November 2025 arXiv papers — page 99

Nathan R. T. Lesnevich

A spline is an assignment of polynomials to the vertices of a graph, where the difference of two polynomials along an edge must belong to the ideal labeling that edge. We consider a ring of splines $\mathcal{M}_{H}$ constructed on a graph whose vertices are the Weyl group $\mathfrak{W}_n$ of signed permutations, and whose edges and edge-ideals are defined us

Xiaoyu Zhang, Mingtao Hu, Sen Lu, Soohyeon Kim

State space models (SSMs) have recently emerged as a powerful framework for long sequence processing, outperforming traditional methods on diverse benchmarks. Fundamentally, SSMs can generalize both recurrent and convolutional networks and have been shown to even capture key functions of biological systems. Here we report an approach to implement SSMs in ene

Vasiliki Tassopoulou, Charis Stamouli, Haochang Shou, George J. Pappas

Despite recent progress in predicting biomarker trajectories from real clinical data, uncertainty in the predictions poses high-stakes risks (e.g., misdiagnosis) that limit their clinical deployment. To enable safe and reliable use of such predictions in healthcare, we introduce a conformal method for uncertainty-calibrated prediction of biomarker trajectori

Parul Khanna, Sameer Prabhu, Ramin Karim, Phillip Tretten

The construction industry is presently going through a transformation led by adopting digital technologies that leverage Artificial Intelligence (AI). These industrial AI solutions assist in various phases of the construction process, including planning, design, production and management. In particular, the production phase offers unique potential for the in

Chalamalasetti Kranti

Following road safety norms is non-negotiable not only for humans but also for the AI systems that govern autonomous vehicles. In this work, we evaluate how well multi-modal large language models (LLMs) understand road safety concepts, specifically through schematic and illustrative representations. We curate a pilot dataset of images depicting traffic signs

Mita D. Ramabulana

For a locally compact group $G$ and compact subgroup $K$, we consider a Delsarte-type extremal problem for $G$-invariant positive definite kernels on the homogeneous space $G/K$, generalising a certain Tur\'an problem for isotropic positive definite kernels on the unit sphere $\mathbb{S}^d$ in $\mathbb{R}^{d+1}$. We exploit a correspondence between $G$-invar

Sriom Chakrabarti, Chuangtao Ma, Arijit Khan, Sebastian Link

SQL-to-Text generation aims at translating structured SQL queries into natural language descriptions, thereby facilitating comprehension of complex database operations for non-technical users. Although large language models (LLMs) have recently demonstrated promising results, current methods often fail to maintain the exact semantics of SQL queries, particul

Bharadwaj Yadavalli

Contemporary large language model deployments typically employ uniform prompting strategies across diverse query types, applying verbose response patterns to both complex analytical tasks and straightforward factual questions. This one-size-fits-all methodology leads to substantial token inefficiency, a concern amplified by the significant cost differential

Amin Heyrani Nobari, Faez Ahmed

Projected Gradient Descent (PGD) methods offer a simple and scalable approach to topology optimization (TO), yet they often struggle with nonlinear and multi-constraint problems due to the complexity of active-set detection. This paper introduces PGD-TO, a framework that reformulates the projection step into a regularized convex quadratic problem, eliminatin

Takayuki Koike

We study the complex-analytic geometry of semi-positive holomorphic line bundles on compact K\"ahler manifolds. In one of our main results, for a $\mathbb{Q}$-effective line bundle satisfying a natural torsion-type assumption, we show the equivalence between semi-positivity and semi-ampleness. More generally, for an effective nef divisor of numerical dimensi

Sara Jensen, Mark L. Lewis

Let $G$ be a group, let $d$ be a character degree, and let $e$ be the integer so that $|G| = d(d+e)$. It has been shown when $e > 1$ that $|G| \le e^4 - e^3$. In this paper, we consider the groups where $|G| = e^4 - e^3$. It is known that $e$ must be a power of a prime. We classify the groups where $e$ is a prime and where $e$ is $4$, $9$, and $25$. In so do

Tomas Ascoli, Dhruba Pariyar Damay, Jing Li, Angela Peace

Ecological processes depend on the flow and balance of essential elements such as carbon (C) and phosphorus (P), and changes in these elements can cause adverse effects to ecosystems. The theory of Ecological Stoichiometry offers a conceptual framework to investigate the impact of elemental imbalances on structured populations while simultaneously considerin

Mihir Gupte, Eshan Dixit, Muhammad Tayyab, Arun Adiththan

The diminishing ability of large language models (LLMs) to effectively utilize long-range context-the "lost-in-the-middle" phenomenon-poses a significant challenge in retrieval-based LLM applications. To study the impact of this phenomenon in a real-world application setting, we introduce GM-Extract, a novel benchmark dataset meticulously designed to evaluat

Mert Onur Cakiroglu, Idil Bilge Altun, Zhihe Lu, Mehmet Dalkilic

Temporal realism remains a central weakness of current generative video models, as most evaluation metrics prioritize spatial appearance and offer limited sensitivity to motion. We introduce a scalable, model-agnostic framework that assesses temporal behavior using motion vectors (MVs) extracted directly from compressed video streams. Codec-generated MVs fro

Juan Carlos Oyola Ballesteros, Paulo M. Carvalho-Neto

In this work, we introduce a novel variational framework for the study of the unsteady Stokes equations in a bounded open Lipschitz domain in R^n, involving a Caputo fractional derivative in time. The nonlocal nature of the fractional derivative presents significant analytical challenges, making classical methods such as the Faedo-Galerkin approach inadequat

Angelos Alexopoulos, Nikolaos Demiris

This paper develops a Bayesian framework for robust causal inference from longitudinal observational data. Many contemporary methods rely on structural assumptions, such as factor models, to adjust for unobserved confounding, but they can lead to biased causal estimands when mis-specified. We focus on directly estimating time--unit--specific causal effects a

Edward Rothberg

Recent enhancements to the Primal-Dual Hybrid Gradient (PDHG) algorithm have enabled GPUs to efficiently solve large linear programming problems, often faster than the long-dominant simplex and interior-point methods. The solutions found by PDHG are typically of much lower quality than those found by the alternatives, which can be remedied by following the P

Kai Chen, Chen Gong, Tianhao Wang

In differentially private (DP) tabular data synthesis, the consensus is that statistical models are better than neural network (NN)-based methods. However, we argue that this conclusion is incomplete and overlooks the challenge of densely correlated datasets, where intricate dependencies can overwhelm statistical models. In such complex scenarios, neural net

Badhan Chandra Das, Md Tasnim Jawad, Md Jueal Mia, M. Hadi Amini

Large Vision Language Models (LVLMs) demonstrate strong capabilities in multimodal reasoning and many real-world applications, such as visual question answering. However, LVLMs are highly vulnerable to jailbreaking attacks. This paper systematically analyzes the vulnerabilities of LVLMs integrated in Intelligent Transportation Systems (ITS) under carefully c

Seyed Mohamad Ali Tousi, Ramy Farag, John A. Lory, G. N. DeSouza

Among soil erosion problems, Ephemeral Gullies are one of the most concerning phenomena occurring in agricultural fields. Their short temporal cycles increase the difficulty in automatically detecting them using classical computer vision approaches and remote sensing. Also, due to scarcity of and the difficulty in producing accurate labeled data, automatic d

Nick Waddoups, Jonah Boe, Arnd Hartmanns, Prabal Basu

Quantitative verification can provide deep insights into reliable Network-On-Chip (NoC) designs. It is critical to understanding and mitigating operational issues caused by power supply noise (PSN) early in the design process: fluctuations in network traffic in modern NoC designs cause dramatic variations in power delivery across the network, leading to unre

Abdul Rehman, Iqra Rasool, Ayisha Imran, Mohsen Ali

Digital hematopathology requires cell-level analysis across diverse disease categories, including malignant disorders (e.g., leukemia), infectious conditions (e.g., malaria), and non-malignant red blood cell disorders (e.g., sickle cell disease). Whether single-task, vision-language, WSI-optimized, or single-cell hematology models, these approaches share a k

Nicolas M. Cuadrado A., Mohannad Takrouri, Jiří Němeček, Martin Takáč

Investment planning in power utilities, such as generation and transmission expansion, requires decisions under substantial uncertainty over decade--long horizons for policies, demand, renewable availability, and outages, while maintaining reliability and computational tractability. Conventional approaches approximate uncertainty using finite scenario sets (

Joseph Tavacoli, Andris P. Stikuts, Mihir Dass, Tim Liedl

We combine colloidal self-assembly and soft-lithography techniques to realize flexible magnetic microcrosses that can be manipulated via external, time dependent magnetic fields. The crosses are characterized by a central domain connected via four flexible arms. When subjected to an in-plane, rotating magnetic field, the crosses transit from a synchronous to

The KM3NeT Collaboration

The KM3NeT collaboration has reported the detection of the highest energy neutrino event observed to date. The energy of the event is of the order of 220 PeV hinting towards a neutrino flux at the highest energies. In this article, the potential blazar origin for this event is explored. The publicly available Astro-Multimessenger Modeling software is used to

Xu Zhou, Ivor Prado, AIFPDS participants, Ilias Tagkopoulos

Artificial intelligence is accelerating a new era of food innovation, connecting data from farm to consumer to improve formulation, processing, and health outcomes. Recent advances in deep learning, natural language processing, and multi-omics integration make it possible to understand and optimize food systems with unprecedented depth. However, AI adoption

Xingxian Liu, Dongshuai Li, Jiahui Wan, Tao Wen

Dense retrieval, as the core component of e-commerce search engines, maps user queries and items into a unified semantic space through pre-trained embedding models to enable large-scale real-time semantic retrieval. Despite the rapid advancement of LLMs gradually replacing traditional BERT architectures for embedding, their training paradigms still adhere to

Govardhan Padmanabhan

The paper presents two approaches submitted to the WMT 2025 Automated Translation Quality Evaluation Systems Task 3 - Quality Estimation (QE)-informed Segment-level Error Correction. While jointly training QE systems with Automatic Post-Editing (APE) has shown improved performance for both tasks, APE systems are still known to overcorrect the output of Machi

Yuetan Chu, Zhongyi Han, Gongning Luo, Xin Gao

Understanding how segmentation performance scales with training data is fundamental for developing data-efficient medical AI systems. In this study, we systematically revisit data scaling behavior across 15 anatomical segmentation tasks spanning four imaging modalities. We observe that medical segmentation follows a structurally stable power-law-like relatio

A. F. Kemper, Antonios Alvertis, Muhammad Asaduzzaman, Bojko N. Bakalov

Quantum computing has traditionally centered around the discrete variable paradigm. A new direction is the inclusion of continuous variable modes and the consideration of a hybrid continuous-discrete approach to quantum computing. In this paper, we discuss some of the advantages of this modality, and lay out a number of potential applications that can make u

Xin Hu, Taotao Jing, Renran Tian, Zhengming Ding

Recent advancements in autonomous driving (AD) have explored the use of vision-language models (VLMs) within visual question answering (VQA) frameworks for direct driving decision-making. However, these approaches often depend on handcrafted prompts and suffer from inconsistent performance, limiting their robustness and generalization in real-world scenarios

Saleh Momeni, Changnan Xiao, Bing Liu

This paper studies the problem of class-incremental learning (CIL), a core setting within continual learning where a model learns a sequence of tasks, each containing a distinct set of classes. Traditional CIL methods, which do not leverage pre-trained models (PTMs), suffer from catastrophic forgetting (CF) due to the need to incrementally learn both feature

Hannah Kravitz

The discrepancy sum $D_N(x,\rho)$ for irrational rotations has been of interest to mathematicians for over a century. While historically studied in an ``almost-everywhere'' or asymptotic sense, $D_N$ for finite N is increasingly an object of interest for its nontrivial properties that depend on the Diophantine properties of $\rho$. This behavior is periodic

Christopher Mantzaris, Ajda Fošner

The Hongkong and Shanghai Banking Co (HSBC) just survived a civil war intermitted by World War II. By the 1950s, it obviously needed to close all its branches in Mao's People's Republic of China, yet could somehow hold its Shanghai branch, which continued likely in the shadows, as non-state banking was illegalised and even simple land owners were executed me

Pandiyaraju V, Abishek Karthik, Jaspin K, Kannan A

This paper proposes a new enhanced model architecture to perform classification of lumbar spine degeneration with DICOM images while using a hybrid approach, integrating EfficientNet and VGG19 together with custom-designed components. The proposed model is differentiated from traditional transfer learning methods as it incorporates a Pseudo-Newton Boosting l

Xiaoyang Wei, Camille Kurtz, Florence Cloppet

Contrastive Language-Image Pretraining (CLIP) has demonstrated strong generalization for vision-language tasks in computer vision and medical domains, yet its text encoder accepts only up to 77 tokens, which limits its ability to represent long and information-rich radiology reports. Recent adaptations using domain-specific encoders, such as PubMedBERT or Cl

Mohamed Chaouch, Thanasis Stengos

This paper investigates how market competition influences poverty dynamics using a functional econometric framework that captures both contemporaneous and lagged effects. Using annual data for 48 countries from 1991-2017, we estimate function-on-function regressions linking poverty headcount ratios to market concentration and other macroeconomic indicators.

Seyedemad Motaghian, Arash Yavari

In this paper we study universal deformations in anisotropic Cauchy elasticity. We show that the universality constraints of hyperelasticity and Cauchy elasticity for transversely isotropic, orthotropic, and monoclinic solids are equivalent. This implies that for each of these symmetry classes the universal deformations and the corresponding universal materi

Hang Thanh Nguyen, Bart Van Der Holst, Phuong Hong Nguyen, Koen Kok

Maintaining system balance becomes increasingly challenging as market design and grid capacity enhancement lag behind the growing share of renewables, requiring greater effort from both the transmission system operator (TSO) and the Balance Responsible Parties (BRPs). An actor can support balancing actively by bidding into reserve markets, or passively by ad

Bukunmi G. Odunlami, Marcos Netto

This paper develops a hybrid system modeling framework for inverters that switch between grid-following and grid-forming control schemes. In particular, such inverters are modeled as hybrid automata with guard conditions on voltage and frequency, and reset maps that maintain consistent phase, frequency, and droop references during mode transitions. The hybri

Michael Rose, David A. Mazziotti

We introduce the correlation-efficient time-evolution (CETE) algorithm for simulating quantum many-body dynamics. CETE recasts each step of time evolution as a time-independent correlation problem: the ansatz begins from a mean-field single Slater determinant and is then correlated to capture the true time-evolved state. We derive this exact ansatz from a co

Zaid Hadach, Hajar El Hammouti, El Houcine Bergou, Adnane Saoud

While classical control theory assumes that the controller has access to measurements of the entire state (or output) at every time instant, this paper investigates a setting where the feedback controller can only access a randomly selected subset of the state vector at each time step. Due to the random sparsification that selects only a subset of the state

Xueyang Li, Zongren Wang, Yuliang Zhang, Zixuan Pan

Bladder cancer is one of the most prevalent malignancies worldwide, with a recurrence rate of up to 78%, necessitating accurate post-operative monitoring for effective patient management. Multi-sequence contrast-enhanced MRI is commonly used for recurrence detection; however, interpreting these scans remains challenging, even for experienced radiologists, du

Mikael von Strauss

Under real-analytic assumptions on decoder-only Transformers, recent work shows that the map from discrete prompts to last-token hidden states is generically injective on finite prompt sets. We refine this picture: for each layer $\ell$ we define a collision discriminant $\Delta^\ell \subset \Theta$ and injective stratum $U^\ell = \Theta \setminus \Delta^\el

Pedro J. Miana

In this paper we are interested to connect Koopman semigroups in Lebesgue funcion spaces $L^p(\R^+)$ and $C_0$-semigroups in Lebesgue sequence spaces $\ell^p$ for $1\le p < \infty$. To get this we use certain Poisson transformation ${\P}: L^p(\mathbb{R}^+)\to \ell^p$ and its adjoint ${\P}^*$ which allows carry semigroup properties from one space to the other

Alessio Di Santo, Gabriella Lanziani

This study examines the quantitative relationship between linguistic regularities and computational search complexity through a hybrid classical-quantum framework applied to Renaissance Italian texts. Using four representative works from the fifteenth and sixteenth centuries-Il Principe (Machiavelli), Il Cortegiano (Castiglione), I Ricordi (Guicciardini), an

James Bono, Beibei Cheng, Joaquin Lozano

AI agents are increasingly deployed to automate complex enterprise workflows, yet evidence of their effectiveness in identity governance is limited. We report results from the first randomized controlled trial (RCT) evaluating an AI agent for Conditional Access (CA) policy management in Microsoft Entra. The agent assists with four high-value tasks: policy me

Jongyeon Yoon, Elyssa M. McMaster, Michael E. Kim, Gaurav Rudravaram

Diffusion MRI (dMRI) provides a distinctive means to probe the microstructural architecture of living tissue, facilitating applications such as brain connectivity analysis, modeling across multiple conditions, and the estimation of macrostructural features. Tractography, which emerged in the final years of the 20th century and accelerated in the early 21st c

Siyuan Li, Kai Yu, Anna Wang, Zicheng Liu

Modeling genomic sequences faces two unsolved challenges: the information density varies widely across different regions, while there is no clearly defined minimum vocabulary unit. Relying on either four primitive bases or independently designed DNA tokenizers, existing approaches with naive masked language modeling pre-training often fail to adapt to the va

Kranti Kumar Parida, Omar Emara, Hazel Doughty, Dima Damen

Humans excel at multisensory perception and can often recognise object properties from the sound of their interactions. Inspired by this, we propose the novel task of Collision Sound Source Segmentation (CS3), where we aim to segment the objects responsible for a collision sound in visual input (i.e. video frames from the collision clip), conditioned on the

Ryan J. French, William H. Ashfield, Cole A. Tamburri, Maria D. Kazachenko

We detect rapid downflows of 150-217 km/s in IRIS Si IV 1402.77 nm measurements of an X9-class solar flare on 2024 October 3rd. The fast redshift values persist for over 15 minutes from flare onset, and can be split into two distinct stages of behavior, suggesting that multiple mechanisms are responsible for the downwards acceleration of flare ribbon plasma.

Linhan Fang, Jesus Silva-Rodriguez, Xingpeng Li

Widespread electric vehicle (EV) adoption introduces new challenges for distribution grids due to large, localized load increases, stochastic charging behavior, and limited data availability. This paper proposes two data-driven methods to estimate residential EV charging profiles using real-world customer meter data from CenterPoint Energy serving the Housto

James Bono

Security operations centers (SOCs) face a persistent challenge: efficiently triaging a high volume of user-reported phishing emails while maintaining robust protection against threats. This paper presents the first randomized controlled trial (RCT) evaluating the impact of a domain-specific AI agent - the Microsoft Security Copilot Phishing Triage Agent - on

Mahan FakouriFard, Mingxi Liu

Distributed multi-agent optimization (DMAO) enables the scalable control and coordination of a large population of edge resources in complex multi-agent environments. Despite its great scalability, DMAO is prone to cyber attacks as it relies on frequent peer-to-peer communications that are vulnerable to malicious data injection and alteration. Existing cyber

Jaden Lopez, Stefano Profumo

A massive ``dark companion'' -- such as an intermediate-mass black hole or other compact dark object -- orbiting the supermassive black hole at the Galactic Center can dynamically reshape the surrounding dark-matter spike. Through gravitational heating and angular-momentum exchange, the companion excavates a ``scoured'' region that lowers the inner density a

Muhammed Can Keles, Bedrettin Cetinkaya, Sinan Kalkan, Emre Akbas

While heatmap-based human pose estimation methods have shown strong performance, they suffer from three main problems: (P1) "Commonly used Mean Squared Error (MSE)" Loss may not always improve joint localization because it penalizes all pixel deviations equally, without focusing explicitly on sharpening and correctly localizing the peak corresponding to the

Jonathon Dilworth, Hui Yang, Jiaoyan Chen, Yongsheng Gao

SNOMED CT is a biomedical ontology with a hierarchical representation, modelling terminological concepts at a large scale. Knowledge retrieval in SNOMED CT is critical for its application but often proves challenging due to linguistic ambiguity, synonymy, polysemy, and so on. This problem is exacerbated when the queries are out-of-vocabulary (OOV), i.e., lac

Banshi Lal, Yogesh Maan, Moaz Abdelmaguid, Visweshwar Ram Marthi

Magnetars are the leading candidate sources of fast radio bursts (FRBs). However, the observational probes of the connections between magnetars and FRBs are severely limited by the paucity of detection of highly energetic radio events from magnetars -- to date, only one radio burst as energetic as FRBs has been detected from a Galactic magnetar. Here, we pre

William A. Simon, Peter J. Love

Quantum simulation is a promising application for quantum computing. Quantum simulation algorithms may require the ability to control the time evolution unitary. Naive techniques to control a unitary can substantially increase the required computational resources. A standard approach to controlling Trotterized time evolution doubles the number of single-qubi

Yassir Benhammou, Suman Kalyan, Sujay Kumar

Broadcast and media organizations increasingly rely on artificial intelligence to automate the labor-intensive processes of content indexing, tagging, and metadata generation. However, existing AI systems typically operate on a single modality-such as video, audio, or text-limiting their understanding of complex, cross-modal relationships in broadcast materi

Xinxin Liu, Zhaopan Xu, Ming Li, Kai Wang

While Chain-of-Thought (CoT) prompting enables sophisticated symbolic reasoning in LLMs, it remains confined to discrete text and cannot simulate the continuous, physics-governed dynamics of the real world. Recent video generation models have emerged as potential world simulators through Chain-of-Frames (CoF) reasoning -- materializing thought as frame-by-fr

Anna Rodum Bjøru, Rafael Cabañas, Helge Langseth, Antonio Salmerón

Recently, Bj{\o}ru et al. proposed a novel divide-and-conquer algorithm for bounding counterfactual probabilities in structural causal models (SCMs). They assumed that the SCMs were learned from purely observational data, leading to an imprecise characterization of the marginal distributions of exogenous variables. Their method leveraged the canonical repres

Thomas Perrin

We study long-time dynamics of the damped focusing cubic Klein-Gordon equation on a compact three-dimensional Riemannian manifold, together with its space-independent reduction, the damped focusing Duffing equation. Under the geometric control condition on the damping and an assumption on the set of stationary solutions, we establish a sharp trichotomy for i

István Juhász, Jan van Mill, Lajos Soukup

We show that there are locally compact spaces that can be condensed onto separable spaces but not onto compact separable spaces. We also show that for every cardinal $\kappa$ there is a locally compact topological group of cardinality $2^\kappa$ that can be condensed onto a compact space but not onto a compact topological group. These answer some questions o

Matteo Spreafico, Ludovica Tassini, Camilla Sancricca, Cinzia Cappiello

Large language models have recently demonstrated their exceptional capabilities in supporting and automating various tasks. Among the tasks worth exploring for testing large language model capabilities, we considered data preparation, a critical yet often labor-intensive step in data-driven processes. This paper investigates whether large language models can

Duncan Farrah, Kiana Ejercito, Andreas Efstathiou, David Leisawitz

Infrared-luminous galaxies are important sites of stellar and black hole mass assembly at most redshifts. Their luminosities are often estimated by fitting spectral energy distribution (SED) models to near- to far-infrared data, but the dependence of these estimates on the data used is not well-understood. Here, using observations simulated from a well-studi

Matteo Della Rocca, Thomas F. M. Spieksma, Francisco Duque, Leonardo Gualtieri

Black holes in our Universe are rarely truly isolated, being instead embedded in astrophysical environments such as plasma or dark matter. A particularly intriguing possibility is that light scalar fields form bound states around black holes, producing extended ''clouds'' known as gravitational atoms. When these clouds become sufficiently compact, the spacet

William A Wheeler, Samuel Chevalier, Amritanshu Pandey

Electrification of residential heating and transporta- tion has the potential to overload transformers in distribution feeders. Strategic scheduling of transformer upgrades to antici- pate increasing loads can avoid operational failures and reduce the risk of supply shortages. This work proposes a framework to prioritize transformer upgrades based on predict

Xin Guan, Bernhard Mistlberger, Michael Ruf

We compute three-loop corrections to the three-point form factor of the operator $\operatorname{tr}(\phi^2)$ in $\mathcal{N} = 4$ Super Yang-Mills theory. In particular, our result is valid beyond the leading-color limit and will consequently be an important input towards extending the amplitude-bootstrap program beyond the leading-color approximation. We fi

Bence Bakó, Zoltán Kolarovszki, Zoltán Zimborás

Quantum generative learning is a promising application of quantum computers, but faces several trainability challenges, including the difficulty in experimental gradient estimations. For certain structured quantum generative models, however, expectation values of local observables can be efficiently computed on a classical computer, enabling fully classical

Julian G. Soltes

High-dimensional numerical optimization presents a persistent challenge in computational science. This paper introduces Quasi-Adaptive Search with Asymptotic Reinitialization (QUASAR), an evolutionary algorithm to accelerate convergence in complex, non-differentiable problems afflicted by the curse of dimensionality. QUASAR expands upon the core principles o

Eryk Buk, Denis Dobkowski-Ryłko, Jerzy Lewandowski, Maciej Ossowski

We provide a construction of a new class of axisymmetric extremal isolated horizons admitting a structure of U(1)-principal fiber bundle over a two-sphere. In contrast to the previous examples, the null generators are assumed to be transversal to the bundle fibers. We impose the Einstein equations at the horizon and explicitly derive all intrinsic geometries

Zelei Shao, Vikranth Srivatsa, Sanjana Srivastava, Qingyang Wu

Reinforcement learning(RL) post-training has become essential for aligning large language models (LLMs), yet its efficiency is increasingly constrained by the rollout phase, where long trajectories are generated token by token. We identify a major bottleneck:the long-tail distribution of rollout lengths, where a small fraction of long generations dominates w

Sayed Shafaat Mahmud, Sayantan Auddy, Neal Turner, Jeffrey S. Bary

Dust-continuum observations of many protoplanetary disks reveal rings and gaps that are widely interpreted as evidence of ongoing planet formation. Here we present the first framework for inferring planet and disk parameters from such images using variational autoencoder (VAE) based generative machine learning (ML). The new framework is called VADER (Variati

A. de Oliveira Junior, Rafael A. Macedo, Jakub Czartowski, Jonatan Bohr Brask

Quantum advantage arises from quantum states that cannot be efficiently simulated on a classical computer. Such states are characterised by a property known as nonstabiliserness. In this work, we investigate whether nonstabiliserness can be generated by placing an initially stabiliser state in contact with a heat bath. Under minimal thermodynamic assumptions

Claudia Ojeda-Aristizabal, Xiaohu Zheng, Changsong Xu, Zohar Nussinov

Quantum spin liquids can arise from Kitaev magnetic interactions, and exhibit fractionalized excitations with the potential for a topological form of quantum computation. This review surveys recent experimental and theoretical progress on the pursuit of phenomena related to Kitaev magnetism in layered and exfoliatable materials, which offer numerous opportun

Carlo Heissenberg, Rodolfo Russo

Gravitational observables for binary systems exhibit a simple polynomial dependence on the masses $m_1$, $m_2$ of the two scattering objects when they are written in terms of the appropriate kinematic variables in the post-Minkowskian (PM) regime. We point out that this property, combined with particle interchange symmetry, allows one to reconstruct the lead

Nora B. Linzer, Lucia Armillotta, Eve C. Ostriker, Eliot Quataert

We use models of spectrally resolved cosmic ray (CR) transport in TIGRESS MHD simulations of the local ISM to produce synthetic synchrotron emission and to test, on scales from a few kpc down to ~10 pc, the traditional estimate of magnetic field strength based on the assumption of equipartition between the magnetic and total CR energy densities. Our analysis

Aswin Parayil Mana, Yaman Sanghavi

Defects associated with non-invertible symmetries have attracted significant attention in recent years. Among them, Kramers-Wannier (KW) duality defects have been investigated in both classical statistical systems and quantum Hamiltonian models. Aasen et al. analyzed duality defects in the 2D Ising model and in statistical models built from fusion categories

Douglas R. Alves, James S. Jenkins, José I. Vinés, Maximilano Moyano

Observations have shown that planets similar to Neptune are rarely found orbiting Sun-like stars with periods up to ~4 days, defining the so-called Neptune desert region. Therefore, the detection of each individual planet in this region holds a high value, providing detailed insights into how such a population came to form and evolve. Here we report the dete

Mahmoud Jalali Mehrabad, Alireza Parhizkar, Lida Xu, Gregory Moille

Multiscale synergy -- the interplay of a system's distinct characteristic length, time, and energy scales -- is becoming a unifying thread across many contemporary branches of science. Ranging from moir\'e and super-moir\'e materials and cold atoms to DNA-templated superlattices and nested photonic networks, multiscale synergy produces behaviors not obtainab

Savvas Constantinou, Nikku Madhusudhan, Måns Holmberg

The first explorations of temperate sub-Neptune exoplanets have been the hallmark of early JWST observations. The bulk properties of such planets are consistent with a range of possible internal structures, which can be distinguished through their interactions with the observable atmospheres. JWST observations of TOI-270 d, a temperate sub-Neptune, have prev

A. Traina, C. Gruppioni, I. Delvecchio, B. Magnelli

We investigate how the obscured IR-derived and the dust-corrected UV star formation rate functions (SFRFs) compare with each other, and with predictions from state-of-the-art theoretical models of galaxy formation and evolution. We derive the IR-SFRF from the ALMA A$^3$COSMOS survey, by converting the IR luminosity functions (IR-LFs) into SFRF after correcti

Sebastian Leontica, Alberto Baiardi, Julian Schuhmacher, Francesco Tacchino

We propose a hybrid quantum-classical algorithm for approximating the ground state of two-dimensional quantum systems using an isometric tensor network ansatz, which maps naturally to quantum circuits. Inspired by the density matrix renormalization group, we optimize tensors sequentially by diagonalizing a series of effective Hamiltonians. These are construc

Humza Nusrat, Omar Nusrat

Agentic AI "scientists" now use language models to search the literature, run analyses, and generate hypotheses. We evaluate KOSMOS, an autonomous AI scientist, on three problems in radiation biology using simple random-gene null benchmarks. Hypothesis 1: baseline DNA damage response (DDR) capacity across cell lines predicts the p53 transcriptional response

Dimitri A. Gadotti

Deconstructing galaxies through two-dimensional decompositions has been shown to be a powerful technique to derive the physical properties of stellar structures in galaxies. However, most studies employ fitting algorithms that are prone to be trapped in local minima, or involve subjective choices. Furthermore, when applied on samples beyond the nearby Univer

Albert Rico, Dmitry Grinko, Robin Krebs, Lin Htoo Zaw

We present a method to detect entanglement partitions of multipartite quantum systems, by exploiting their inherent symmetries. Structures like genuinely multipartite entanglement, $m$-separability and entanglement depth are detected as very special cases. This formulation enables us to characterize all the entanglement partitions of all three- and four- par

Basile Coron, Luis Ferroni, Shiyue Li

We introduce and develop the theory of UMEL-shellable posets. These are posets equipped with an edge-lexicographical labeling satisfying certain uniformity and monotonicity properties. This framework encompasses classical families of combinatorial geometries, including uniform matroids, projective and affine geometries, braid matroids of type A and B, and al

Jaya Doliya, Deep Jyoti Das, Subhadip Bouri, Pooja Bhattacharjee

The origin and acceleration mechanism of cosmic rays (CRs) remain fundamental open questions. Galaxy mergers are proposed as very high-energy CR accelerators, which are expected to produce high-energy (HE) $\gamma$ rays and neutrinos through interactions with the ambient gas and low-energy background radiation fields. For the first time, we systematically st

Osase Omoruyi, Grant Tremblay, Stefi A. Baum, Tracy E. Clarke

To examine how AGN feedback shapes the intracluster medium (ICM) and fuels black hole accretion in the cool-core galaxy cluster Abell 2597, we present deep ($\sim$600 ks) Chandra X-ray observations complemented by archival GMRT radio and SINFONI near-infrared data. Radio-mode AGN activity has inflated seven X-ray cavities and driven one to three potential we

Christoph A. Keller, Ashley Winter Roberts

Motivated by sparseness conditions for holographic CFTs, we investigate sparseness of modular forms, lattices, and codes. For this we investigate the free energy of such objects as their weight, dimension or size goes to infinity. We construct families of modular forms that are sparse, such as the Eisenstein series $E_{2k}(\tau)$. We then investigate lattice

Francisco R. Candón, Damiano F. G. Fiorillo, Ángel Gil Muyor, Hans-Thomas Janka

QCD axions would be copiously produced in the proto-neutron star formed in a core-collapse supernova (SN). After escaping, they would convert into gamma rays in the Galactic magnetic field and, as recently shown, in that of the progenitor star itself. Here, we show that Type Ibc SNe -- whose progenitors have lost their hydrogen or even helium envelopes -- ar

Vinicius Nevoa, Sanjay Raman, Cumrun Vafa

In this note, we conjecture and provide evidence that quantum gravity cobordism classes are trivialized by elementary generators with tension at most order 1 in Planck units. Motivated by the attractor mechanism and a number of key examples, we propose that there is always a regime in the field space of the EFT in which the intrinsic tension of such an objec

Yonatan Kahn, Michael Wentzel, Nicolás Yunes

Finite density corrections to the lighter-than-QCD axion can invert the effective axion potential, sourcing a non-trivial axion field inside dense objects. We perform the first numerical study of the complete dynamics of the lighter-than-QCD axion in a neutron star in 1+1 general relativity, extending the region of analysis to low-mass axions with kilometer-

Maria Werhahn, Christoph Pfrommer, Joseph Whittingham, Léna Jlassi

Cosmic ray (CR) electrons are key tracers of non-thermal processes in galaxies, yet their spectra are often modelled under the untested assumption of steady state between injection and cooling. In this work, we present a time-dependent modelling of CR electron spectra in a galactic context using the CREST code, applied to magnetohydrodynamical simulations of

Tianhong Li, Kaiming He

Today's denoising diffusion models do not "denoise" in the classical sense, i.e., they do not directly predict clean images. Rather, the neural networks predict noise or a noised quantity. In this paper, we suggest that predicting clean data and predicting noised quantities are fundamentally different. According to the manifold assumption, natural data shoul

Zhongang Cai, Ruisi Wang, Chenyang Gu, Fanyi Pu

Despite remarkable progress, multimodal foundation models still exhibit surprising deficiencies in spatial intelligence. In this work, we explore scaling up multimodal foundation models to cultivate spatial intelligence within the SenseNova-SI family, built upon established multimodal foundations including visual understanding models (i.e., Qwen3-VL and Inte

Philip Boyle Smith, Joe Davighi

Gauging fermion parity and summing over spin structures are subtly distinct operations. We introduce 'bosonisation cohomology' groups $H_B^{d+2}(X)$ to capture this difference, for theories in spacetime dimension $d$ equipped with maps to some $X$. Non-trivial classes in $H_B^{d+2}(X)$ contain theories for which $(-1)^F$ is anomaly-free, but spin structure s

Xunjie Wang, Jiacheng Shi, Zihan Zhao, Yang Yu

Large Language Models (LLMs) deployed on mobile devices offer benefits like user privacy and reduced network latency, but introduce a significant security risk: the leakage of proprietary models to end users. To mitigate this risk, we propose a system design for protecting on-device LLMs using Arm Trusted Execution Environment (TEE), TrustZone. Our system ad

Benjamin K. Capistrant, Jason Dittmann

M dwarfs are magnetically active stars that frequently produce flares, which have implications for both stellar evolution and exoplanet studies. Flare occurrence rates and activity levels of M dwarfs correlate with stellar characteristics such as age, mass, and rotation period. We search TESS observations of a known active population of M dwarfs as well as a