Peer Revenue Publish or profit

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Monday 7 September

Dynamical Systems

Semi-global exact prescribed-time stabilization of linear systems by bounded linear time-varying feedback

This paper addresses the problem of semi-global exact prescribed-time stabilization of linear systems by bounded controls. To solve such a problem, one needs to construct a parameterized quadratic control Lyapunov function whose corresponding level set can be made arbitrarily large (for achieving semi-global stabilization) and such that the real parts of the poles of the closed-loop system approach minus infinity as the parameter approaches infinity (for achieving prescribed-time stabilization). This objective is achieved by proposing a nested parametric Lyapunov equation (PLE)-based approach, through which only two linear matrix equations need to be solved in the controller design stage without using real-time state information. Under the proposed structural condition, the nested PLE-based approach guarantees semi-global exact prescribed-time stabilization for any prescribed bounded set of initial conditions, provided that the prescribed settling time is sufficiently large. A numerical example is provided to demonstrate the effectiveness of the designed controller.
Read the full abstract

Cryptography and Security

Merging Cyber Threat Intelligence Through Retrieval-Augmented Generation and Small Language Models for Rich Threat Representation

Modern cybersecurity operations rely on CTI collected from heterogeneous sources, including semi-structured threat representations, IoCs, and narrative technical reports. However, these artifacts are often insufficient in isolation to reconstruct how an attack unfolds, under which conditions each step is feasible, and which traces it leaves behind. In practice, analysts must manually correlate partial evidence scattered across multiple and only partially structured sources, delaying the design of effective prevention, detection, and response actions. To address this gap, we propose an automated pipeline that derives an actionable representation of a cyberattack from heterogeneous CTI sources. The pipeline combines a RAG architecture with a locally deployable SLM, used to consolidate such evidence and infer missing operational details. Starting from a semi-structured threat representation and auxiliary CTI documents, the pipeline produces an enriched Attack Graph that captures a coarse, tactic-aligned progression of the attack and annotates each step with explicit pre-conditions and post-conditions, and an enriched description. This representation supports prevention by exposing execution requirements, detection by highlighting observable traces, and response by clarifying the temporal progression of the attack. Then, due to the lack of validated datasets with ground-truth information on the temporal evolution of real-world attacks, we test the complete pipeline on 10 real-world case studies spanning multiple threat types, including backdoors and staged downloaders delivered via phishing. A manual assessment across 10 real-world case studies provides initial evidence that the generated graphs are consistent with expected attack progressions, indicating that the proposed approach can support analysts by consolidating dispersed CTI evidence into a structured and actionable view of attacks.
Read the full abstract

All authors Nicola Deidda, Leonardo Regano, Alessandro Sanna, Davide Maiorca, Giorgio Giacinto

Analysis of PDEs

Generalized functions in a Non-Archimedean field

Ultrafunctions are generalized functions defined on a non-Archimedean field extending R. They provide a framework for treating problems that fall outside the classical theory of distributions. In this paper we introduce an improved notion of ultrafunctions which refines the previous constructions. This new approach allows a more flexible functional framework than the one used in other papers and yields a more flexible functional setting. In particular, this approach allows us to prove the existence of ultrafunction solutions for several classes of ill-posed evolution problems arising in partial differential equations.
Read the full abstract

Earth and Planetary Astrophysics

The effect of the atmospheric C/O ratio on interiors of hot Jupiters

The atmosphere is the outer boundary of a gas giant exoplanet's interior. Therefore, compositional changes in the atmosphere can affect interior inferences. Typically, only metallicity is considered in atmospheric boundary conditions, while other elemental ratios, in particular the C/O ratio, are assumed to be solar. In light of the observational constraints now achievable with JWST, this assumption might no longer be justified. In this work, we investigate the effect of the C/O ratio on atmospheric boundary conditions, interiors and radii of hot Jupiters. We construct an atmospheric boundary grid, consisting of one-dimensional atmospheric models in radiative-convective and thermochemical equilibrium. This grid is coupled to a static interior structure model at the radiative-convective boundary (RCB). Because in the temperature regime of hot Jupiters, the C/O ratio determines the dominant species in the atmosphere, it can significantly alter the opacity and, consequently, the pressure and temperature at the RCB. This temperature change propagates into the convective interior, thus affecting the calculated radius. The radius difference relative to a solar C/O-atmosphere significantly exceeds observed radius uncertainties of 3 per cent for planets with equilibrium temperatures $\gtrsim$ 1500 K and super-solar atmospheric metallicities. As an example, we demonstrate that, for WASP-19~b, higher C/O ratios result in higher inferred intrinsic temperatures. These results highlight the importance of treating atmospheric composition and interior structure as a coupled system. As atmospheric constraints improve, incorporating measured C/O ratios into atmospheric boundary conditions is necessary to obtain robust inferences of hot Jupiter interiors.
Read the full abstract

Mathematical Physics

Renormalization group bounds on Araki relative entropy

This paper derives a multiscale Bakry-Émery criterion for an upper bound on the Araki relative entropy of perturbations of an equilibrium state in relativistic QFT. Using stochastic positivity, affiliated perturbations of the relativistic bosonic field can be put in correspondence with perturbations of the massive Euclidean free field. This gives a Feynman-Kac formula, which in turns is used to derive a probabilistic expression for the Araki relative entropy. The description of the massive Euclidean free field as an abstract Wiener space provides the infinite-dimensional framework to derive a Polchinski equation. Extending the methods of Bauerschmidt, Bodineau, and Dagallier to this context yields a probabilistic bound on the Araki relative entropy, which can be formulated entirely in operator-algebraic terms. Using the criterion, it is shown that the Lorentzian massive Sine-Gordon model in the ultraviolet finite regime satisfies the relative entropy bound.
Read the full abstract

Information Retrieval

Matryoshka Hash Representations for Model-Aware Compact Semantic Retrieval

Retrieval-augmented generation (RAG) depends on dense retrieval: each document is stored as a learned vector, and queries are answered by nearest-neighbor search. Keeping one full-precision vector per document dominates index cost at corpus scale, so systems replace each vector with a short code of a few bytes (quantization). Standard quantizers such as product quantization (PQ) pick the code that best reconstructs the original vector, but reconstruction is only a proxy for retrieval: at a few bytes per document, a model-aware code trained on the retriever's ranking signal uses the budget better. A single code is even more useful if its short prefixes are each directly searchable, letting a deployment set its efficiency-quality operating point without re-encoding the corpus. But training all prefixes under one objective makes the early bits a compromise across budgets: short codes improve while the full-width code degrades. Binary codes sharpen the conflict, since a gradient step either flips a stored bit or leaves it unchanged. We introduce Matryoshka Hash Representations (MHR), a two-stage procedure that separates full-width training from prefix organization. MHR first learns a longer binary code, then freezes the model and trains zero-initialized residual code adaptors for directly searchable prefixes. Documents are stored at one bit per coordinate, queries keep continuous logits as in PQ, and search runs on FAISS FastScan. Trained on MS MARCO and zero-shot transferred to seven BEIR datasets, one MHR code reaches 0.502 NDCG@10 at 16 bytes and 0.556 at 32 bytes, against 0.404 and 0.524 for the best compact baselines; the gains hold across training seeds, on all 13 public BEIR datasets, and against off-the-shelf binary Matryoshka embeddings with stronger backbones. The same code also helps shortlist candidates for reranking and prune low-storage graph indexes such as LEANN.
Read the full abstract

Performance

Mathematical Modeling of a Cognitive Continuum Digital Shadow for Large-Scale, Cross-Facility Workflows

We present the mathematical foundations of a Cognitive Continuum Digital Shadow (CCDS), a decision-support layer between users and the cross-facility infrastructure—instruments, networks, data stores and compute centers—of exascale and post-exascale scientific workflows. The CCDS couples a state-space representation of the continuum with multistage stochastic programming, so that deployment scenarios can be explored and optimized before jobs are launched. This allows operators and users to quantify the cost, makespan and energy trade-offs of a workflow under uncertain resource availability, and hedge their decisions accordingly. We formulate the underlying optimization as a multimode, resource-constrained, stochastic supply-chain network design problem and demonstrate it on a realistic genomics workflow scheduled across heterogeneous HPC and data-center resources. This is the first of three papers; the second treats the underlying software architecture and the third reports large-scale use-cases.
Read the full abstract

Robotics

LightSplat: Real-Time High-Fidelity 3D Gaussian SLAM with Loop Closure

SLAM systems based on 3D Gaussian Splatting (3DGS) have recently demonstrated promising reconstruction accuracy for dense 3D scene representations. However, current 3DGS systems struggle to meet the strict demands of real-world deployments due to severe limitations in operational performance and map adaptability. To this end, we propose LightSplat, a hybrid-representation RGB-D SLAM framework. It synergizes local sparse features for robust and fast tracking with a dual-thread backend that progressively constructs dense Gaussian submaps. Crucially, we enable online loop closure through feature-accelerated 3DGS registration, refining overall map consistency through pose graph optimization. Ultimately, LightSplat achieves the online reconstruction of high-fidelity Gaussian map. Extensive experiments on multiple datasets and real-world robotic platform demonstrate that our method achieves near state-of-the-art reconstruction quality and the capability to accommodate practical camera motions, maintaining an average framerate of 8 FPS. Overall, LightSplat provides an efficient and robust foundation for deploying high-fidelity 3DGS in real-world environments.
Read the full abstract

All authors Junze Bao, Ye Gao, Yiming Huang, Xiaolong Yu, Chen Dong, Qing Gao, Wei Wang, Jinhu Lü

Information Retrieval

Task-Blind No MORE: Multi-Task Information Flow in Unified Ranking Backbones

Industrial ranking models for recommendation have scaled feature interaction and sequence modeling separately; recent architectures such as HyFormer and MixFormer unify both in a stackable backbone. Real-world recommender systems, however, nearly always require multi-task learning, yet existing unified architectures confine multi-task modeling to shallow post-backbone towers, leaving the backbone without task-aware information flow. We propose MORE (Multi-task cO-evolving Ranking modEl), which embeds multi-task information flow inside the backbone, enabling task-specific signals to co-evolve with sequence and feature representations at every layer rather than in a post-hoc fusion. It introduces Anchor Tokens that persist across backbone layers: Shared Anchors encode cross-task commonalities, while Private Anchors capture task-specific priors. In each block, Anchor Tokens (1) read task-conditioned signals from behavior sequences, (2) mix with non-sequential features under a task-boundary mask, and (3) refine per-task representations through independent branches; as blocks stack, each task obtains a differentiated representation refined through all backbone layers. Experiments on large-scale industrial datasets show that MORE consistently outperforms baselines across all tasks under comparable parameter and FLOPs budgets, and scales well with model size. Online A/B tests on Momo, a leading Chinese social discovery platform with tens of millions of monthly active users, yield 3% improvement in usage duration, 3.6% in interaction rate, and 2% in deep-chat rate. MORE is deployed in production with request-level shared computation reducing scoring latency by about 30%.
Read the full abstract

All authors Yuchen Wang, Feng Niu, Qing Tan, Junting Lu, Baoxin Wu, Jun Gao

Chemical Physics

Mixed quantum-classical evolution in open molecular systems

We consider a mixed quantum-classical formulation for open nonequilibrium molecular systems. We employ the pseudoparticle nonequilibrium Green's function (PP-NEGF) method as a fully quantum description and a starting point for introducing classical nuclear dynamics for an open system. We use this formulation to derive a density-matrix equation of motion describing the quantum-classical evolution. We compare our results with previous formulations and highlight important differences related to the nonequilibrium and open character of the molecular system. We also analyze the approximations needed to reduce full quantum-classical dynamics to the fewest-switches surface hopping (FSSH) method and show that these approximations are unreasonable when adiabatic surfaces come close to each other.
Read the full abstract

Statistics Theory

Quantiles for multivariate distribution: a short survey of population concepts

Quantiles are among the most fundamental concepts in Probability and Statistics, from descriptive to inferential. However, quantile functions are well-defined and well-understood in the context of one-dimensional probability distributions, where their definition as the inverse of distribution functions—a definition which is intimately related to the canonical ordering of the real line. Starting with dimension $d=2$, such a canonical ordering is no longer available in~${\mathbb R}^d$; nor is it available in nonlinear manifolds such as hyperspheres, tori, or polyspheres (vectors of directional variables). This results in the absence of an obvious and widely accepted definition of quantiles. The need to extend the concept of quantile beyond the classical univariate context nevertheless has sparked a large body of literature, giving rise to a variety of more or less satisfactory quantile concepts; the recent years have been particularly active in this respect. This short review is an attempt to complement and update Serfling's 25-year-old survey (Serfling 2008) by summarizing and unifying some of these concepts in light of recent contributions.
Read the full abstract

Atomic Physics

Magnetic characterization of electronic components for portable atomic sensors using a zero-field optically pumped magnetometry platform

We present a zero-field optically pumped magnetometry platform for magnetic characterization of a photodetector (PD) board and a resistance temperature detector (RTD) used in a portable atomic magnetometer. For each component, the static magnetic field along the measurement axis is determined from the shift in the center of the dispersive response, while response distortion caused by off-axis magnetic-field components is assessed from the absorptive admixture. Magnetic-field noise is evaluated based on the quadrature difference in the amplitude spectral density (ASD). The platform achieved a -3 dB bandwidth of 71.5 Hz and a median ASD of 57.6 $\mathrm{fT}/\sqrt{\mathrm{Hz}}$ over 20-70 Hz. For the unpowered PD board, static magnetic fields of approximately 8 nT in magnitude and opposite signs were measured in the front- and back-facing orientations. The back-facing orientation also exhibited a degraded dispersive response consistent with magnetic-field inhomogeneity. The magnetic-noise contributions associated with the PD board were estimated at 36.0 and 63.2 $\mathrm{fT}/\sqrt{\mathrm{Hz}}$ in the front- and back-facing orientations, respectively. By contrast, operation of the RTD readout generated a static magnetic field of -0.6 nT, with no measurable response degradation or additional magnetic-field noise. These results provide guidance for the design and placement of electronic components in portable atomic magnetometers targeting sensitivities below 0.1 $\mathrm{pT}/\sqrt{\mathrm{Hz}}$.
Read the full abstract

All authors Hyeonjae Kim, Sangkyung Lee, Sin Hyuk Yim, Taek Jeong, Younghoon Lim

Robotics

Singularity-Free Guiding Vector Fields on SO(3) with Designer-Specified Progression Behavior

This paper develops a singularity-free guiding vector field (SF-GVF) for path following on the special orthogonal group SO(3). First, we lift the Euclidean SF-GVF construction to SO(3), integrating the augmented-state approach with the intrinsic Lie-group geometry and obtaining a closed-form geometric guidance law whose integral curves converge to a designer-specified attitude path. The field is defined on a dense open subset of SO(3), excluding only the measure-zero antipodal set - a manifestation of the topological obstruction to continuous global stabilization on SO(3). The construction requires no per-step optimization and produces a control input intrinsically in so(3) as body angular rates. Second, we formalize the progression behavior along the path as a designer-supplied function \nu(\xi), promoting the parametric speed from an implicitly resolved degree of freedom to a first-class design specification. In contrast to the Euclidean condition v = 0, which excludes vehicles with minimum-speed constraints, the corresponding condition \omega = 0 on SO(3) is physically admissible for most platforms with active attitude control, making the progression behavior a design freedom structurally available on SO(3) but absent in the Euclidean setting. The framework's structural results are established under a bi-invariant Riemannian metric and hold uniformly across choices of path, progression, and Lyapunov gain. The framework is illustrated in simulation on self-intersecting paths under both constant and point-convergence progression behaviors.
Read the full abstract

Combinatorics

The exponent of harmonic LCM avoidance

Fix $k\ge 3$, and let $f_k(N)$ be the largest harmonic sum of a subset of $[N]$ containing no $k$ distinct integers with a common pairwise least common multiple. We prove that $f_k(N)=(\log N)^{\gamma_k+o(1)}$ for a well-defined exponent $\gamma_k\in(0,1]$. Following the weighted-pressure idea of Chojecki, we give a self-contained proof of variational formulas for $\gamma_k$ in terms of weighted sunflower-free families. We then eliminate the continuous weight: if $M_k(n,r)$ is the largest size of an $r$-uniform $k$-cosunflower-free family on $[n]$, then \[ \gamma_k=\sup_{n\ge1,\ 1\le r\le n}\frac{r}{en}M_k(n,r)^{1/r}. \] This finite-block formula yields a direct transfer principle from uniform set-system constructions, recovers the Tang–Zhang bounds, and gives $0.438899\ldots<\gamma_3\le0.889881\ldots$.
Read the full abstract

Quantum Physics

Spectral Criterion for Disorder-Free Localization of Quantum Walks on Hypercube: QBN Approach

We study disorder-free localization in a spin-coupled quantum walk on the hypercube within the quantum Bernoulli noises framework. We derive a general criterion for disorder-free localization: it occurs if and only if some spectral subspace of a reduced unitary evolution contains a vector whose position marginal probability is non-uniform. For the Grover walk with distance-dependent coupling \(\phi_\sigma=|\sigma|\pi/(n+1)\) in dimension \(d=4\), we explicitly construct a non-uniform fixed point in the all \(+1\) spin configuration. This implies that disorder-free localization occurs in our model.
Read the full abstract

Combinatorics

External Difference Families Arising from Two or Three Cyclotomic Classes

We study external difference families arising from cyclotomic classes in finite fields from the viewpoint of a fixed number of blocks. For families consisting of even-indexed cyclotomic classes, the EDF condition can be expressed in terms of relations among cyclotomic numbers. We first study the two-block case and recover a classical characterization in terms of quadratic forms. Our main result shows that, for a prime $p=12k+1$, the family $\{C_0^6,C_2^6,C_4^6\}$ forms an EDF in $\mathbb{F}_p$ if and only if $k$ is a square. The proofs combine symmetry relations of cyclotomic numbers with their explicit evaluations.
Read the full abstract

Dynamical Systems

On the computation of fold and cusp bifurcations in mass action networks via dual equations

This paper presents some results useful in identifying or ruling out fold and cusp bifurcations in mass action networks, and hence in studying multistationarity in these networks. The main result is that the occurrence of fold and cusp bifurcations in mass action networks can be confirmed or ruled out using alternative systems of equations, obtained using principles of Gale duality, rather than the original mass action equations. This can, for certain classes of networks, greatly simplify the calculations involved, avoiding the need for explicit centre manifold reduction or Lyapunov-Schmidt reduction when checking bifurcation, nondegeneracy and transversality conditions. Moreover, the approach leads to immediate corollaries on the non-occurrence of bifurcations in classes of networks with certain combinatorial properties. We present a number of examples illustrating the results and corollaries.
Read the full abstract

Machine Learning

Dense Structural Compression of Transformers via Gauge-Correct Channel Removal

Inference energy per token drives the cost and carbon footprint of deployed transformers. It is dominated by dense matrix products that incur fused multiply-accumulate (FMA) operations and memory traffic. To reduce these computations while retaining dense tensors for high GPU throughput, we develop a methodology from first principles to adapt structural complexity during training to maximize inference utility per unit compute. Channel penalties drive entire tensor slices to zero to enable physical removal while preserving density and the network function. The natural approach, penalizing the norm of operator components acting through each channel, is provably destabilized by gauge freedom. We resolve this pathology with GaugeLasso: additive symmetric group-lasso penalties that recover a monotone function of product-norms when the network converges to gauge balance. Our equilibrium analysis enables per-channel calibration to correctly suppress slices that under-perform in inference utility per unit compute. Under adaptive pressure, the network reorganizes into depth-dependent structural profiles that can be far smaller than the architecture required to learn the task. On polynomial long division over $\mathbb{F}_{31}$, compute compresses from 148 to 255 times with perfect accuracy. On character-level language modeling, compressed models outperform the hand-designed baseline at equal FMA. On masked autoencoding, a compression trial exposes which axes were over-provisioned and which saturated, guiding a better second design. Compaction also accelerates training monotonically as the model progresses. Post-hoc pruning with the same utility ranking cannot reach these structures, showing that sustained pressure is central to discovery of efficient models. Retraining a discovered architecture recovers baseline quality on our statistical tasks, but fails on our exact algorithmic task.
Read the full abstract

All authors Jed A. Duersch, Naïm Es-Sebbani, Nathanaël Haas, Zied Bouraoui

Algebraic Topology

Degrees of Maps between Generalized Dold Manifolds

We study the existence of maps of nonzero Brouwer degree between two orientable, equal-dimensional generalized Dold manifolds (GDMs) fibered by complex partial flag manifolds over real projective spaces. The GDMs considered here are obtained as orbit spaces of the diagonal involution on the product of a sphere and a complex partial flag manifold, acting antipodally on the sphere and by complex conjugation on the flag manifold. We show that no map of nonzero degree exists between two distinct GDMs when their base real projective spaces are different, except in one exceptional case. Furthermore, when the base real projective spaces coincide but fibers are distinct, we prove the nonexistence of maps of nonzero degree in most cases, including those in which at least one of the fibers is not a Grassmannian. As an application of our study, we establish cohomological rigidity for the class of products of a sphere with a complex or quaternionic partial flag manifold.
Read the full abstract

Information Retrieval

Query-Aware Token Budgeting for Efficient Late-Interaction Visual Document Retrieval

Late-interaction visual document retrievers preserve fine-grained page evidence by storing many token embeddings per page, but the resulting storage and query-time interaction costs make large-scale deployment expensive. Pooling document tokens before indexing offers a natural remedy, yet static pooling must decide which visual evidence to preserve before the query is known. We study an alternative: a heavily compressed hot-path index generates candidates, after which query-aware token budgeting operates on the original token sets of the shortlisted pages. We formulate this stage-two selection as a budgeted MaxSim coverage problem, show that a clipped version is monotone submodular, and compare coverage-only, cluster-guided, token-wise, and marginal-gain policies. On ten ViDoRe tasks with ColModernVBERT, direct static pooling reduces macro normalized discounted cumulative gain at rank five from 0.6309 without compression to 0.4738 at a thirty-two-fold pool factor. Under the same candidate-generation regime and a pool-factor-eight-equivalent reranking budget, token top-k recovers 93.93 percent of the full-token score, while greedy marginal-gain selection recovers 98.39 percent. Held-out and leave-one-dataset-out evaluations yield positive greedy improvements over token top-k on every dataset. The latency analysis reveals two useful operating points: token top-k for interactive retrieval and the naive greedy implementation as a quality upper envelope. Together, these results show that late-interaction visual retrieval benefits from query-aware allocation rather than query-agnostic pooling alone.
Read the full abstract

Game Theory

Improved Randomized Approximations for Strategic Obnoxious Facility Location

We study randomized strategyproof mechanisms for strategic obnoxious facility location on a line segment, where agents wish the facility to be located as far away from them as possible and their utility is their distance from the facility, under the social utility and minimum utility objectives. For social utility, we propose a novel randomized mechanism that breaks the previously best known \(\frac32\)-approximation of [Cheng, Yu, and Zhang, TCS 2013], achieving an approximation ratio of at most \(1.47359\). We also raise the lower bound on the approximation ratio of randomized strategyproof mechanisms from \(\frac{2}{\sqrt{3}}\approx1.15470\) [Feigenbaum et al., JAAMAS 2020] to \(\frac{105}{88}\approx1.19318\). For minimum utility, following the profile-independent approach of [Chan, Lin and Wang, AAMAS 2026], we design a simple randomized mechanism that reduces the approximation guarantee from \(\sqrt{2n}+O(1)\) to \(\sqrt n+O(1)\), where \(n\) is the number of agents. Finally, we prove that no randomized strategyproof mechanism can achieve an asymptotic approximation ratio strictly smaller than \(2\), strengthening the previous asymptotic lower bound of \(\frac32\) [Feigenbaum et al., JAAMAS 2020]. Thus, all four bounds considered in this paper strictly improve upon the corresponding previously known results.
Read the full abstract

Functional Analysis

A Family of Completely PPT Maps

Given $\alpha\in \mathbb R$ and $ B_1,\ldots,B_q\in M_{m,n}$, consider the map \[\Phi^{\pm}_{\alpha,\mathfrak{B}}(X)=\alpha \mbox{tr}(X) I_m\pm \sum_{s=1}^qB_sXB_s^*\ \mbox{for all}\ X\in M_n.\] In this study, necessary and sufficient conditions on $\alpha $ and $B_1,\ldots,B_q$ are given that guarantee $\Phi_{\alpha,\mathfrak{B}}^\pm$ is $k$-positive, $k$-copositive, and $k$-PPT, respectively. As application, known results about the map $\alpha \mbox{tr}(X)I_n\pm X$ are recovered. Furthermore, singular value inequalities associated with these positivity constraints are considered.
Read the full abstract

General Relativity and Quantum Cosmology

Thin accretion-disc transfer functions in the q-metric spacetime: quadrupolar lensing signatures

We study geometrically thin equatorial disc images in the static, axisymmetric q-metric spacetime using backward ray tracing. The Schwarzschild solution is recovered at $q=0$; for the nonzero deformations considered here, $r=2M$ is a curvature singularity rather than an event horizon. For a distant nearly face-on observer, we construct the first three equatorial transfer functions $r_m(b;q)$. In the {Schwarzschild} limit, the resolved third-crossing interval contains the critical impact parameter $b=3\sqrt{3}M$. As $q$ increases from $-0.3$ to $0.3$, the branch center shifts from $b_{3,c}/M\simeq3.716$ to $6.702$. In units of the physical monopole mass $\mathcal M=(1+q)M$, it instead decreases from $b_{3,c}/\mathcal M\simeq5.309$ to $5.156$. Much of the displacement in metric-parameter units therefore reflects the changing monopole scale. At fixed $\mathcal M$, the inclined first-intersection maps show that the contour dimensions increase with $q$, with larger relative changes near the inner disc. A prescribed power-law emissivity and relativistic frequency-shift weighting yield direct images whose integrated flux differs from the Schwarzschild value by at most $2.4\%$. Between the endpoint models, the horizontal median-flux position increases by $8.8\%$ and the interquartile width by $14.5\%$. Within this emission prescription, the residual deformation dependence appears mainly as a displacement and broadening of the brightness distribution. Higher-order radiative contributions are not included.
Read the full abstract

All authors Saken Toktarbay, Manas Khassanov, Aray Muratkhan, Zhantay Muratkhan, Aliya Taukenova, Hernando Quevedo

Computer Vision

MV-STRIDE: Enabling MLLMs to Master Multi-View Spatial Reasoning via Hierarchical Capability Modeling

Despite the rapid progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, robust multi-view spatial reasoning remains a fundamental bottleneck due to the lack of structured 3D cognitive pathways in existing datasets. To address this, we introduce MV-STRIDE, a Multi-View hierarchical SpaTial Reasoning dataset with Interdependent and DEcomposed capabilitiEs. Moving beyond flat data structures, MV-STRIDE explicitly models the dependency relationships between foundational perception, scene understanding, and complex contextual reasoning, providing a coherent learning pathway aligned with human spatial cognition. We develop a systematic QA generation pipeline leveraging diverse 3D scene sources that enforces cross-view dependency constraints to prevent single-view solvability, generating multi-level spatial reasoning tasks supported by cognitively grounded chain-of-thought supervision for complex inference. Extensive evaluations demonstrate that our multi-stage training framework based on our hierarchical dataset achieves state-of-the-art performance across multiple spatial reasoning benchmarks, notably the multi-view oriented MMSI-Bench. Our approach enables MLLMs to maintain robust, 3D-consistent spatial reasoning across diverse viewpoints. The code and dataset are available at https://co1dspring.github.io/MV-STRIDE/.
Read the full abstract

All authors Jin Xu, Xiaojian Huang, Zhuodong Luo, Zhihong Zhang, Xin Liu, Jiansheng Wei, Xinzhi Wang, Jie Zhao, Xuejin Chen

Statistical Methodology

Bayesian Generalized Network Autoregressive Model with Structured Shrinkage and Persistence Priors

We propose a Bayesian generalized network autoregressive (BGNAR) model for multivariate time series whose component series are associated with the nodes of a known network. The proposed framework combines the parsimonious network structure of the generalized network autoregressive (GNAR) model with structured shrinkage and persistence priors adapted from Bayesian vector autoregressive (BVAR) modeling. We adapt Minnesota-type shrinkage to both own-lag and network-lag coefficients, with prior variances decreasing over temporal lags and, for network effects, neighborhood orders. A hierarchical prior on the own-lag coefficients allows information to be shared across nodes while retaining node-specific heterogeneity. We further adapt the sum-of-coefficients and dummy-initial-observation priors to the GNAR parameterization. Posterior inference is performed using a Gibbs sampler. Simulation studies show that BGNAR can use the same deliberately over-specified temporal and neighborhood structure across datasets without dataset-specific BIC order selection, while maintaining forecasting accuracy comparable to BIC-selected GNAR and outperforming the unrestricted BVAR benchmark in the settings considered. The structured prior regularizes weakly supported coefficients toward zero within this fixed model. Posterior distributions for the dynamic coefficients and posterior predictive distributions for future observations provide direct quantification of parameter and predictive uncertainty. An application to a wind-speed network demonstrates that BGNAR achieves point-forecast performance comparable to GNAR while additionally providing posterior inference on own-lag and network-lag effects and posterior predictive uncertainty.
Read the full abstract

Analysis of PDEs

Liouville Theorem for the Constant $Q/R$-Curvature Equation

We study positive entire solutions of the flat constant $Q/R$-curvature equation in dimension $n\geq5$. We prove that the weak superharmonicity condition $-\Delta u\geq0$ already forces every nonconstant solution to be admissible. With this automatic admissibility, we classify all such solutions as positive constants or standard bubbles. We also show that the weak superharmonicity condition is structurally necessary. Once it is removed, there is a one-parameter family of positive radial entire solutions near every positive constant. These solutions converge to a positive constant at infinity, while both $-\Delta u$ and $-\Delta(u^{(n-2)/(n-4)})$ change sign infinitely many times.
Read the full abstract

Artificial Intelligence

SkillAlign: Aligning Skill Interfaces for LLM-based Agents

Language-model agents increasingly rely on skills: reusable procedural knowledge for reasoning, tool use, and interaction. Existing work studies how skills are acquired, retrieved, compressed, or composed, but often assumes that once a skill is selected, its interface to the agent is fixed. We argue that this overlooks a key source of skill utility: the same skill can help, distract, or mislead depending on how it is exposed. We propose SkillAlign, a provider-agnostic framework that represents candidate skills as multi-view procedural cards and renders them through alternative exposure interfaces, including full instructions, hints, compressed summaries, workflows, or no exposure. This enables counterfactual evaluation where the task, agent, and candidate skills are fixed while only the exposure interface varies. Across ALFWorld and SkillsBench, we show that exposure form substantially affects task success and rendered context cost, and that compact top-k exposure can outperform full-library injection. We further conduct a replay-based policy-learning analysis on ALFWorld, showing that adaptive exposure contains learnable signal but remains far from oracle selection. Our results suggest that skill-augmented agents should optimize not only which skills to use, but also how those skills are presented.
Read the full abstract

Optimization and Control

Mathematical Programming in Machine Learning and Artificial Intelligence: A Unified Taxonomy of Models and Applications

Mathematical programming provides a common language for many decisions embedded in modern machine-learning (ML) and artificial-intelligence (AI) systems: selecting retrieval context, routing tokens, allocating inference compute, fitting structured predictors, protecting against distribution shift, and balancing competing objectives. However, the relevant literature is fragmented across optimization, information retrieval, recommendation, natural-language processing, computer vision, and learning theory. This paper organizes various applications under common mathematical programming paradigms: linear, quadratic, binary and mixed-integer, conic, bilevel, multi-objective, inverse, distributionally robust, submodular, and min–max optimization. We normalize the models with a mostly unified notation and, for every application, identify inputs, decision variables, a principal formulation, structural properties, solution strategies, and limitations. Across paradigms, we compare tractability, relaxation quality, decomposition, approximation guarantees, and scalability bottlenecks. The paper shows that mathematical programming is most useful not as a claim that all learning is LP or MIP, but as a disciplined interface between predictions and constrained decisions.
Read the full abstract

Distributed Computing

The Maximum Mutual Visibility Set on a Cactus Graph and the Self-stabilizing Constructions

Given a graph $G=(V,E)$, let $S$ ($\subseteq V$) be a set of vertices. Two vertices are mutually visible if there exists a shortest path in $G$ between them that does not contain any other vertex of $S$. A set $S$ is a Mutual Visibility Set (\MVS) if every pair of vertices in $S$ is mutually visible. The concept of \MVS s in graphs has attracted significant attention since its introduction, as it provides an important structural property of graphs. However, determining a maximum \MVS\ in general graphs is computationally intractable; the decision problem of whether a graph admits an \MVS\ of size at least $k$ has been shown to be NP-complete. Thus, prior work has focused on finding maximal \MVS s or restricting attention to specific graph classes. Cactus graphs form a fundamental low-treewidth class, yet the maximum \MVS\ problem for this class remains open. In this paper, we first determine the size of maximum \MVS~in cactus graphs, and introduce two self-stabilizing algorithms that construct such sets. The first algorithm uses a single BFS tree and stabilizes in $O(D)$ rounds with $O(\log n)$ bits per process on average; the second one uses parallel BFS trees and stabilizes in $O(|C_{\max}|+|T_{\max}|)$ rounds, which we show to be asymptotically tight as a function of these two parameters, even on graphs where $|C_{\max}|+|T_{\max}| = o(D)$.
Read the full abstract

Differential Geometry

Curvature Sign Rigidity and Sharp Pointwise Pinching Thresholds

We study connected Riemannian manifolds on which either the sectional curvature or the Ricci tensor is, at each point, strictly positive, strictly negative, or zero, and ask whether the two signs can coexist under pointwise pinching. A general support-rigidity theorem for positive semidefinite divergence-free symmetric tensors is the common analytic mechanism. For sectional curvature in dimension $n\ge 3$, any locally uniform positive lower bound for the absolute pointwise pinching ratio rules out a change of sign. With a fixed pinching constant $\delta_0$, the flat set has no $C^1$ hypersurface piece; it is empty when $\delta_0>1/2$, and is locally porous when $\delta_0=1/2$. These conclusions are sharp in several senses: there are smooth conformally flat local metrics whose sectional curvature changes sign across a flat hypersurface when the pinching degenerates, and there are closed exactly $1/q$-pinched metrics on spheres with isolated flat points. For Ricci curvature, the sharp threshold is \[ \delta_c=\frac1{n-1}. \] A locally uniform gap $\delta_{\mathrm{Ric}}>\delta_c$ forces one Ricci sign globally. Conversely, for every $0<\delta<\delta_c$ there are local sign-changing metrics with exact pointwise pinching $\delta_{\mathrm{Ric}}\equiv\delta$, and exact local examples also exist at $\delta=\delta_c$. Pointwise strictness $\delta_{\mathrm{Ric}}>\delta_c$ is insufficient if the gap collapses at a Ricci-flat interface. For every subcritical $\delta$ we give explicit closed metrics on $\mathbb{S}^1\times\mathbb{S}^{n-1}$, and complete periodic lifts to $\mathbb{R}\times\mathbb{S}^{n-1}$, whose optimal global lower pinching constant is exactly $\delta$. Finally, we prove two ansatz-specific obstructions to compactifying the exact local constructions. The main content of the proof is generated by ChatGPT 5.6 sol and verified by the authors.
Read the full abstract