Peer Revenue Past citations are no guarantee of future returns

Papers

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

All of arXiv

Friday 11 September

Optimization and Control

Controlling the energy jump of multistable structures using shape optimization

Multistable systems admit distinct stable equilibria for the same loading and boundary conditions. Shape optimization of multistable structures offers a promising route to engineer the mechanical response of advanced materials and metamaterials. In this work, we develop a formulation and algorithm for controlling the energy jump associated with snap-through behaviour in hyperelastic metamaterials by optimizing the shape of the domain. The key challenge is that classical shape optimization theory assumes a single-valued domain-to-solution map, which does not hold in the multistable setting. We address this by extending the theory to settings with non-unique solutions, treating the scaling of the domain deformation as a continuation parameter and applying the implicit function theorem to give sufficient conditions for the continued existence of multiple solution branches as the shape varies. This theoretical foundation underpins a practical algorithm that targets a prescribed energy jump between stable states, enabling the systematic design of structures whose snap-through response can be tuned on demand.
Read the full abstract

All authors Arselane Hadj Slimane, Patrick E. Farrell, Alberto Paganini, Àlex Ferrer

Atmospheric and Oceanic Physics

4D Parallelism Unlocks Exascale Bayesian Neural Networks for High-Fidelity Atmospheric Modeling

We present BEAST, the first-ever Bayesian Swin Transformer for atmospheric forecasting on 0.25$^\circ$ global resolution able to accurately quantify both aleatoric and epistemic uncertainty. To overcome the associated computational bottlenecks, we devise an orthogonal 4D-parallelization scheme that introduces a unique domain-tensor-parallelism strategy and a novel uncertainty parallel method, enabling us to fully leverage GPU capacity and efficiently scale model training. For a 2.4-billion-parameter model, we achieve a peak performance of 3.96 EFLOP/s on 20,480 NVIDIA GH200 GPUs on the JUPITER supercomputer. We train BEAST as a 700-million-parameter model with 96 random weight samples on 384 nodes on 40 years of data for nearly one million gradient updates. This model achieves predictive skill scores competitive with state-of-the-art probabilistic atmospheric AI models and numerical models, and can predict extreme events with exceptional skill, while generating large ensembles 3 to 4 times faster than the current-best AI model. Our contribution unlocks the potential of high-fidelity uncertainty quantification in atmospheric AI models, heralding a new era for AI-based models in climate and Earth system sciences.
Read the full abstract

All authors Deifilia Kieckhefen, Juan Pedro Gutiérrez Hermosillo Muriedas, Lars Helge Heyen, Mathis Bode, Iida Hakulinen, Andreas Herten, Chelsea Maria John, Thorsten Kurth, Anni Moisala, Asena Karolin Özdemir, Kaleb Phipps, Oskar Taubert, Arvid Weyrauch, Markus Götz, Charlotte Debus

Machine Learning

RunningTensor: Generalizing Linear Attention to Higher-Order Recurrent States

Linear attention and state-space models provide linear-time sequence modeling, but their recurrent memory remains a second-order tensor (a matrix), limiting the order of interactions that can be represented in the state. We introduce the RunningTensor, which generalizes this memory to an order-$o$ tensor, updated by a rank-1 outer product and read by contracting against $o-1$ vector queries. Order $2$ recovers linear attention; we study order $3$ as a proof of concept, retaining both recurrent and parallel forms while remaining linear in sequence length $T$ and improving working memory capacity from $\mathcal{O}(W^2)$ to $\mathcal{O}(W^o)$. On synthetic multi-query associative recall, RunningTensor outperforms linear-attention and SSM baselines. After pretraining, it also improves performance on language-understanding and non-synthetic retrieval tasks, suggesting that higher-order recurrent state can provide useful additional memory capacity beyond matrix-valued state.
Read the full abstract

High Energy Physics (Theory)

Dimensional reduction of Carroll-Field-Jackiw theory

This article investigates the dimensional reduction of the massive Carroll-Field-Jackiw (CFJ) theory from $(3+1)D$ to $(2+1)D$ dimensions with the objective of constructing and analyzing the resulting effective field theory, termed pseudo-Carroll-Field-Jackiw (PCFJ). The theoretical framework relies on quantum field theory, pseudo-quantum electrodynamics (PQED) and Lorentz-violating models. The methodological procedure projects the three-dimensional dynamics onto the plane by rigorously integrating out the gauge field degrees of freedom along the transverse momentum. Subsequently, the study calculates the static interaction potential, applies both the generalized optical theorem and the Källén-Lehmann spectral representation to test probability conservation. The results reveal that the effective theory exhibits an intricate non-local structure, governed by nested pseudo-differential operators, which generates a static potential that transitions from a Coulomb-like behavior at short distances to a long-range decay modified by a topological mass. The research analytically proves that the integration of the extra-dimensional topology introduces ineliminable branch cuts in momentum space, which preclude the causal definition of asymptotic states and determine the breakdown of unitarity. The work concludes that this quantum instability emerges from the direct interplay between the spatial projection and the pre-existing Lorentz-violating background in the bulk.
Read the full abstract

High Energy Physics (Theory)

Stochastic dynamics from O(2N) fractional Laplacian vector model and O(N) vector model free energy in finite temperature

We explore O(2N) vector model with fractional Laplacian, $\sqrt{-\nabla^2}$ in $d$-dimension and its Hamiltonian dynamics which is described by a Schrodinger type equation. This equation is a kind of current conservation equation, where one can define a current $j(\phi^a)$ of a probability $P(\phi^a)$, where $\phi^a$ is the O(2N) vector field. Naturally, Gibbs entropy $S=-\int [D\phi^a] P(\phi^a)\log P(\phi^a)$ can be considered to explore the system. We realize that this Gibbs entropy of the O(2N) vector model with fractional Laplacian is matched with free energy of O(N) vector model at finite temperature, $1/\beta$ with a deformation, $\mu$ in $d$-dimension. The precise map between the stochastic fictitious time $t$ and the inverse temperature $\beta$ is $\beta=2t$. Therefore, the temperature dependence of the thermal O(N) vector model can be realized as the dynamics of a time dependent solution satisfying Schrodinger type equation. This free energy is obtained by putting O(N) vector model in $S^1\times \mathbb R_d$, where $S^1$ is thermal circle with its periodicity $\beta$. To get $d$-dimensional theory, we sum up all possible frequencies on the circle(so called Matsubara frequency summation) which gives $d$-dimensional thermal partition function. We note that the nontrivial $t$-dependence appears beyond classical limit. To take into account quantum effects, we solve the Hamiltonian dynamics by keeping $\hbar$ corrections. The spectral deformation is mediated by a parameter $l$ such that $\mu=\beta^{-1}\log l$ and so we call this $l$-deformation. This is related to the initial boundary condition of the Schrodinger equation. We also note that the two theories are not equivalent to each other and we just check their correspondence at the level of one-loop determinant, i.e. zero point function in the note.
Read the full abstract

All authors WooCheol Shin, Jun Hyuk Lee, Ji-seong Chae, Jae-Hyuk Oh

Group Theory

Synchronizing groups and weakly perfect graphs: a survey

The concept of synchronization has been transferred from finite automata, via transformation semigroups, to permutation groups: a permutation group $G$ on a finite set $\Omega$ is said to be synchronizing if, for any non-permutation $t$ on $\Omega$, the transformation monoid $\langle G,t\rangle$ contains an element of rank~$1$. Synchronizing permutation groups are primitive and basic, and so are affine, diagonal, or almost simple (in the O'Nan–Scott classification). In an earlier paper, the author, with João Araújo and Ben Steinberg, surveyed what was known at the time about synchronizing permutation groups. Since then, several new classes of groups have been tested for the synchronizing property, using the fact that a group is non-synchronizing if and only if it preserves a nontrivial weakly perfect graph. Cases studied include diagonal groups with more than two socle factors, two natural actions of symmetric groups, and some actions of classical groups; the proofs are scattered in the literature. My primary purpose is to collect these results with their proofs. As a second objective, I will explain the background, and comment on some related topics. (Reading this paper does not require knowledge of the earlier paper.)
Read the full abstract

Cryptography and Security

Batten the Hatches: Cybersecurity with Military Mariners

Cyberwarfare has become a key component of contemporary geopolitical conflict. However, there has been extremely limited systematic investigation into how cybersecurity is handled by military organizations and personnel. The military context is unique compared to other operational ones, with immense resource availability (U.S. military spending approached 1 trillion dollars in 2024), a rigid chain of command, and extraordinary consequences for its actions. Thus, military cybersecurity is a distinct yet understudied topic. In this paper, we take an early step at understanding military cybersecurity by investigating how service members understand, recognize, and respond to cyber risk. We focus on maritime services and carefully consider organizational barriers to design an unclassified study and conduct semi-structured interviews with 20 military mariners from U.S. Navy and Coast Guard vessels. Through our investigation, we identify unique consequences of compromising military systems, including weapon takeover and purposeful geopolitical escalation. We find that cybersecurity is organizationally abstract on ships, so mariners build cyber risk models from informal experience rather than formal instruction. They nonetheless make cybersecurity actionable by recognizing operational impacts and responding with a safety-oriented incident-response model that creates resilience but may delay cyber attribution and containment. These findings inform actionable recommendations to help military operators frame cyber threats, merge longstanding nautical doctrine with modern systems, and apply military insights to the civilian sector, all to secure the broader maritime environment.
Read the full abstract

All authors Ryan Von Brock, Anna Raymaker, Animesh Chhotaray, Frank Li, Saman Zonouz, Raheem Beyah

Geometric Topology

Non-Hitchin Borel Anosov representations from surface groups to $\mathrm{SL}_{3k}\mathbb{R}$

We use the Labourie–Wentworth's formula and the thermodynamic formalism to show that, along the slice constructed by Bronstein–Davalo, the logarithmic top-eigenvalue length spectrum has a uniformly positive second variation near the Barbot representation. As a major application, we show that every closed surface group admits a non-Hitchin Borel Anosov representation into $\mathrm{SL}_{3k}\mathbb{R}$ for every $k\geqslant 1$. In particular, we obtain the first such examples in the even dimensions $6k$. We also study the local behavior of related objects of this slice around the Barbot representation, including the Lyapunov exponent of the flat bundle, the Hausdorff dimension of the limit set, and the Hilbert entropy of the representation.
Read the full abstract

Artificial Intelligence

K-Bench: A Benchmark for LLM Unlearning in Agentic Deployments

Unlearning benchmarks such as TOFU and MUSE certify forgetting by reading the model's final answer, where a model that refuses to answer already counts as having forgotten. We show that this model-level certificate does not transfer once the model is deployed as an agent. We introduce K-Bench, a benchmark that scores LLM unlearning under agentic deployment. K-Bench inspects all six channels a ReAct agent exposes, including its chain-of-thought (CoT), tool calls and tool observations, and elicited summary. A query counts as leaked if the secret appears in any of them. Each experiment places the secret in exactly one of the agent's three sources (the weights, the prompt, or the retrieval store). The K-Score is computed separately for each source and credits forgetting only when the agent remains usable. Clearing the answer channel does not make the secret unrecoverable. On structured retrieval, the secret stays verbatim in the tool-observation channel and the aggregate leak rate is unchanged. When the secret lives in the prompt or the retrieval store, TOFU and MUSE report no leakage, while the deployed agent still leaks it on 22–86% of queries. When the secret is in the weights, none of the twenty evaluated published methods demonstrably removes it, and only an input-corruption intervention reaches selective forgetting under the evaluated observer. The top-ranked method changes across base models. A refusal-tuning method resists the evaluated extraction without verified knowledge removal.
Read the full abstract

All authors Guangsheng Yu, Yanna Jiang, Qin Wang, Baihe Ma, Xu Wang

Probability

Large deviations for the CRT

We establish large deviation principles both for the CRT in the Gromov-Hausdorff topology and for the CRT weighted by its mass measure in the Gromov-Hausdorff-Prohorov topology. In the first case, the rate function of a tree $\mathcal{T}$ is half the square of the length of $\mathcal{T}$. In the second case, the rate function of a pair $(\mathcal{T},\mu)$ is half the integral with respect to the length measure of $\mathcal{T}$ of the inverse of the absolutely continuous part of $\mu$.
Read the full abstract

High Energy Physics (Phenomenology)

Reconstructing the Dark Matter Equation of State with Compact Object Inspirals

We investigate the gravitational wave (GW) signatures of compact binaries embedded in extended dark matter (DM) configurations in hydrostatic equilibrium. If a neutron star is surrounded by a sufficiently massive and extended envelope, an inspiraling companion experiences dynamical friction (DF) in addition to the standard GW energy loss. We show that this environmental effect can be isolated through a GW observable, the $D$-function, which directly characterizes the additional dissipative power induced by the surrounding medium. Since the density of the envelope is directly determined by the DM equation of state together with the stellar boundary conditions, the frequency dependence of the $D$-function provides direct information about the macroscopic properties of the dark sector. We develop a regression-based framework to reconstruct both the density profile and the DM equation of state over the density range probed by observations of the GW inspiral. Using representative DM models as benchmarks, we demonstrate that the method accurately recovers the input equation of state while remaining largely independent of the microscopic realization of DM. Our results show that future GW observations of compact binaries can provide a probe of the equation of state of DM.
Read the full abstract

Algebraic Geometry

Data-Adaptive Grassmann Manifold Representations for Spatial Transcriptomics Alignment

Spatial transcriptomics measures gene expression together with spatial coordinates, but many existing analysis methods represent each spot primarily by a single feature vector. We propose GrassST, a subspace method for spatial transcriptomics analysis and cross-slice alignment. For each spatial spot, GrassST constructs a neighborhood from tissue coordinates, fits a low-dimensional subspace to the embedded expression profiles in that neighborhood, and represents the spot by the resulting subspace. These representations are points on a Grassmann manifold and can be compared using distances between subspaces. GrassST selects the neighborhood size and subspace rank from the spectral energy of the data, allowing these parameters to vary across datasets rather than being fixed globally. Experiments on four spatial transcriptomics datasets show that GrassST achieves competitive clustering and cross-slice integration performance under a unified evaluation pipeline.
Read the full abstract

Mathematical Physics

Random field approximation and local counting statistics for the weakly interacting thermal Bose gas

We study weakly interacting bosons on the three-dimensional torus at temperatures proportional to the critical temperature for Bose–Einstein condensation. We show that, as the expected particle number tends to infinity, the grand canonical Gibbs state is asymptotically described by the coherent state quantization of a random field that converges to a novel Bogoliubov random field. Moreover, the point process associated with the Gibbs state can be approximated by a Cox process. These results are based on and extend the approximation of the Gibbs state recently obtained by the first two authors and Nam. Using these approximations, we compute limiting distributions of the number of particles outside the condensate, the joint occupation statistics of finitely many momentum modes, and particle-number fluctuations in macroscopic subsets of the torus, all governed by the Bogoliubov field. We further prove convergence of the empirical measure of the associated point process to the uniform measure and convergence of the microscopic particle-number statistics to the boson point process. The latter result establishes the universality of particle-number statistics on microscopic length scales.
Read the full abstract

Signal Processing

Multi-Label 12-Lead ECG Classification on the PTB-XL Dataset: A Comparative Evaluation of Deep Learning Architectures and Heterogeneous Ensemble Approaches

This study aimed to compare the performance of different deep learning architectures and heterogeneous ensemble learning approaches for multi-label 12-lead ECG classification on the PTB-XL dataset. Five different models, namely 1D-ResNet18, Bidirectional Mamba, xLSTM, CWT-ViT-KAN, and the pre-trained ECGFounder, were evaluated. Utilizing the recommended split structure of the PTB-XL dataset, folds 1-8 were allocated as the training set, fold 9 as the validation set, and fold 10 as the independent test set. In addition to the individual models, three different ensemble approaches were investigated: Equal-Weight Soft Voting, Validation-Weighted Soft Voting, and stacking. Among the individual models, the highest performance was achieved by ECGFounder, with a Macro AUROC of 0.930 and a Macro AUPRC of 0.823. For the ensemble models, the highest values in the primary macro performance metrics were obtained by the stacking approach, achieving a Macro AUROC of 0.936, a Macro AUPRC of 0.836, and a Macro F1 of 0.763. The highest subset accuracy of 0.630 was achieved using the Validation-Weighted Soft Voting method. The findings indicate that heterogeneous ensemble models, which combine different representation learning approaches, can provide additional performance improvements over individual models in multi-label ECG classification.
Read the full abstract

Quantum Physics

Testing and Diagnosing Parameter Miscalibration in Probabilistic Error Cancellation via Direct Fidelity Estimation

Probabilistic Error Cancellation (PEC) is a standard technique for mitigating noise in quantum computations, but its unbiasedness relies on accurate knowledge of the underlying noise parameters. In practice, these parameters are inferred from experimental data and may be inaccurate. We introduce a protocol that tests PEC parameter consistency on hardware by applying Direct Fidelity Estimation (DFE) to the effective mitigated state. The resulting DFE overlap serves not only as a witness of parameter miscalibration but, under appropriate conditions, also reveals whether the PEC parameters have been overestimated or underestimated. We establish this behavior for arbitrary circuits under global depolarizing noise and for Clifford circuits under general Pauli-stochastic noise. For arbitrary circuits subject to Pauli-stochastic noise, the underestimation result holds generally, while the corresponding overestimation result is established perturbatively for sufficiently small consistent miscalibrations under an additional nondegeneracy condition. We further show that the purity of the effective mitigated operator provides a more general diagnostic, allowing the miscalibration direction to be identified for arbitrary circuits and arbitrary parameter deviations. Finally, we describe a two-copy PEC procedure for measuring this quantity and show that, together with the DFE overlap, it determines the Hilbert-Schmidt distance from the ideal state. The sampling overhead incurred by our protocol is the same as that of PEC, and we discuss strategies to improve it.
Read the full abstract

All authors Francesc Sabater, Tristan Kraft, Carlos A. Riofrío, Barbara Kraus

Artificial Intelligence

Interpreting the predictions of neural network classification based on a Taylor Coefficient Analysis (TCA)

We introduce a rigid and comprehensive taxonomy and paradigm for characterizing the influence of the input feature space $X$ on the predictions $\hat{y}$ of a neural network (NN) used for event classification, based on a Taylor expansion of $\hat{y}$ in $X$. The complete process of introspection we refer to as Taylor Coefficient Analysis (TCA). Based on two simplistic example tasks, which can be easily understood and bencmarked, we illustrate the power of the TCA when it comes to revealing, what properties of $X$ have led to what value of $\hat{y}$, of a given NN model, building up intuition for the method. A more complex application is meant to represent $X$ of a typical classification task at a CERN LHC experiment. Based on this application, we play through the different levels of introspection that the TCA offers and discuss a number of practical aspects for a TCA application typical for the analysis of CERN LHC data. We conclude with a study to support the assumption that those properties of $X$ most relevant for tasks of the complexity typical for a CERN LHC experiment, are usually caught by a TCA up to the second order.
Read the full abstract

All authors Markus Klute, Artur Monsch, Lars Sowa, Roger Wolf

High Energy Physics (Phenomenology)

Associated $W$ + charm production: indications for PDF analysis

The strange quark and antiquark contents of the proton remain weakly constrained compared to the other light quarks, particularly the asymmetry between the two. Production of a $W^\mp$ boson in association with a charmed meson $D^{(*)\pm}$ could provide additional constraints in future fits of parton distribution functions. We calculate this process in general-mass variable-flavor-number scheme at next-to-leading order in perturbative Quantum Chromodynamics. We investigate the production ratio between events with oppositely charged $W$ bosons, in which various theoretical uncertainties largely cancel while sensitivity to PDFs remains. We compare our predictions for this quantity with the recent ATLAS data at $\sqrt{s} = 13$ TeV and find that CT18ANLO, which sets the strangeness asymmetry to zero, yields good agreement with the data while MSHT20NLO and NNPDF4.0NLO, which have positive strangeness asymmetry in the relevant range of momentum fraction $x$, show more tension with the data. An approximate PDF-level analysis of the production ratio indicates that this tension could be attributed to a too large strangeness asymmetry.
Read the full abstract

Instrumentation and Detectors

Surface Commissioning and Performance of the sMDT Muon Chambers for the ATLAS HL-LHC Upgrade

To improve the first-level muon trigger efficiency at the HL-LHC, the Monitored Drift Tube (MDT) chambers in the inner small sectors of the ATLAS Barrel Muon Spectrometer will be replaced with integrated modules of small-Diameter Muon Drift Tube (sMDT) and Resistive Plate Chambers (RPCs). This paper reports on the surface commissioning of 102 new sMDT chambers at CERN in 2025 following the installation of the final front-end electronics, including the new Amplifier-Shaper-Discriminator (ASD), Time-to-Digital Converter (TDC) chips, and Chamber Service Module (CSM) developed for the HL-LHC. The commissioning included measurements of gas leak rates and high-voltage dark currents, tests of the integrated planarity monitoring system, noise characterization, and measurements of detector efficiency and spatial resolution using cosmic rays. Fewer than 0.1% of the 49152 tubes were found to be non-functional due to broken sense wires or gas leaks. The chambers exceed the ATLAS design requirements, with gas leak rates a factor of five below the specified limit of $9.3\times 10^{-3}~\mathrm{mbar\cdot liter/s}$ per chamber, dark currents of only around 0.2~nA per tube, average channel noise hit rates below 30~Hz, and average drift tube efficiency and spatial resolution of 99% and $82~\mu\mathrm{m}$, respectively, at an effective threshold of 15 primary electrons.
Read the full abstract

All authors Jiajin Ge, Elena Voevodina, Hubert Kroha, Tatiana Azaryan, Kwok Ching Cheung, Tiesheng Dai, Edward Diehl, Claudio Ferretti, Yuxiang Guo, Oliver Kortner, Chihao Li, Chun Kit Lo, Nick Meier, Emmett Salzer, Can Suslu, Cecilia Vanesa Imthurn, Samuel Hugo Venetianer, Curtis Weaverdyck, Chuanshun Wei, Bastian Michael Wesely, Ruslan Yakubovych, Bing Zhou, Junjie Zhu, Jörg Zimmermann

Computer Vision

LGFN: Lightweight Gated RGB-Polarization Fusion with Modality-Availability Conditioning for Camouflaged Object Detection

Camouflaged object detection (COD) is an important engineering task in intelligent optical perception, but it remains challenging when targets closely resemble their surroundings. Polarization imaging provides complementary physical cues, whereas existing methods typically assume fixed multimodal input configurations and entangle intra-polarization coordination with interaction between red-green-blue (RGB) and polarization representations. We propose LGFN, a lightweight gated RGB-polarization fusion framework supporting separately optimized RGB-only and polarization-assisted configurations. A deterministic Modality Router selects the appropriate configuration according to polarization availability. In the multimodal configuration, an availability-conditioned Modality Gate calibrates the available polarization branches; the Gated Polarization Hub coordinates learned degree of linear polarization (DoLP) and angle of polarization (AoP) representations with explicit polarization cues; and RGB-Polarization Cross Fusion introduces the coordinated representation into the RGB hierarchy through controlled residual interaction. The multimodal configuration requires neither sample-dependent statistics nor handcrafted quality descriptors during inference. On the complete 230-image PCOD_1200 test set, the RGB-only configuration achieves a mean absolute error of 0.0090, a Dice score of 0.8806, and an intersection over union of 0.8144, obtaining the best results on all six metrics among the evaluated RGB-based methods. Under a common local reevaluation protocol, the multimodal configuration outperforms PolarNet and IPNet on all six metrics. Relative to IPNet, it reduces the parameter count, floating-point operations, and latency by 53.1%, 73.6%, and 63.0%, respectively. The source code will be available at https://github.com/1hzf/LGFN.
Read the full abstract

All authors Zhuangfan Huang, Xiaosong Li, Yang Liu, Tao Ye, Haishu Tan

General Relativity and Quantum Cosmology

The Wavelet Detection Filter: a real time unmodeled pipeline for gravitational wave transients, ranking coincidences with a graph neural network

Accurate waveform models are unavailable for many potential sources of gravitational wave transients, motivating searches that do not assume a predefined template family. The Wavelet Detection Filter (WDF) is an unmodeled pipeline that identifies excess energy in the wavelet coefficients of whitened detector data. Here, we develop WDF into a low latency pipeline for both detection and reconstruction and this paper describes and validates that pipeline. We use a zero phase streaming autoregressive whitening filter that preserves transient morphology with a fixed latency and we use the retained wavelet coefficients to provide both a detection statistic and a sparse representation of the signal. By combining coefficients across consecutive analysis windows, the pipeline reconstructs transients longer than a single window. We represent candidate coincidences between detectors with a trigger graph and rank them with supervised and unsupervised graph neural networks (GNNs). We estimate the significance of the ranked coincidences from time shifted data. We validate the pipeline on Gaussian noise colored according to the Advanced LIGO sensitivity curve and on real O4 strain containing compact binary and core collapse supernova injections. Applied to GW250114, WDF identifies the event as the loudest zero lag candidate in the analyzed segment without using waveform templates. We measure the computational cost of the pipeline and the latency it introduces. Both are compatible with a low latency search,so WDF can run on a stream in real time.
Read the full abstract

Materials Science

Agent-authored deposition recipes for X-ray multilayer mirrors: schema-bound LLM control of a magnetron sputtering system with reflectivity-verified outcomes

Periodic multilayer mirrors for X-rays require a layer period controlled to about 0.1 nm, set by deposition rates held in the laboratory's own notebook. A supervising model session gave a large language model agent a target multilayer specification and calibration constants derived from a laboratory archive. The agent authored recipes in the magnetron sputtering system's native recipe language and submitted them through a Model Context Protocol bridge of 17 tools in three tiers, with enforcement in the control application. The target was a 30-bilayer Ru/C multilayer with a period of 6.85 nm, a pair chosen because each period is fitted as two layers. The agent could read and write only its own working project, so it could alter neither the calibration record nor the measurements that scored it. We scored mirrors by X-ray reflectivity and fixed the 3% acceptance threshold on relative period difference before deposition. Across four scored depositions, the relative period difference fell from 20% on the calibration as given to 2.26% after one measured-feedback iteration and 0.79% after a second. We identified sixteen failure modes of the agent-facing interface. The result is bounded to the period, pair, and chamber observed.
Read the full abstract

All authors Oleksiy V. Penkov, Haoyu Fu, Yueying Zhan, Pengyuan Wu, Jiong Jin

Robotics

High-Fidelity Multi-Body Simulator for Autonomous Racing

We present a custom high-fidelity vehicle dynamics simulation environment for testing and validation of Autonomous Racing software. The digital twin of the autonomous vehicle is developed in Dymola, using racecar dynamics modeling libraries to build a complete multi-body model. A 3D road surface, including elevation profiles and curbs, is implemented using the Curved Regular Grid (CRG) standard. The model is exported from Dymola as a Functional Mock-up Unit (FMU) and integrated into a custom software-in-the-loop simulator, where communication interfaces with the autonomous racing stack were developed in C++. A calibration procedure based on experimental data is also presented, along with a validation study to further support the quality of the proposed framework. The simulator runs in real time on a portable computer and provides reliable ground truth for algorithms validation prior to real-world deployment.
Read the full abstract

All authors Nicola Musiu, Francesco Iacovacci, Fausto Lupo, Matteo Pini, Giovanni Scapicchi, Francesco Moretti, Eugenio Mascaro, Pietro Musso, Ayoub Raji, Marko Bertogna, Vincenzo Maria Arricale, Angelo Lo Sapio, Alessandro Piccarelli, Garron Fish

Differential Geometry

A Classification of Complete Self-shrinkers

Let $X:M^n\to\R^{n+1}$ be an $n$-dimensional complete self-shrinker. We obtain a complete classification of complete self-shrinkers with positive constant scalar curvature. More precisely, we prove that the round sphere $S^n(\sqrt n)$ and the standard generalized cylinder $S^k(\sqrt k)\times \R^{n-k}$ for $2\leq k\leq n-1$ are the only complete self-shrinkers with positive constant scalar curvature. The key difficulty is to characterize the residual case in Cheng-Li-Wei \cite{CLW}: $R>0$, $S<1$, and $\sup_M S=1$, where $R$ and $S$ denote the scalar curvature and the squared norm of the second fundamental form, respectively. For this case, it seems a hard task that the generalized maximum principle yields a useful information. In order to overcome this substantial difficulty, our key ingredient is to get a uniform positive lower bound for the Bakry-Émery Ricci curvature so that we can make use of the comparison theorem of Wei-Wylie \cite{WeiWylie} to conclude that the Gaussian volume is finite. Furthermore, the gap theorems on $S$ are given.
Read the full abstract

Machine Learning

VertiFuseX: Generalizable Financial Forecasting via Multi-Stream Temporal Fusion

Stock price prediction remains challenging due to the non-stationary and noisy nature of financial time series. Existing deep learning models often rely on rigid decision-level fusion, ad hoc hyperparameter tuning, and compressed final-layer outputs, causing information loss, overfitting, and limited cross-market generalization. We propose VertiFuseX, a hybrid LSTM architecture using penultimate-layer vertical fusion of multi-scale temporal representations. VertiFuseX stacks and reweights penultimate features from LSTM, Bi-LSTM, and St-LSTM branches, integrates a parallel DNN stream, and jointly optimizes all components via backpropagation under a fixed hyperparameter configuration. This preserves richer intermediate temporal information across scales. Evaluated on 15 years (2010-2024) of closing prices from 10 global equity indices using strict chronological out-of-sample testing with the final 365 trading days held out, VertiFuseX achieves 30-54% MAPE reductions and over 40% improvements in MAE and RMSE versus LSTM-based baselines, and outperforms seven state-of-the-art models across 33 metric-dataset comparisons. Ablation studies confirm penultimate-layer fusion drives these gains over final-layer fusion and decision-level ensembling. Gradient-based saliency analysis shows consistent emphasis on mid-range dependencies at lags 9-15 days. Economic validation via algorithmic trading simulation under extreme market regimes shows reduced maximum drawdowns and superior risk-adjusted returns. With 675k parameters, a 2.6 MB memory footprint, and 1.5 ms/sample inference latency, VertiFuseX offers a lightweight, interpretable, deployment-ready framework for robust financial forecasting.
Read the full abstract

Game Theory

A Randomized $\frac32$-Approximation for Strategic Facility Location on a Circle

We study strategyproof mechanisms for locating a single facility on a circle so as to serve a set of strategic agents under the utilitarian social cost objective. We analyze a simple parity-dependent mechanism that randomizes between the Random Dictator (RD) mechanism and the Proportional Circle Distance (PCD) mechanism. For an odd number \(n\) of agents, this mechanism mixes RD and PCD with equal probability, as originally proposed by Rogowski and Dziubiński (IJCAI 2025). For even \(n\), it mixes RD with a random-deletion extension of PCD, in which one agent is removed uniformly at random before PCD is applied to the remaining agents. Our main result shows that this mechanism is strategyproof and achieves an approximation ratio of \(\frac32\) for every \(n\ge 3\). This improves the \(\frac74\) upper bound of Rogowski and Dziubiński, which applied only to odd \(n\), and extends the guarantee to even numbers of agents. Finally, we establish a computer-assisted lower bound of \(1.088187\ldots\) on the approximation ratio of any randomized strategyproof mechanism, improving on the previous lower bound of \(1.0456\) due to Meir (SAGT 2019).
Read the full abstract

Computation and Language

Cognition on Graph: Navigating Massive Knowledge Space via Cognitive Cycles and Bidirectional Graph-Text Synergy

Retrieval-Augmented Generation (RAG) has empowered Large Language Models (LLMs) to tackle knowledge-intensive tasks. However, navigating global, heterogeneous knowledge bases (large-scale knowledge graphs and text corpora) for complex reasoning remains a challenge. Existing methods typically employ reactive, graph-driven exploration strategies, which blindly follow graph topology without adapting to the question context or evolving exploration progress, and lack deep bidirectional synergy between graph and text. To address these limitations, we propose CoG (Cognition on Graph), a cognitive-inspired, training-free framework for adaptive knowledge exploration. Drawing inspiration from human problem-solving, CoG performs a continuous plan-explore-reflect cycle, where it proactively formulates investigation plans, performs dual-source retrieval, and dynamically reflects on progress to adjust strategies. Crucially, it establishes deep bidirectional synergy between structured graph and unstructured text, where entities extracted from text dynamically guide graph exploration to bridge knowledge gaps. Extensive experiments on seven multi-hop QA benchmarks demonstrate that CoG significantly outperforms state-of-the-art methods while achieving superior exploration efficiency. Our code and datasets are available at https://github.com/zhougengxian/CoG.
Read the full abstract

All authors Gengxian Zhou, Jian Xu, Zichen Tang, Shiming Xiang, Haihong E, Cheng-Lin Liu

Probability

One marked-sum law determines random mass partitions and $\Xi$-coalescents

Let $P=(P_j)$ be a random mass partition and let $(X_j)$ be iid real marks, independent of $P$. We construct a fixed mark distribution for which the law of the real random variable $\sum_j P_j X_j$ determines the law of $P$ on the Kingman simplex. The observation map is an affine topological embedding. One construction uses an infinite convolution of stable laws with rationally independent indices; for every prescribed $q\in(0,\infty)$, a determining mark can instead be chosen symmetric, centered, compound Poisson, and in $L^q$. At any known positive time, the law of the ranked block frequencies of a $\Xi$-coalescent started from singletons$\unicode{x2014}$equivalently, the exchangeable partition probability functions of all its finite restrictions at that time$\unicode{x2014}$determines the finite collision measure $\Xi$. Composing the two results therefore identifies $\Xi$ from the law of one real marked sum. Within the $\Lambda$ subclass, an asymmetric Bernoulli coloring determines the normalized collision measure, as does any centered nonconstant mark whose absolute moment of some order $a\in(1,4)$ is finite while every higher absolute moment is infinite. We also prove that, for every $d\ge 2$, a pair of distinct laws supported on strictly decreasing $(d+1)$-tuples of positive masses with a fixed total mass remains indistinguishable for every mark supported on at most $d$ points. For every fixed number $d\ge 2$ of types, this obstruction yields distinct normalized collision measures with the same mutation-free neutral $d$-type $\Xi$-Fleming$\unicode{x2013}$Viot transition semigroup.
Read the full abstract

2609.12790 · 11 Sep · 32 pages, no figures. Companion paper to "Breiman's conjecture and normalized jumps of subordinators" (arXiv:2608.24616v3). Identification from one marked-sum law is developed for random mass partitions and $\Xi$-coalescents, together with finite-support and finite-type non-identifiability results

Number Theory

Gauss Genus Theory in Characteristic 2

We extend Gauss composition and Gauss genus theory over $\mathbf{Z}$ to $\mathbb{F}_{2^n}[T]$, a polynomial ring over a finite field $\mathbb{F}_{2^n}$ of characteristic 2. We find new invariants of binary quadratic forms over $\mathbb{F}_{2^n}[T]$ by using Arf invariant and introduce new definitions of proper equivalence and direct composition, and prove that the direct composition makes the set of proper equivalence classes of binary quadratic forms with the same invariants into a finite Abelian group, which is isomorphic to a Picard group of a corresponding extension ring of $\mathbb{F}_{2^n}[T]$. Building on this, we develop genus theory in characteristic 2 and prove that the kernel of the generalized Gauss's map is the subgroup of all squares in the class group.
Read the full abstract

Computational Engineering

Transducer Placement and the Limits of a Four-State Reduced Model in Post-Flutter Piezoelectric Energy Harvesting from a Pitch-Plunge-Flap Aerofoil

Aeroelastic ?utter is normally a failure mode to be designed against, yet the limit-cycle oscillations (LCOs) that follow it convert flow energy into sustained structural motion that a piezoelectric transducer can turn into electrical power. A transducer is embedded in a three-degree-of-freedom pitch-plunge aerofoil with a finite-mass trailing-edge flap and unsteady strip-theory aerodynamics, giving a fifteen-state electro-aeroelastic system with a cubic hardening pitch spring, and the system is reduced by biorthonormal projection of the Taylor-expanded residual onto eigenvectors of the coupled Jacobian. The testbed is deliberately low order, so that every reduced prediction can be checked against the full-order system it replaces. The degree of freedom that carries the transducer is a first-order design variable, since it sets both the sign of the shift in the flutter boundary and the magnitude of the harvested power. The error of the reduced model is dominated not by the size of the retained basis but by how the reduced operator is made to depend on low speed. Expanding the retained eigenvalues leaves the reduced operator block-diagonal, with no coupling between the retained modes at any order of the expansion, whereas re-projecting the exact Jacobian onto the same frozen basis supplies that coupling and restores the structural response without enlarging the basis. It is found that the error that remains lies in the harvested voltage, which the equilibrium eigenvectors carry in a fixed proportion to the motion, and this is not removed by adding the electrical modes to the basis, whereas the flap detection, which the same basis also misses, is recovered by retaining the flap modes. The accuracy of a harvester reduced-order model must therefore be reported on the harvested quantity itself.
Read the full abstract

Computer Vision

LG-PF: Lightweight Confidence-Guided Polarization Image Fusion

Polarization image fusion combines the stable luminance and structural information of the total- intensity image S0 with the material-sensitive details of the degree of linear polarization (DoLP) image. However, the reliability of DoLP varies spatially, and indiscriminate polarization transfer may amplify unstable responses or disturb the structural appearance anchored by S0. We therefore propose LG-PF, a lightweight confidence-guided framework that formulates polarization fusion as a selective residual transfer process. A Polarization Confidence Prior estimates spatially reliable polarization responses, a Mask-guided Multi-scale Fusion module regulates their transfer across three feature scales, and a Lightweight Context-aware Bounded Correction Head stabilizes local photometric and structural transitions. Confidence guidance is also incorporated into the optimization objectives to preserve reliable polarization details while suppressing unsupported responses. We also construct MSP, a multi-scene polarization fusion dataset containing 1000 pixel-aligned image pairs from 17 indoor and outdoor scene categories. LG-PF achieves the best results across all six evaluated metrics on MSP, while subset-based evaluations on PIF and GAND show promising transferability without fine-tuning. With only 0.2936 M parameters and an inference time of 21.712 ms per image, LG-PF achieves competitive fusion quality with low computational cost. The source code will be available at https://github.com/1hzf/LG-PF.
Read the full abstract

All authors Zhuangfan Huang, Zhenyu Kuang, Gao Wang, Yang Liu, Haishu Tan, Xiaosong Li