Search arXiv⌕ Search

arXiv subjects

Search papers

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

At least 1,261 records · Page 70Linked to original sources
arXiv2026

BelT Block Cipher

BelT is a block cipher with a 128-bit block size and a 256-bit key, developed in Belarus and adopted as part of the Belarusian national standard STB 34.101.31 for encryption and integrity control. Its specification was issued as a preliminary standard in 2007, becoming a final standard in 2011. BelT employs an eight-round structure combining elements of a Feistel network and the Lai-Massey scheme. This paper presents a complete description of the cipher and explains the principles underlying its design. Software performance and possible optimizations of BelT are discussed in comparison with those of candidates in the Advanced Encryption Standard (AES) competition.

cs.CR↗
arXiv2026

Degenerations of Jordan algebras

We describe all degenerations of complex $4$-dimensional Jordan algebras and of complex $5$-dimensional nilpotent Jordan algebras. The proofs do not use other papers on degenerations, and all degenerations and non-degenerations were formally verified in the proof assistant Lean~4. The graph of primary degenerations of $4$-dimensional Jordan algebras has $73$ vertices and $160$ edges. The variety of $5$-dimensional nilpotent Jordan algebras is stratified into $63$ normal forms, five of which are one- or two-parameter families; together with two special members of a family this gives $65$ objects, and the graph of primary degenerations has $154$ edges. In particular, we describe all irreducible components of the corresponding varieties: the variety of $4$-dimensional Jordan algebras has $10$ irreducible components, and the variety of $5$-dimensional nilpotent Jordan algebras has dimension $22$ and $5$ irreducible components.

math.RA↗
arXiv2026

Track finding and particle identification performance with particle physics foundation models

Modern high energy nuclear and particle physics experiments produce large amounts of sparse data that must be reduced significantly in size before performing physics analysis. Recently, there has been an effort to evaluate Foundation Models as a potential tool for performing data analysis. These models have been shown to learn features in high dimensional datasets based on the multi-modal data they are trained on. To show their effectiveness at performing complex pattern recognition tasks, such as reconstructing charged particles in a magnetic field, we evaluated the performance of a Foundation Model based reconstruction on several downstream tasks with a dataset composed of spacepoints in a Time Projection Chamber. We show that the performance of the Foundation Model based data analysis significantly improves upon that of traditional methods, and furthermore can be adapted to learn additional tasks that are not necessarily related to the initial training objective.

physics.data-an↗
arXiv2026

A cosmological intercept tension in local distance ladders

The Hubble tension motivates direct tests of the internal consistency of the local distance ladder. We perform forward and backward consistency checks between Type Ia supernovae in Cepheid host galaxies and the Hubble-flow supernova sample using Pantheon+ data. Within flat $Λ$CDM, we impose Planck-CMB priors on one sample, infer the supernova absolute-magnitude calibration, and propagate it to the other sample to recover the Hubble constant, which should match the input value if the second and third rungs of the distance ladder align perfectly. We examine the impact of host-galaxy peculiar velocities using redshift corrections based on the Manticore-Local reconstruction. These corrections shift the recovered $H_0$ from $74.30^{+1.18}_{-1.20}$ to $71.37^{+1.23}_{-1.18}$ $\mathrm{km\,s^{-1}Mpc^{-1}}$ in the forward analysis and from $60.83^{+0.98}_{-0.97}$ to $63.25^{+0.95}_{-0.93}$ $\mathrm{km\,s^{-1}Mpc^{-1}}$ in the backward analysis, bringing both results closer to the input CMB constraint. Comparing the cosmological intercepts of the magnitude-redshift relation inferred from two rungs of the distance ladder before and after peculiar-velocity corrections, we find that the significance of the discrepancy decreases from $6.8σ$ to $3.9σ$ in the forward check and from $7.3σ$ to $4.3σ$ in the backward check. Peculiar-velocity corrections therefore partially alleviate the intercept tension, while a residual discrepancy persists within the adopted framework. Our results highlight the importance of local velocity modeling and motivate further tests of sample selection and calibration systematics in assessing distance-ladder consistency.

astro-ph.CO↗
arXiv2026

The XMM-Newton view of the DEM L36 superbubble in the Large Magellanic Cloud

We present the first multi-wavelength analysis of the DEM L36 superbubble in the Large Magellanic Cloud. We compare the DEM L36 superbubble with other Galactic and Magellanic Cloud superbubbles. We use data from XMM-Newton, supplemented by data from the Magellanic Cloud Emission Line Survey, the Dark Energy Camera Magellanic Cloud Emission Line Survey, the Spitzer Space Telescope, and the Australia Telescope Compact Array. We perform a multi-wavelength morphological analysis to understand the shape and structure of the superbubble and X-ray spectral analysis to derive physical parameters of the hot gas. We find significant X-ray emission in DEM L36, dominated by a central bubble region mostly confined by a thin H$α$ filamentary shell. This has two distinct blowout regions to the east and west, with a possible smaller blowout in the south. Spectral analysis shows the X-ray emission is well fit with a thermal plasma model in collisional ionisation equilibrium with a temperature of $kT=0.17~(0.16-0.18)$ keV and unabsorbed luminosity $L_{X~(0.5-7.0~\mathrm{keV})}=3.62\times 10^{35} $erg s$^{-1}$. Unlike other LMC superbubbles, we find no clear powering cluster at the centre, though relatively small clusters near the edge are likely supplying the mechanical input at the present epoch. We use photometric data of these clusters to estimate cluster masses, ages, and mechanical inputs to compare with the energy stored in the superbubble. We find that these clusters have supplied more energy over their lifetime than is currently observed in the superbubble, indicative of the so-called growth rate discrepancy problem. It is likely the DEM L36 superbubble was initially shaped by a central massive stellar population that has since dispersed and largely lost to supernovae, with smaller clusters at the edges of the swept-up shell being currently supplying the energy input.

astro-ph.HE↗
arXiv2026

Unconstrained learning of equivariant atomistic properties

Machine-learned interatomic potentials based on graph neural networks (GNNs) have made first-principles accuracy routine for the potential energy surface (PES). Their backbones, often O(3)-equivariant by design, are increasingly used to learn other atomistic properties. Many of these are of high angular order or pseudotensorial character, defined per atom, or expanded on large atom-centered basis sets. Such targets strain equivariant GNNs, whose tensor products grow steeply in cost with angular order and can fail to represent the target if truncated, forcing difficult trade-offs between hyperparameters. Unconstrained models like the Point-Edge Transformer (PET) instead replace explicit O(3) algebra with edge-to-edge attention and learn equivariance from data, so that the same backbone can in principle learn any equivariant property. We demonstrate how to use PET to learn molecular dipoles and polarizabilities, NMR chemical shielding tensors, and auxiliary-basis expansions of the electron density. Increasing the backbone size systematically improves both accuracy and equivariance error, and targeted architectural and training choices reduce the symmetry breaking further. PET matches or improves on the released reference models at lower inference cost and, compared with equivariant models, a smaller memory footprint. The predictions give accurate IR and Raman spectra and isotropic shieldings; the predicted densities accelerate self-consistent field calculations and give one-shot ground-state properties, and pretrained density models transfer to higher levels of theory by fine-tuning on little data. The impact of approximate equivariance is small for most observables, further validating this approach to train fast, accurate, and useful surrogate models for properties beyond the PES.

physics.chem-ph↗
arXiv2026

Anomalous polar chains in halide perovskites

Local polar distortions, their collective ordering, and their field-driven reversal have been studied extensively in ferroelectric perovskites for decades. Here we identify an unusual behavior in the ferroelectric halide perovskite CsGeBr3 using large-scale molecular dynamics with first-principles-trained interatomic potentials. In its nominally cubic phase the Ge off-centering forms one-dimensional polar chains that extend over at least tens of nanometers and relax on nanosecond timescales, four orders of magnitude slower than in BaTiO3 and PbTiO3, with CsSnBr3 in between. The cubic CsGeBr3 phase is a dynamically disordered state in which oppositely polarized chains flip throughout the crystal, and a long-lived one-dimensional local polarization coexists with zero net polarization. A simple three-dimensional perovskite thus hosts long-ranged one-dimensional polar correlations with remarkably slow dynamics. In the ferroelectric phase the same chains set a one-dimensional switching mechanism, in which the polarization reverses chain by chain.

cond-mat.mtrl-sci↗
arXiv2026

Realization of a Superleak in a Quantum Gas

Ultracold quantum gases offer a controllable and microscopically accessible platform for understanding and engineering superfluidity. Yet, realizing a superleak that selectively transports the superfluid while suppressing normal flow has remained elusive. Here, we harness an optical superlattice to realize such a superleak through dynamical selectivity. The superfluid undergoes coherent Josephson transport, whereas the normal component exhibits slow, thermodynamically limited response, resulting in a flow-rate contrast of 500(49). Exploiting this contrast, we demonstrate repeated filtering of the normal component and dynamical purification of the superfluid, while sustaining coherent driven oscillations for up to 100 cycles. Our results establish a controllable atomic superleak and open possibilities for manipulating transport and thermodynamics in quantum fluids.

cond-mat.quant-gas↗
arXiv2026

RaSFNet: Ray-aligned and Scan-aware Fusion Network for LiDAR Semantic Segmentation

LiDAR point cloud semantic segmentation is a fundamental component of 3D perception for autonomous driving and robotics. Projection-based methods map 3D point clouds into 2D representations for feature extraction using 2D networks. However, aggregating multiple points during projection can obscure geometric differences within each frustum, while local feature extraction limits the modeling of spatial relationships across scanning regions. Moreover, proximity in the projected representation does not necessarily imply proximity in 3D space, and interactions between point and projected features may mix features from different ranges, reducing semantic discriminability. To address these challenges, we propose the Ray-aligned and Scan-aware Fusion Network (RaSFNet), which incorporates point cloud geometry and LiDAR scanning characteristics into projection-based semantic segmentation. Specifically, the Ray-Aligned Geometric Distribution Encoding (RAGDE) module characterizes point distributions within each frustum using a ray-aligned coordinate system. To model spatial context, a Scan-Topology-Aware Mamba (STAM) block organizes frustum features according to the LiDAR azimuth sampling pattern and captures long-range dependencies across scanning regions. In addition, a Range-Constrained Point-Frustum Fusion (RCPFF) mechanism uses shared range information to guide feature exchange between point and frustum representations while maintaining spatial correspondence. RaSFNet achieves mIoU scores of 70.2% and 77.7% on the SemanticKITTI and nuScenes validation sets, respectively, indicating competitive segmentation performance compared with the evaluated methods.

cs.CV↗
arXiv2026

Entanglement certification of attosecond light fields with photoelectron interferometry

The emergence of entangled quantum optical states in high-harmonic generation raises a fundamental problem about its characterization, namely, \textit{how can entanglement of extreme-ultraviolet radiation be certified for attosecond light fields?} Here, we provide experimentally accessible separability tests, particularly showing that attosecond photoelectron interferometry provides a direct solution to this problem. This is achieved by a quantum optical treatment of mapping photoelectron interferometry to the measurement of optical correlation functions. This yields a negative-partial-transpose witness for infrared-XUV entanglement and a sufficient witness of genuine multipartite entanglement without auxiliary optical measurements. We evaluate both witnesses on explicit optical states and identify how the protocol can be extended to XUV-XUV entanglement certification. Hence, we establish attosecond photoelectron interferometry as a general framework for entanglement certification across a wide range of optical frequencies.

quant-ph↗
arXiv2026

ChainSEED: A Scalable, High-Fidelity Blockchain Emulator

Researchers need blockchain experimental environments with configurable workloads, controllable network conditions, and observable outcomes. Public blockchains and testnets limit experimental control, model-based simulators omit real client implementations, and real-client environments face challenges in combining network control with resource-efficient deployment. We present ChainSEED, a scalable, high-fidelity blockchain emulator built on the SEED Internet Emulator. It separates blockchain configuration from client placement on a configurable routed network. For Ethereum PoS, it organizes separately configured execution, consensus, and validator clients and provides APIs for customizing genesis state, peer connections, and client software. A transaction generator informed by Ethereum mainnet activity, application interfaces, and chain- and node-level monitoring support experiments. ChainSEED also supports Monero and Solana. We examine beacon-node memory reduction, execution-client memory allocation, and validator-client multiplexing to support larger experiments within limited hardware budgets. On a single server with approximately 1 TiB of memory and 384 logical CPU cores, ChainSEED deploys 400 beacon nodes, 400 execution nodes, and 60,000 validator identities while sustaining approximately 20 TPS. It also emulates a 1,200-node Monero network with up to 200 active single-threaded PoW miners and a 200-node Solana network with sustained finalization. Case studies demonstrate routing attacks, blockchain amplification attacks, Monero eclipse attacks, topology-discovery benchmarking against known peer graphs, and network-configuration experiments.

cs.NI↗
arXiv2026

Holographic surface defect RG flows without supersymmetry

We consider $\mathcal{N}=4$ supersymmetric Yang--Mills theory with gauge group SU$(N)$, which admits families of conformal surface defects defined by singular boundary conditions for scalar fields that completely break supersymmetry. At large $N$ and strong coupling, these defects are holographically dual either to a probe D5-brane or to one of two new D7-brane embeddings that we introduce in this work. We also construct embeddings of both D5- and D7-branes describing defect renormalisation group (RG) flows, for which the ultraviolet and infrared fixed points are both dual to the same D$p$-brane, with different values of a parameter. Our embeddings dual to RG flows are mostly constructed numerically, but we find a regime in which we can derive approximate analytic embeddings using a controlled perturbative expansion. We demonstrate that the $b$-theorem is satisfied by all of the RG flows that we study.

hep-th↗
arXiv2026

Site and bond percolation thresholds of all $k$-uniform tilings with $k\le 10$

We determine the site and bond percolation thresholds of all 23,396 $k$-uniform tilings of the plane by regular polygons with $k\le 10$, the complete set according to the known enumeration. We use Monte Carlo simulation on tori and the condition that no wrapping and a cross-wrapping cluster are equally probable, the Monte Carlo form of the critical-polynomial criterion, which needs no knowledge of the large-scale anisotropy that most of these lattices have. The 486 thresholds with $k\le4$ have uncertainties of a few units in the sixth decimal and reproduce the 22 Archimedean reference values (rms deviation 0.71 standard errors) and unpublished bond thresholds of the 20 two-uniform lattices ($χ^2=19.5$ for 20 values); the others have median statistical uncertainties of $2\times10^{-5}$ ($k\le7$) to $7\times10^{-5}$ ($k\ge8$) and no detectable finite-size bias. The vertex-type fractions alone fix the thresholds for $k\ge5$ to about $2\times10^{-3}$ (linear rule fitted on $k\le4$), while tilings with equal composition differ by up to $6\times10^{-3}$ (site) and $2\times10^{-2}$ (bond). In the range of domain sizes covered, descriptors of radius one account for more than half of this difference, and stackings of identical strips have thresholds linear in the fraction of each boundary type. For tilings that differ only by the registry of strips separated by $h$ rows of squares, the bond threshold difference decreases by a factor of 6-8 per row, comparable to the leading momentum-$π$ eigenvalue of the square-band transfer matrix.

cond-mat.stat-mech↗
arXiv2026

Timelike Supersymmetric Geometries in Seven-Dimensional Supergravity

Seven-dimensional supergravity poses a uniquely intricate supersymmetry problem, combining topologically massive higher-form gauge dynamics, distinct Clifford algebra properties, and rich geometric structures. We first establish the complete formulation of the maximal Cartan-gauged theory in terms of the gauge-invariant higher-form flux, including both the bosonic action and supersymmetry variations. This formulation, which also makes the smooth ungauged limit manifest, is achieved by requiring a consistent truncation to the minimal Cartan-gauged theory in the equal-charge limit. We construct the complete bilinear system for a timelike Killing vector, where additional internal orientation of the symplectic-Majorana spinor arises already in the scalar sector, uniquely among pure minimal ungauged theories. In the equal-charge sector, we decouple the algebraic constraints into SU$(3)$-isotypic components, and analyse the differential bilinear relations, determining the permitted dilaton gradient, fluxes, fibration twist and intrinsic torsion of the base. We prove that with gauging and active Cartan flux, the complex scalar bilinear vanishes identically. We also construct a Killing gauge for the higher-form potential compatible with the self-duality condition. We show that with gauging but no active Cartan flux, the Killing vector also acts as a generator of the gauged R-symmetry, and the base structure is enlarged from SU$(3)$ to U$(3)$. We find that BPS-flow solutions realize this setting. With supersymmetry, we isolate a single independent component of the 3-form equation across an exhaustive four-class partition for both theories within this setup. Applied to known and proposed solutions in both theories, this classification exposes unexplored sectors and provides an accessible toolbox for constructing and testing solutions and developing new geometric and holographic applications.

hep-th↗
arXiv2026

Large Language Models Meet Bayesian Optimization: Towards Self-evolving Wireless Digital Twins

Wireless digital twins (DTs) are increasingly integrated with artificial intelligence to support online modeling, prediction, and network optimization. However, most existing approaches perform adaptation within predefined learning or optimization structures, which may become unsuitable when wireless tasks and operating conditions change. In this article, we investigate the integration of large language models (LLMs) and Bayesian optimization (BO) to extend wireless DTs from online adaptation towards algorithm-level self-evolution. We introduce an LLM-assisted BO framework in which BO provides a stable and sample-efficient optimization backbone, while the LLM determines whether selected BO components should be retained or adapted based on the current task and accumulated optimization experience. In simulations, we apply the framework to Gaussian process (GP) kernel discovery for a multi-user wireless beam-selection problem. Simulation results show that the LLM-discovered kernel improves the final BO performance by 19.7% on the training tasks compared with the strongest standard kernel baseline. The results demonstrate the potential of combining LLM flexibility with structured BO for self-evolving wireless DTs.

eess.SY↗
arXiv2026

When More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions

Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields should improve grounding. Museum description generation challenges this assumption because catalogue records are heterogeneous, partially visual, and unevenly documented across geographic regions. We study how metadata composition shapes the generation of artefact descriptions through a two-phase framework. Phase 1 analyses natural metadata variation in a large museum collection, where field count, field type, image availability, and Place co-vary. Phase 2 introduces controlled field probes that hold object identity fixed while varying metadata quantity, field type, and image access. Our results show that metadata quantity alone does not explain generation behaviour. Field type determines whether models ground, omit, or infer information, while images can both support visual grounding and increase unsupported interpretation. More broadly, our findings show that metadata-conditioned data-to-text evaluation should assess field- and claim-level support, rather than relying only on aggregate input coverage or reference similarity.

cs.CL↗
arXiv2026

Inside-JEPA: From Prediction to Relational Grounding in Latent World Models

Joint-embedding predictive architectures (JEPAs) learn world models in latent space, enabling efficient planning but leaving their states opaque. We introduce Inside-JEPA, which places relational channels for ordered entity pairs inside the JEPA latent state and shows that an inverse-dynamics loss can ground these channels without relation labels when its head reads them, either alone or as part of the full state. On Something-Something V2, latent prediction alone performs no better than a random relation head, while inverse-dynamics supervision that recovers actions from channel changes improves relational AUROC by .065-.103 across ten settings with annotated boxes, by .081 on RH20T robot episodes with held-out tasks, and by .050-.061 across five relations using off-the-shelf detection and tracking. An inverse head on entity latents decodes actions more accurately but barely improves relational grounding; shuffled labels, two label-free objectives, and action labels on only 5% of the pairs also fail to ground the channels. Given an action, predicted relational changes approach the quality of observed future relations (AUROC .599 vs .621). For one-step relational goals, a supervised readout of the 16 channels ranks the correct action among the top five of 65 in 43% of cases, compared with 24% for prediction-only latents and 17% for a frequency prior.

cs.LG↗
arXiv2026

Constrained Sparsest Input Selection for Structural Controllability is NP-Hard

We study the problem of selecting the fewest admissible input--state links that make a structured linear system structurally controllable. This objective counts nonzero entries from prescribed positions of the input matrix and is different from minimizing its number of nonzero columns. Prior work gives polynomial-time algorithms for several restricted cases, including single or dedicated inputs and prescribed input patterns satisfying additional graph-theoretic conditions. We give a polynomial-time reduction from perfect three-dimensional matching, establishing for the first time that the decision problem is NP-complete and the optimization problem is NP-hard. A refinement of the reduction yields an exact optimum relation and shows that the problem is APX-hard. An example also exhibits a strict integrality gap for the natural linear programming relaxation.

eess.SY↗