution-time coverage, SkelOT needs far fewer artifacts than per-hash AOT, and the advantage holds at every coverage target.

Biological visual systems achieve continuous, low-latency motion perception by processing sparse, asynchronous spiking signals, enabling real-time tracking under strict energy constraints. Event-based cameras, inspired by the mammalian retina, replicate this efficiency by capturing only local brightness changes as asynchronous events, offering a natural substrate for spiking neural networks (SNNs) to parallelise computation and adapt to fast-changing scenes. Pinball provides a controlled yet dynamic testbed, requiring precise motion estimation and fast reaction to a small, rapidly moving target. This work presents a fully spiking, real-time perception-to-action pipeline for closed-loop pinball gameplay. A dynamic vision sensor observes a small, fast-moving ball, and a network of spiking Time-Difference Encoders on the SpiNNaker neuromorphic platform jointly estimates its position, speed, and direction. The system is characterised across receptive field size, accumulation window, and angular tuning width for real-time operation, and benchmarked in closed loop against human players across two flipper regimes of increasing physical realism. It achieves a hit rate of 56.1%, nearly double the human average, reacting within 21.7 ms (5 ms network latency) and consuming an estimated 148 {\mu}W using fewer than 25k neurons, among the fastest and most energy-efficient event-based closed-loop demonstrators benchmarked. Under more realistic flipper dynamics, tuning a single interpretable policy parameter reproduces the full spectrum of human play styles, from cautious to aggressive, with no change to the perception pipeline. A physical demonstrator, tracking a real ball and actuating real flippers in closed loop, confirms the principle operates beyond simulation. Its fully spiking, learning-free design offers a compact, energy-efficient example of real-time neuromorphic perception-to-action.

The logistic map is a quintessential model in the study of low-dimensional chaos. High-dimensional chaos, on the other hand, presents itself in disordered systems with many interacting and heterogeneously coupled components. Here, we formulate a system of many logistic maps, interacting through disordered couplings. Using a combination of dynamic mean-field theory, random matrix theory and numerical simulations, we show that even the smallest amount of disorder can remove the period-doubling cascade in the conventional logistic map. Instead we find a transition to high-dimensional chaos, marked by an oscillatory instability not previously reported for disordered systems. We also show that with sufficiently strong homogeneous coupling between the maps one recovers elements of the conventional period-doubling cascade. Our findings indicate that well-known phenomena in dynamical systems can be fragile in the face of disorder. At the same time, new phenomena emerge that are neither found in simple low-dimensional dynamics nor in high-dimensional disordered systems.

Structured pruning is a model compression technique that is used to reduce the computational cost of deploying deep neural networks on resource-constrained devices. Popular methods of pruning rely on opaque heuristics or weight-based criteria that give no indication as to the structural dependencies in the network. To address these limitations we present Artificial Structure Function Search (ASF-S): a novel structured pruning framework. ASF-S utilizes Principle Gradient Importance (PGI): a novel prune-candidate selection criteria that is inspired by structure-function relationships in the brain. By ensuring the pruned structure of the model respects topographical organization of the output layer, we define Artificial Functional Connectivity (AFC) for artificial neural networks. AFC provides evidence to demonstrate that accurate smaller networks can be found using careful prune candidate selection criteria. We present results for PGI as a selection criterion and for ASF-S as a pruning framework against recent benchmarks, demonstrating that our method yields model variants with 70% parameter reduction, that can recover baseline accuracy without re-training the pruned layers.

We investigate the Joule Thomson expansion of a symmergent black hole within an extended phase-space formalism in which the thermodynamic pressure is introduced as an independent variable. The symmergent entropy contains the parameter alpha-hat, which modifies the enthalpy and thermodynamic volume relative to their standard forms. We show that the Joule Thomson coefficient depends only on the combination 8piP*r_+^2, leading to an inversion curve P_i = (pi/2) * I_i^2 that is independent of alpha-hat. The isenthalpic curves, however, retain a parametric dependence on alpha-hat, indicating that the symmergent sector affects the thermodynamic trajectories but not the inversion boundary itself. The inversion pressure and temperature are derived explicitly, and the cooling regime P > P_i is distinguished from the heating regime P < P_i. Since the effective geometry is asymptotically AdS-like for positive pressure, the results also admit a holographic reading, in which the Joule Thomson expansion corresponds to an isenthalpic process in the dual field theory. The results provide a consistent thermodynamic framework for symmergent black holes and clarify the role of the symmergent parameters in the Joule Thomson process.

Heffter arrays, introduced by Archdeacon in 2015, are combinatorial structures with applications to cyclic cycle systems and biembeddings of graphs on surfaces. Costa, Morini, Pasotti and Pellegrini proposed in 2020 the notion of relative Heffter arrays as a generalization of classical Heffter arrays, inspired by the concept of relative difference families. In their work, the existence problem of integer relative Heffter arrays $\mathrm{H}_k(n;k)$ was solved for all $k\neq 5$, while the case $k=5$ and $n \equiv 0 \pmod{4}$ remained open, apart from two sporadic examples with $n=8,16$. In this article, we consider this open problem and we construct an $\mathrm{H}_5(n;5)$ for every $n\equiv 0 \pmod{4}$, $n\geq 12$. As a consequence, the existence problem of integer relative $\mathrm{H}_k(n;k)$ is completely settled.

Coordinated Inauthentic Behavior (CIB) has become a major concern in online social platforms, yet its actual impact on information diffusion remains poorly understood. Existing research has primarily focused on detecting coordinated activity, while comparatively little attention has been devoted to quantifying its influence once detected. In this work, we introduce two complementary frameworks for the post-hoc evaluation of coordinated accounts. First, we formulate the problem on information cascades as a constrained influence maximization problem over directed trees and develop a polynomial-time dynamic programming algorithm that computes the optimal placement of coordinated nodes, providing an upper bound on their achievable influence. Second, motivated by the limited availability of diffusion cascades in real-world platforms, we propose a network-based framework that estimates influence directly from retweet networks using the independent cascade model and compares the observed placement of coordinated accounts against established heuristic baselines. We evaluate both approaches on Twitter/X data from the 2019 UK General Election and on a collection of verified state-backed information operation campaigns spanning multiple countries. While coordinated accounts exhibit limited influence in the UK cascades, the network-based analysis reveals substantial differences across campaigns, with several operations achieving influence comparable to or exceeding that of structurally central seed sets. Finally, by reconstructing cascades from the retweet networks, we show that the two frameworks produce consistent results, suggesting that the observed effects reflect intrinsic structural properties of coordinated activity rather than artifacts of the underlying methodology.

The conventional testing process quickly fails when applied to ML-based systems such as obstacle detection in vehicles: if an obstacle is not detected in a test, classical bug fixing is impossible and an AI system will always retain shortcomings. Test results can therefore only be interpreted statistically, which in turn requires test sets that are not only complete with respect to the operational design domain (ODD) of the system, but also representative of it. To this end, we introduce probabilistically extended ontologies (PEONs): ontologies describing the ODD, augmented with a probability distribution over the partitioning they induce. Instead of unmaintainable conditional probability tables, only marginal distributions and functionally described dependencies need to be specified; algorithms based on couplings and optimal transport complete this specification to a Bayesian network. From a PEON we derive the sampling of representative test cases, rigorous end-of-test criteria for given quality targets and significance levels, and methods for re-evaluating existing test results and for assessing the balance of training data. We demonstrate the practical modelling of a complex ODD using the example of automatic train operation.

We study a stochastic Navier–Stokes system in which the spatial modes of the transport noise evolve dynamically and are coupled to the resolved velocity. The model couples a Navier–Stokes stochastic PDE to an infinite family of linearized Navier-Stokes equations for the noise correlation modes. These latter equations may contain a hyperviscosity. We prove existence of martingale solutions in two dimensions for hyperviscosity exponent $s\geq1$, and pathwise uniqueness for $s>1$ under additional regularity of the initial noise modes. In three dimensions, we establish existence of martingale solutions for $s>3/2$. The proof combines Galerkin approximations and compactness with estimates adapted to the coupled drift and stochastic terms.

Instrumental quality models for video quality prediction are usually trained for a fixed set of codecs or other input features, and every new input variable requires retraining. Novel, general-purpose decision models can answer questions without task-specific training, but it is unclear whether they can judge video quality. We evaluate Jev, a commercial “System One” model that returns probability distributions over a provided answer scale, as a zero-shot video quality model. We call the resulting method JEVQA. In a first study on 1,936 AV1, H.264, HEVC, and VP9 encodes of 22 sources, scored against VMAF as ground truth, using encoding metadata only, JEVQA reached a Pearson correlation of 0.737, on par with the standardized ITU-T P.1204.1 model (0.733). Giving the model bitstream data raised the accuracy to 0.797, and combined pixel-based and bitstream features raised it to 0.824. A pixel-only variant failed in our tests. In a second study, using H.264, HEVC, and VP9 encodes in the AVT-VQDB-UHD-1 database, the metadata-only model reached a correlation of 0.879 with MOS, close to P.1204.1 (0.898). Bitstream statistics did not help there. Our results show that trained models on the same features remain clearly ahead in both studies, but that zero-shot classifiers are promising.

Climate variability influences whether a market disruption escalates into a food crisis, yet broad climate patterns like El Niño, tracked months before they alter hydro-climatic conditions, are still not incorporated as an early-warning component in food-security responses. We address this gap by introducing sensitivity regimes, a stratification of regions by the direction and strength of their vegetation response to the El Niño Southern Oscillation, and using them to estimate how food price spikes affect acute food insecurity across sub-Saharan Africa. Integrating remote sensing, socioeconomic data, and causal machine learning, we find that in regions where ENSO systematically suppresses vegetation, a price spike raises the share of the population at acute risk by 5.4 percentage points in the following month. In regions where vegetation is unaffected by or positively linked to ENSO, the estimated effect is smaller (around 2 percentage points) and statistically insignificant. These results demonstrate that climate context is critical for understanding food security vulnerabilities. Sensitivity regimes can be combined with operational price-spike triggers to stage anticipatory action: the ENSO state flags vulnerable regions months ahead, and a pre-positioned response in those regions to a price spike would avert the largest jump in acute food insecurity.

This paper evaluates a passive Hybrid Network-based Intrusion Detection System (Hybrid-NIDS) prototype that combines Suricata with Random Forest flow classification and centralized ELK-based alert handling. The study explicitly separates benchmark evaluation from PCAP/live operational validation and controls exact feature-duplicate leakage using feature hashing and group-aware splitting. From 2,540,047 UNSW-NB15 records, 453 conflicting-label hash groups containing 1,879 rows were removed; the resulting Development and Hold-out sets have zero exact feature-hash overlap. RF-41 achieved F1 = 0.971360 and ROC-AUC = 0.999671, while the NFStream-compatible RF-21 achieved F1 = 0.970148 on the same prepared hold-out boundary. However, operational validation revealed substantial benchmark-to-deployment domain shift: on a labeled laboratory PCAP, RF-21 and the strictly correlated branch achieved recall of only 0.0095, and RF-21 produced no alerts in five additional 60-second attack sessions. An unlabeled normal-traffic test produced 439 alerts from 2,375 flows; this value is reported only as an alert ratio and is not interpreted as a false-positive rate. These results show that strong performance on a public benchmark does not directly translate into operational effectiveness. Accordingly, the current Hybrid-NIDS should be interpreted as a passive prototype and evaluation framework, and the reported experiments do not demonstrate that Suricata-Random Forest correlation provides better operational detection than Suricata alone.