November 2025 arXiv papers — page 190

Alvin Wei Ming Tan, Ben Prystawski, Veronica Boyce, Michael C. Frank

Iterated reference games - in which players repeatedly pick out novel referents using language - present a test case for agents' ability to perform context-sensitive pragmatic reasoning in multi-turn linguistic environments. We tested humans and vision-language models on trials from iterated reference games, varying the given context in terms of amount, orde

Liam Bakar, Zachary Englhardt, Vidya Srinivas, Girish Narayanswamy

Food logging, both self-directed and prescribed, plays a critical role in uncovering correlations between diet, medical, fitness, and health outcomes. Through conversations with nutritional experts and individuals who practice dietary tracking, we find current logging methods, such as handwritten and app-based journaling, are inflexible and result in low adh

Simon Escobar Steinvall, Hampus Thulin, Nico Kawashima, Francesco Salutari

To enable lightweight and flexible solar cell applications it is imperative to develop direct bandgap absorber materials. Moreover, to enhance the potential sustainability impact of the technologies there is a drive to base the devices on earth-abundant and readily available elements. Herein, we report on the epitaxial growth of Zn3P2 nanowires using exclusi

G. D. Muro, C. M. S. Cohen, Z. Xu, R. A. Leske

During mid-May 2024, active region (AR) 13664 produced a series of M- and X-class flares along with several coronal mass ejections (CMEs) that resulted in exceptionally strong aurora at Earth. This study presents in-situ solar energetic particle (SEP) ion composition data from Solar Terrestrial Relations Observatory Ahead (STA), Advanced Composition Explorer

Zheng-Hao Xu, Yi-Xian Chen, Douglas N. C. Lin

Stars embedded in the inner pc region of an active galactic nucleus (AGN) experience extreme accretion conditions that significantly alter their evolution. We present one-dimensional MESA simulations of stars growing and decaying within AGN disks, implementing radiative-feedback-regulated accretion which limits stellar growth near the Eddington luminosity, a

Niloufar Yousefi, John W. Simpson-Porco

Feedback-based optimization (FBO) provides a simple control framework for regulating a stable dynamical system to the solution of a constrained optimization problem in the presence of exogenous disturbances, and does so without full knowledge of the plant dynamics. However, closed-loop stability requires the controller to operate on a sufficiently slower tim

Abdesslam Arhrib, Ayoub Hmissou, Stefano Moretti, Larbi Rahili

In this paper, pair production of Standard Model (SM)-like Higgs bosons, $hh$, is studied through $\gamma\gamma$ scattering at future electron-positron colliders, in the framework of the Inert Doublet Model with two Active Doublets, i.e., the I(1+2)HDM for short. The relevance of the process $\gamma\gamma\to hh$ for such a Beyond the SM (BSM) scenario stems

Manh Nguyen, Sunil Gupta, Dai Do, Hung Le

Hallucination mitigation remains a persistent challenge for large language models (LLMs), even as model scales grow. Existing approaches often rely on external knowledge sources, such as structured databases or knowledge graphs, accessed through prompting or retrieval. However, prompt-based grounding is fragile and domain-sensitive, while symbolic knowledge

N. J. Lambert, A. Schumer, J. J. Longdell, S. Rotter

The amplitude of resonant oscillations in a non-Hermitian environment can either decay or grow in time, corresponding to a mode with either loss or gain. When two coupled modes have a specific difference between their loss or gain, a feature termed an exceptional point emerges in the excitations' energy manifold, at which both the eigenfrequencies and eigenm

Itai Maimon

Classic Rock-Paper-Scissors, RPS, has seen many variants and generalizations in the past several years. In the previous paper, we defined playability and balance for games. We used these definitions to show that different forms of imbalance agree on the most balanced and least balanced form of playable two-player n-object RPS games, referred to as (2,n)-RPS.

Thomas J. Harris, Prescott C. Alexander, Anh B. D. Pham, Joseph Tuccillo

Social contact patterns are a key input to many infectious disease models. Contact surveys, where participants are asked to provide information on their recent close and casual contacts with others, are one of the standard methods to measure contact patterns in a population. Surveys that require detailed sociodemographic descriptions of contacts allow for th

Haithem Taha, Kshitij Anand

Gauss's principle of least constraint transforms a dynamics problem into a pure minimization problem, where the total magnitude of the constraint force is the cost function, minimized at each instant. Newton's equation is the first-order necessary condition for minimizing the Gaussian cost, subject to the given kinematic constraints. The principle of minimum

Sraddha Agrawal, Stephen Whitelam, Pierre Darancet

We address the problem of active optical steering of structural phase transitions in solids. We demonstrate that existing reinforcement learning approaches can derive optimal time-dependent electric fields in optically-driven dissipative classical systems far beyond the harmonic regime, enabling the stabilization of non-thermal structural phases. Our approac

Martin Guillot, Cédric Blanchard, Martina Morassi, Aristide Lemaître

Recent discoveries in semi-metallic multi-gap systems featuring band singularities have galvanized enormous interest in particular due to the emergence of non-Abelian braiding properties of band nodes. This previously uncharted set of topological phases necessitates novel approaches to probe them in laboratories, a pursuit that intricately relates to evaluat

Adam M. Saunders, Lucas W. Remedios, Elyssa M. McMaster, Jongyeon Yoon

Diffusion-weighted magnetic resonance imaging allows for reconstruction of models for structural connectivity in the brain, such as fiber orientation distribution functions (ODFs) that describe the distribution, direction, and volume of white matter fiber bundles in a voxel. Crossing white matter fibers in voxels complicate analysis and can lead to errors in

Chiraag Kaushik, Justin Romberg, Vidya Muthukumar

We present simple, user-friendly bounds for the expected operator norm of a random kernel matrix under general conditions on the kernel function $k(\cdot,\cdot)$. Our approach uses decoupling results for U-statistics and the non-commutative Khintchine inequality to obtain upper and lower bounds depending only on scalar statistics of the kernel function and a

Hlali Azzeddine, Majid Ben Yakhlef, Soulaiman El Hazzat

Small, imbalanced datasets and poor input image quality can lead to high false predictions rates with deep learning models. This paper introduces Class-Based Image Composition, an approach that allows us to reformulate training inputs through a fusion of multiple images of the same class into combined visual composites, named Composite Input Images (CoImg).

Linchen Qian, Jiasong Chen, Ruonan Gong, Wei Sun

Accurate geometric modeling of the aortic valve from 3D CT images is essential for biomechanical analysis and patient-specific simulations to assess valve health or make a preoperative plan. However, it remains challenging to generate aortic valve meshes with both high-quality and consistency across different patients. Traditional approaches often produce tr

M. M. Katsova, V. N. Obridko, D. D. Sokoloff, N. V. Emelianov

Solar activity is a process driven by many independent but interconnected phenomena. Although the 11-year cycle is the result of operation of the dynamo mechanism, the cause of longer secular variations is not clear. In search of such a cause, it was proposed to take into account the influence of the planetary system. In order to verify the idea, we consider

Robert Alonzo Lyman

Perhaps the fundamental theorem of geometric group theory, the Milnor--Schwarz lemma gives conditions under which the orbit map relating the geometry of a geodesic metric space and the word metric on a group acting isometrically on the space is a quasi-isometry. Pioneering work of Rosendal makes these and other techniques of geometric group theory applicable

Erdem Sucu, Kuantay Boshkayev, Yassine Sekhmani, İzzet Sakallı

We explore charged black holes in Scalar-Tensor-Vector Gravity (STVG), unveiling their distinctive features across multiple physical domains. Our topological analysis reveals that the STVG coupling parameter $\alpha$ bolsters thermal stability while electromagnetic charge $Q$ weakens it. Using the Gauss-Bonnet theorem, we find that $\alpha$ amplifies light d

Shayne Waldron

If a (weighted) spherical design is defined as an integration (cubature) rule for a unitarily invariant space P of polynomials (on the sphere), then any unitary image of it is also such a spherical design. It therefore follows that such spherical designs are determined by their Gramian (Gram matrix). We outline a general method to obtain such a characterisat

Doron Sayag, Avichai Snir, Daniel Levy

In 1991 and 2008, Israel abolished the equivalents of 1-cent and 5-cent coins, respectively, effectively eliminating low-denomination coins and introducing rounding in cash transactions. When totals were rounded up, shoppers incurred a small rounding tax. Using detailed data on price endings and basket sizes across supermarkets, drugstores, small groceries,

Joaquin Torres Henestroza

Following the ideas of Bossinger and Fang, Fourier, and Littelman, we study iterated sequences for the Grassmannian $\operatorname{Gr} (3, n)$ as a special class of birational sequences. For each iterated sequence $S$, there is a weighting matrix $M_{S}$ corresponding to a valuation on the rational coordinate ring and we show that the initial form of a Pl\"{

Kaihua Ji, Luning Sun, Shusen Liu, Fei Zhou

Microstructural pattern formation, such as dendrite growth, occurs widely in materials and energy systems, significantly influencing material properties and functional performance. While the phase-field method has emerged as a powerful computational tool for modeling microstructure dynamics, its high computational cost limits its integration into practical m

Chengxuan Xia, Qianye Wu, Hongbin Guan, Sixuan Tian

Large language models (LLMs) have achieved impressive results in high-resource languages like English, yet their effectiveness in low-resource and morphologically rich languages remains underexplored. In this paper, we present a comprehensive evaluation of seven cutting-edge LLMs -- including GPT-4o, GPT-4, Claude~3.5~Sonnet, LLaMA~3.1, Mistral~Large~2, LLaM

Jorge de Blas, Monica Dunford, Emanuele Bagnaschi, Ayres Freitas

The European Strategy for Particle Physics (ESPP) reflects the vision and presents concrete plans of the European particle physics community for advancing human knowledge in fundamental physics. The ESPP is updated every five-to-six years through a community-driven process. It commences with the submission of specific proposals and other input from the commu

Anton Nazarov, Matvey Sushkov

We consider random Young diagrams with respect to the measure induced by the decomposition of the $p$-th exterior power of $\mathbb{C}^{n}\otimes \mathbb{C}^{k}$ into irreducible representations of $GL_{n}\times GL_{k}$. We demonstrate that transition probabilities for these diagrams in the limit $n,k,p\to\infty$ with $p\sim nk$ converge to the large $N$ lim

Hellina Hailu Nigatu, Bethelhem Yemane Mamo, Bontu Fufa Balcha, Debora Taye Tesfaye

As low-resourced languages are increasingly incorporated into NLP research, there is an emphasis on collecting large-scale datasets. But in prioritizing quantity over quality, we risk 1) building language technologies that perform poorly for these languages and 2) producing harmful content that perpetuates societal biases. In this paper, we investigate the q

Sergey N. Solodukhin, Vagif Tagiev

We identify a general class of spacetime metrics that mimic the properties of black holes without possessing a true event horizon. These metrics are constrained by the requirements of being singularity-free and geodesically complete. Specifically, we study metrics that do not possess $Z_2$ symmetry and may deviate slightly or significantly from the symmetric

Suraj Prasai, Mengnan Du, Ying Zhang, Fan Yang

We develop KnowThyself, an agentic assistant that advances large language model (LLM) interpretability. Existing tools provide useful insights but remain fragmented and code-intensive. KnowThyself consolidates these capabilities into a chat-based interface, where users can upload models, pose natural language questions, and obtain interactive visualizations

Julian Frattini, Hans-Martin Heyn, Robert Feldt, Richard Torkar

As any scientific discipline, the software engineering (SE) research community strives to contribute to the betterment of the target population of our research: software producers and consumers. We will only achieve this betterment if we manage to transfer the knowledge acquired during research into practice. This transferal of knowledge may come in the form

Fariba Hosseinynejad, Pavithran Iyer, Guillaume Dauphinais, David L. Feder

Physical Gottesman-Kitaev-Preskill (GKP) states are inherently noisy as ideal ones would require infinite energy. While this is typically considered as a deficiency to be actively corrected, this work demonstrates that imperfect GKP stabilizer states can be leveraged in order to apply non-Clifford gates using only linear optical elements. In particular, Gaus

Lucas E Sant'Anna, Rohita Roy, Janella C Schwab, Julian I Perez

Chimeric Antigen Receptor (CAR) T cell therapy has transformed immunotherapy for resistant cancers, yet it faces major limitations such as lack of persistence, toxicity, exhaustion, and antigen-negative relapse. Enhancing CAR T cells with genetic circuitry and synthetic receptors offers solutions to some of these problems, but often the theoretical design sp

Greg Markowsky, Clayton McDonald

In this paper, we use purely complex analytic techniques to prove two results of the first author which were hitherto given only probabilistic proofs. A general form of the Phragm\'en-Lindel\"of principle states that if the $p$\textsuperscript{th} Hardy norm of the conformal map from the disk to a simply connected domain is finite, then an analytic function

Hokuto Iwakiri, Keita Kanno

The hardware-efficient ansatz (HEA) is one of the most important class of parametrized quantum circuits for near-term applications of quantum computing. We show that the problem of simulating some major classes of the HEA is BQP-complete by explicitly demonstrating that any relevant quantum circuit can be efficiently represented as an HEA circuit of those cl

Ganna Kudryavtseva

We introduce a general framework, based on \'etale topological categories, for studying discrete restriction semigroups and their algebras. Generalizing Paterson's universal groupoid of an inverse semigroup, we define the universal category ${\mathscr C}(S)$ of a restriction semigroup $S$ with local units as the category of germs of the spectral action of $S

J. R. Meyer, I. Vurgaftman, S. -Q. Yu, R. Q. Yang

Semiconductor photonic devices operating in the midwave infrared (mid-IR, which we roughly define here as wavelengths spanning 3 to 14 microns) uniquely address a wide range of current practical needs. These include chemical sensing, environmental monitoring, industrial process control, medical diagnostics, thermal imaging, LIDAR, free space optical communic

Arijit Bhattacharjee, Ali TehraniJamsaz, Le Chen, Niranjan Hasabnis

Recent advances in large language models (LLMs) have significantly accelerated progress in code translation, enabling more accurate and efficient transformation across programming languages. While originally developed for natural language processing, LLMs have shown strong capabilities in modeling programming language syntax and semantics, outperforming trad

H. Gashmard, H. Shakeripour, M. Alaei

Inspired by nature, this study employs the Materials Genome Initiative to identify key components of HTSC superconductors. Integrating AI with high-throughput screening, we uncover crucial superconducting "genes". Through HTS techniques and advanced machine learning models, we demonstrate that Functional Convolutional Neural Networks (CNNs) ensure accurate e

Noga Alon, Martin Milanič, Paweł Rzążewski

We study two graph parameters defined via tree decompositions: tree-independence number and induced matching treewidth. Both parameters are defined similarly as treewidth, but with respect to different measures of a tree decomposition $\mathcal{T}$ of a graph $G$: for tree-independence number, the measure is the maximum size of an independent set in $G$ incl

Olha Silina

We study the perfect matching lattice of a matching covered graph $G$, generated by the incidence vectors of its perfect matchings. Building on results of Lov\'asz and de Carvalho, Lucchesi, and Murty, we give a polynomial-time algorithm based on polyhedral methods that constructs a lattice basis for this lattice consisting of perfect matchings of $G$. By de

Giulio Del Felice, Ludovico Theo Giorgini

Multiscale dynamical systems characterized by interacting fast and slow processes are ubiquitous across scientific domains, from climate dynamics to fluid mechanics. Accurate modeling of such systems requires capturing both the long-term statistical properties governed by slow variables and the short-term transient dynamics driven by fast chaotic processes.

Christian Roettger, Xuyi Ren

The \textit{ternary digits of $2^n$} are a finite sequence of 0s, 1s, and 2s. It is a natural question to ask whether the frequency of any string of 0s, 1s, and 2s in this sequence approaches the same limit for all strings of the same length, as the exponent $n$ approaches infinity (\textit{Uniform Distribution in the limit}). Currently the answer to this qu

Claudio U. Hail, Lior Michaeli, Harry A. Atwater

Achieving spatiotemporal control of light at subwavelength and subcycle scales is an important milestone in the development of new photonic materials and technologies. Ultrafast spatiotemporal light modulation currently relies on electronic interband and intraband transitions that yield pronounced refractive index changes but typically suffer from slow, pico

Tammy Mackenzie, Sukriti Punj, Natalie Perez, Sreyoshi Bhaduri

This paper examines how decision makers in academia, government, business, and civil society navigate questions of power in implementations of artificial intelligence. The study explores how individuals experience and exercise levers of power, which are presented as social mechanisms that shape institutional responses to technological change. The study repor

Walter Quiñonez, María José Sánchez, Diego Rubi

We present a systematic study of memristor based neural networks trained with the hardware-friendly Manhattan update rule, focusing on the trade offs between learning performance and energy consumption. Using realistic models of potentiation/depression (P/D) curves, we evaluate the impact of nonlinearity (NLI), conductance range, and number of accessible lev

M. Lupi, R. Ballabriga, F. N. Bandi, G. Bergamin

The FastRICH ASIC provides high-precision, triggerless readout for the LS3 Enhancements and Upgrades II of the LHCb RICH detector. The demands of continuous data acquisition and varying hit rates across the detector impose unique challenges on the ASIC's design and verification. This work presents the verification strategy for FastRICH, focusing on functiona

Yanahan Paramalingam, Adam Noel

A biofilm is a self-contained community of bacteria that uses signaling molecules called autoinducers (AIs) to coordinate responses through the process of quorum sensing. Biofilms exhibit a dual role that drives interest in both combating antimicrobial resistance (AMR) and leveraging their potential in bioprocessing, since their products can have commercial

Duong Mai, Lawrence Hall

Deep learned (DL) models for image recognition have been shown to fail to generalize to data from different devices, populations, etc. COVID-19 detection from Chest X-rays (CXRs), in particular, has been shown to fail to generalize to out-of-distribution (OOD) data from new clinical sources not covered in the training set. This occurs because models learn to

Giorgio Stefani

Given $\alpha\in(0,1)$ and a set $E\subset\mathbb{R}^N$ with locally finite fractional $\alpha$-variation, we show that for $|D^\alpha\mathbf 1_E|$-a.e. $x$, every non-trivial tangent set of $E$ at $x$ with locally finite integer perimeter is a half-space oriented by the fractional inner unit normal of $E$ at $x$.

Joong M. Park, Martin Mootz, Richard H. J. Kim, Zhixiang Chong

Even in the quantum limit, non-equilibrium quasiparticle (QP) populations induce QP poisoning that irreversibly relaxes the quantum state and significantly degrades the coherence of transmon qubits. A particularly detrimental yet previously unexplored mechanism arises from QP multiplication facilitated by vortex trapping in superconducting quantum circuits,

Gabriel Rabelo-Soares, Gojko Vujanovic, Giorgio Torrieri

The experimental observation of collective behavior in proton-proton and proton-nucleus collisions poses a fundamental theoretical question regarding the proper characterization of the initial state underlying hydrodynamic evolution. While relativistic hydrodynamics requires an initial condition characterized by an entropy current, corresponding to a maximal

Shakil Ahmed, Amika Tabassum, Ibrahim Almazyad, Ashfaq Khokhar

With the rapid advancement of 6G technology, the Tactile Internet is emerging as a novel paradigm of interaction, particularly in intelligent transportation systems, where stringent demands for ultra-low latency and high reliability are prevalent. During the transmission and coordination of autonomous vehicles, malicious adversaries may attempt to compromise

Phuc Nguyen, Josiah Couch, Rahul Bansal, Alexandra Morgan

Shannon entropy is not the only entropy that is relevant to machine-learning datasets, nor possibly even the most important one. Traditional entropies such as Shannon entropy capture information represented by elements' frequencies but not the richer information encoded by their similarities and differences. Capturing the latter requires similarity-sensitive

Arthemy V. Kiselev

The generalised Wronskian of differential order $k\geqslant 1$ for $N$ functions $f_1$, $\ldots$, $f_N$ in $d\geqslant 1$ independent variables $x^1$, $\ldots$, $x^d$ is the determinant of the matrix with these functions' derivatives $\partial^{|\sigma_i|} f_j / \partial (x^1)^{\sigma_i^1}\cdots \partial (x^d)^{\sigma_i^d}$ (of orders $0 \leqslant |\sigma_i|

David Altizio

The classical Stern sequence of positive integers was extended to a polynomial sequence $S_n(\lambda)$ by Klav\v{z}ar et. al. by defining $S_0(\lambda) = 0$, $S_1(\lambda) = 1$, and $$S_{2n}(\lambda) = \lambda S_n(\lambda),\quad S_{2n+1}(\lambda) = S_n(\lambda) + S_{n+1}(\lambda).$$ Dilcher et. al. conjectured that all roots of $S_n(\lambda)$ lie in the half

Ilya Vilkoviskiy, Michael Sonner, Qi Camm Huang, Wen Wei Ho

Temporal entanglement (TE) of an influence matrix (IM) has been proposed as a measure of complexity of simulating dynamics of local observables in a many-body system. Foligno et al. [Phys. Rev. X 13, 041008 (2023)] recently argued that the TE in chaotic 1d quantum circuits obeys linear (volume-law) scaling with evolution time. To reconcile this apparent high

Tianning Dong, Luyi Ma, Varun Vasudevan, Jason Cho

Multimodal Large Language Models (MLLMs) are reshaping how modern agentic systems reason over sequential user-behavior data. However, whether textual or image representations of user behavior data are more effective for maximizing MLLM performance remains underexplored. We present \texttt{BehaviorLens}, a systematic benchmarking framework for assessing modal

Yuran Ding, Xinwei Chen, Xiaofan Zhang, Zongwei Zhou

Optimizing large-language model (LLM) training on distributed domain-specific accelerator systems presents significant challenges due to its complex optimization space. Existing optimization methods, however, rely on time-consuming manual tuning or resource-intensive black-box searches, which struggle to keep pace with the rapidly evolving LLM domain, leadin

Ira Wolfson

Loschmidt's paradox asks why macroscopic irreversibility is universal despite the time-reversal symmetry of microscopic dynamics. We argue that irreversibility is not a property of the dynamics but of accessibility: chaotic evolution drives phase-space structure below the quantum resolution scale $\ell_\hbar$, at a critical time $t_c = \lambda^{-1}\ln(\delta

Niels Böttner, Roman Schnabel, Mikhail Korobko

Disturbances in gravitational wave (GW) observational data are often caused by non-stationary noise in the detector itself, such as back-scattering of laser stray light into the signal field. Unlike GW signals, non-stationary noise can appear in both the GW-signal quadrature and the orthogonal quadrature, which is usually not measured. Simultaneous sensing o

Yedidel Louck, Ariel Stulman, Amit Dvir

Multi-agent systems (MAS) powered by artificial intelligence (AI) are increasingly foundational to complex, distributed workflows. Yet, the security of their underlying communication protocols remains critically under-examined. This paper presents the first empirical, comparative security analysis of the official CORAL implementation and a high-fidelity, SDK

Sicheng He, Max Howell, Daning Huang, Eirikur Jonsson

Many physical systems exhibit limit cycle oscillations induced by Hopf bifurcations. In aerospace engineering, limit cycle oscillations arise from undesirable Hopf bifurcation phenomena such as aeroelastic flutter and transonic buffet. In some cases, the resulting limit cycle oscillations can themselves be unstable, leading to amplitude divergence or hystere

Corey Sargent, William Clark, Antonia Seifert, Alicia Mand

We examine the claimed observations of a gravitational external field effect (EFE) reported in Chae et al. We show that observations suggestive of the EFE can be interpreted without violating Einstein's equivalence principle, namely from known correlations between morphology, environment and dynamics of galaxies. While Chae et al's analysis provides a valuab

Eugene Koreshin, Denis Sakhno, Jim A. Enriquez, Pavel A. Belov

We present a single-shot near-field technique, where the near-field scan is performed on a single sample without repeating measurements or averaging over multiple samples, to reconstruct the isofrequency surfaces of metamaterials in the microwave regime. In our approach, we excite resonant modes using a fixed source in a resonator composed of the material un

Saúl Fenollosa, Narcis Cardona, Wenfei Yang, Jian Li

Integrated Sensing and Communication (ISAC) is critical for efficient spectrum and hardware utilization in future wireless networks like 6G. However, existing channel models lack comprehensive characterization of ISAC-specific dynamics, particularly the relationship between mono-static (co-located Tx/Rx) and bi-static (separated Tx/Rx) sensing configurations

Brandon Redding, Joseph B. Murray, Matthew J. Murray, Ross T. Schermer

Designing miniaturized optical spectrometers is an increasingly active area of research as spectrometers are crucial components for a wide range of applications including chemical and material analysis, medical diagnostics, classical and quantum sensing, characterization of light sources, and radio frequency (RF) spectrum analysis. Among these applications,

Daniel C. M. Palumbo, George N. Wong, Angelo Ricarte

Light orbiting an accreting black hole may impact the disk or jet multiple times before escaping to the observer, at a variety of angles with respect to the local magnetic field. In this letter, we characterize the imprints of these long path lengths and disparate magnetic field impacts in synchrotron spectra of hot accretion disks, as the strongly lensed ``

Christian Kuehn, Giacomo Landi

We study interacting particle systems of Kuramoto-type. Our focus is on the dynamical relation between the partial differential equation (PDE) arising in the continuum limit (CL) and the one obtained in the mean-field limit (MFL). Both equations arise when we are considering the limit of infinitely many interacting particles but the classes of PDEs are struc

Catherine C. Blume, Bradley W. Hindman

Though Rossby waves have been observed on the Sun, their radial eigenfunctions remain a mystery. The prior theoretical work either considers quasi-2D systems, which do not apply to the solar interior, or only considers fully radiative or fully convective atmospheres. This project calculates the radial eigenfunctions for Rossby waves in a deep atmosphere for

James Enouen, Yujia Zheng, Ignavier Ng, Yan Liu

Causal structure learning has long been the central task of inferring causal insights from data. Despite the abundance of real-world processes exhibiting higher-order mechanisms, however, an explicit treatment of interactions in causal discovery has received little attention. In this work, we focus on extending the causal additive model (CAM) to additive mod

Mikołaj Langner, Jan Eliasz, Ewa Rudnicka, Jan Kocoń

We introduce a method for efficient multi-label text classification with large language models (LLMs), built on reformulating classification tasks as sequences of dichotomic (yes/no) decisions. Instead of generating all labels in a single structured response, each target dimension is queried independently, which, combined with a prefix caching mechanism, yie

Ilia Moiseenko, Zhanna Devizorova, Olga Polischuk, Viacheslav Muravev

We demonstrate that the slot between parallel metal gates placed above two-dimensional electron system (2DES) forms a plasmonic cavity with unconventional mode quantization. The resonant plasmon modes are excited when the slot width $L$ and the plasmon wavelength $\lambda$ satisfy the condition $L = \lambda/8 +n \times \lambda/2$, where $n=0, 1, 2 \ldots$. T

Mohammad Atif Quamar, Mohammad Areeb, Mikhail Kuznetsov, Muslum Ozgur Ozmen

Aligning large language models (LLMs) with human values is crucial for safe deployment. Inference-time techniques offer granular control over generation; however, they rely on model uncertainty, meaning an internal estimate of how likely the model believes its next tokens or outputs are correct, for segmentation. We show that this introduces two critical lim

Ahmed Mostafa, Raisul Arefin Nahid, Samuel Mulder

Tokenization is fundamental in assembly code analysis, impacting intrinsic characteristics like vocabulary size, semantic coverage, and extrinsic performance in downstream tasks. Despite its significance, tokenization in the context of assembly code remains an underexplored area. This study aims to address this gap by evaluating the intrinsic properties of N

Ryien Hosseini, Filippo Simini, Venkatram Vishwanath, Rebecca Willett

Graph Neural Networks learn on graph-structured data by iteratively aggregating local neighborhood information. While this local message passing paradigm imparts a powerful inductive bias and exploits graph sparsity, it also yields three key challenges: (i) oversquashing of long-range information, (ii) oversmoothing of node representations, and (iii) limited

Jan Kocoń, Maciej Piasecki, Arkadiusz Janz, Teddy Ferdinan

Large Language Models (LLMs) play a central role in modern artificial intelligence, yet their development has been primarily focused on English, resulting in limited support for other languages. We present PLLuM (Polish Large Language Model), the largest open-source family of foundation models tailored specifically for the Polish language. Developed by a con

Ayush Subedi, Anthony Torres, Jeff Ganley

In Part I of this study, vibration assisted heat treatments of ZBLAN glass revealed irregular crystallization at higher vibration levels, attributed to intermittent loss of thermal contact between the sample and the inner silica ampoule wall. The present work (Part II) investigates this mechanism through finite element modeling (FEM) and experimental validat

Zhiguo Ding, Robert Schober, H. V. Poor

This paper exploits the dynamic features of wireless propagation environments as the basis for a new multiple access technique, termed environment division multiple access (EDMA). In particular, with the proposed pinching-antenna-assisted EDMA, the multi-user propagation environment is intelligently reconfigured to improve signal strength at intended receive

Ozan Kanbertay, Richard Vogg, Elif Karakoc, Peter M. Kappeler

Computer vision methods are increasingly used for the automated analysis of large volumes of video data collected through camera traps, drones, or direct observations of animals in the wild. While recent advances have focused primarily on detecting individual actions, much less work has addressed the detection and annotation of interactions -- a crucial aspe

Ryan Stees

If a rational homology 3-sphere $M$ bounds a rational homology 4-ball $W$, then the kernel of the inclusion-induced homomorphism $H_1(M;\mathbb{Z})\to H_1(W;\mathbb{Z})$ is a Lagrangian for the $\mathbb{Q}/\mathbb{Z}$-valued torsion linking form $\lambda_2$ on $H_1(M;\mathbb{Z})$. In this short paper, we prove that the Freedman-Krushkal triple torsion linkin

Pawel Gajer, Jacques Ravel

We present a geometric framework for regression on structured high-dimensional data that shifts the analysis from the ambient space to a geometric object capturing the data's intrinsic structure. The method addresses a fundamental challenge in analyzing datasets with high ambient dimension but low intrinsic dimension, such as microbiome compositions, where t

Sergey K. Aityan, Abdolreza Mosaddegh, Rolando Herrero, Haitham Tayyar

Medical decision-support and advising systems are critical for emergency physicians to quickly and accurately assess patients' conditions and make diagnosis. Artificial Intelligence (AI) has emerged as a transformative force in healthcare in recent years and Large Language Models (LLMs) have been employed in various fields of medical decision-support systems

Nikolaos Lykousas, Constantinos Patsakis

Child sexual abuse is among the most hideous crimes, yet, after the COVID-19 pandemic, there is a huge surge in the distribution of child sexual abuse material (CSAM). Traditionally, the exchange of such material is performed on the dark web, as it provides many privacy guarantees that facilitate illicit trades. However, the introduction of end-to-end encryp

Iwo Słodczyk, Alexander Frisch, Xufei Fang, Inna Gitman

The growing number of ceramics exhibiting bulk plasticity at room temperature has renewed interest in revisiting plastic deformation and dislocation-mediated mechanical and functional properties in these materials. In this work, a data-driven approach is employed to identify the key parameters governing room-temperature bulk plasticity in ceramics. The model

Jozef Jirásek, Galina Jirásková

We describe witness languages meeting the upper bound on the state complexity of the multiple concatenation of $k$ regular languages over an alphabet of size $k+1$ with a significantly simpler proof than that in the literature. We also consider the case where some languages may be recognized by two-state automata. Then we show that one symbol can be saved, a

Kaushil Patel

To choose between two discrete goods, a consumer pays attention to only those with prices below a threshold. From these, she chooses her most preferred good. We assume consumers in a population have the same preference but may have different thresholds. Similar models of bounded rationality have been studied in the empirical marketing literature. We fully ch

Sneha Banerjee, Shital Lawande, Subhadeep Rana, Kuldeep Saha

We prove that if a closed manifold $B$ is a connected component of the binding of an open book decomposition of a manifold $M$, then every open book decomposition of $B$ spun embeds in $M$. As an application, we prove that every open book decomposition of a simply connected spin $5$-manifold spun embeds in $S^7$ and every $3$-dimensional open book spun embed

Jacob F. Steiner, Larissa Melischek, Felix von Oppen

The dynamics of the superconducting phase difference across a Josephson junction can be described within the resistively and capacitively shunted Josephson junction (RCSJ) model. Microscopic derivations of this model traditionally rely on the tunneling limit. Here, we present a derivation of a generalized version of the RCSJ model, which accounts for dissipa

Egor Gagushin, Marios Mertzanidis, Alexandros Psomas

We study the fundamental problem of fairly allocating a multiset $\mathcal{M}$ of $t$ types of indivisible items among $d$ groups of agents, where all agents within a group have identical additive valuations. Gorantla et al. [GMV23] showed that for every such instance, there exists a finite number $\mu$ such that, if each item type appears at least $\mu$ tim

Huanxiao Wang

This study explores whether large language models (LLMs) can simulate valid student responses for educational measurement. Using GPT -4o, 2000 virtual student personas were generated. Each persona completed the Academic Motivation Scale (AMS). Factor analyses(EFA and CFA) and clustering showed GPT -4o reproduced the AMS structure and distinct motivational su

François Belias, Naser Ezzati-Jivan, Foutse Khomh

Adaptive batch size methods aim to accelerate neural network training, but existing approaches apply identical adaptation strategies across all architectures, assuming a one-size-fits-all solution. We introduce DEBA (Dynamic Efficient Batch Adaptation), an adaptive batch scheduler that monitors gradient variance, gradient norm variation and loss variation to

Bo Zhao, Berkcan Kapusuzoglu, Kartik Balasubramaniam, Sambit Sahu

Reasoning language models perform well on complex tasks but are costly to deploy due to their size and long reasoning traces. We propose a routing approach that assigns each problem to the smallest model likely to solve it, reducing compute without sacrificing accuracy. Using intermediate representations from s1.1-32B, we train lightweight predictors of prob

Shivogo John

Evolving borrower behaviors, shifting economic conditions, and changing regulatory landscapes continuously reshape the data distributions underlying modern credit-scoring systems. Conventional explainability techniques, such as SHAP, assume static data and fixed background distributions, making their explanations unstable and potentially unfair when concept

Linghui Zeng, Ruixuan Liu, Atiquer Rahman Sarkar, Xiaoqian Jiang

Ensuring the privacy of sensitive training data is crucial in privacy-preserving machine learning. However, in practical scenarios, privacy protection may be required for only a subset of features. For instance, in ICU data, demographic attributes like age and gender pose higher privacy risks due to their re-identification potential, whereas raw lab results

P. A. Leonchik, G. S. Mauleshova, A. E. Mironov

In this paper we study two-dimensional discrete operators whose eigenfunctions at zero energy level are given by rational functions on spectral curves. We extend discrete operators to difference operators and show that two-dimensional finite-gap Schrodinger operators at fixed energy level can be obtained from difference operators by passage to the limit.

Mikhail Basok

In his seminal paper published in 2000 Kenyon developed a method to study the height function of the planar dimer model via discrete complex analysis tools. The core of this method is a set of identities representing height correlations through the inverse Kasteleyn operator. In a general setup, such as considered in [Chelkak, Laslier, Russkikh, 23, 22], sca

Stanislav Tsoi, Nicholas Proscia, Marc Christophersen, Joseph Christodoulides

Bound states in the continuum (BIC) are localized waves in electronic, photonic and acoustic systems, which remain decoupled from surrounding propagating waves and hence maintain their oscillation for extraordinary long time [Nat Rev Mater 1, 16048 (2016)]. In photonic crystals, symmetry-protected quasi-BICs (SP-qBIC) have been realized at high symmetry poin

Kenneth J. Duncan

Photometric redshifts (photo-$z$s) are an essential tool for galaxy evolution science with JWST. However, for deep surveys with more limited filter sets (i.e. $N_{\text{filt}} \sim6$) such as large pure parallel surveys, the most commonly used template-fitting based photo-$z$ approaches can yield highly confident but spurious results for high-$z$ populations