October 2025 arXiv papers — page 169

Łukasz Chomienia, Alfred Michel Grundland

The paper contains an analysis of the conditions for the existence of elastic versus non-elastic wave superpositions governed by the Euler system in (1+1)-dimensions. A review of recently obtained results is presented, including the introduction of the notion of quasi-rectifiability of vector fields and its application to both elastic and non-elastic wave su

Ayoub Bouguettaya, Elizabeth M. Stuart

The lexical hypothesis posits that personality traits are encoded in language and is foundational to models like the Big Five. We created a bottom-up personality model from a classic adjective list using machine learning and compared its descriptive utility against the Big Five by analyzing one million Reddit comments. The Big Five, particularly Agreeablenes

Anna Schroeder, Lucas B. Vieira, Jan Nöller, Nikolai Miklin

There is growing interest in developing rigorous tests of quantumness that are feasible even before practical quantum advantages become a reality. Such tests not only aim to certify the quantum nature of a system but also serve as benchmarks for precise quantum control. In this work, we argue that promise problems, studied in the theory of finite automata, p

Daria de tinguy, Tim Verbelen, Emilio Gamba, Bart Dhoedt

Autonomous navigation in unfamiliar environments requires robots to simultaneously explore, localise, and plan under uncertainty, without relying on predefined maps or extensive training. We present Active Inference MAPping and Planning (AIMAPP), a framework unifying mapping, localisation, and decision-making within a single generative model, drawing on cogn

Steve Han, Gilberto Titericz Junior, Tom Balough, Wenfei Zhou

This research introduces the Judge's Verdict Benchmark, a novel two-step methodology to evaluate Large Language Models (LLMs) as judges for response accuracy evaluation tasks. We assess how well 54 LLMs can replicate human judgment when scoring responses from RAG (Retrieval-Augmented Generation) or Agentic pipelines against ground truth answers. Our methodol

Siphiwe Shandu, Thabiso Moropa, Alain R. Ndjiongue

This letter investigates contactless power line communications (CPLC) for underground mining by modeling power wires as long-wire antennas. A system-level framework is developed, comprising a cascade of RF and power line channels. The model accounts for multipath propagation, frequency-dependent attenuation, and Rician fading. Simulations from 1-20 GHz revea

Gustavo P. de Brito, Manuel Reichert, Marc Schiffer

We present the first complete next-to-leading-order analysis of a Yukawa system within the framework of asymptotically safe quantum gravity. Our results are obtained through a systematic resummation of higher-order operators, revealing two distinct resummation mechanisms -- one of which has not been explored previously. In addition, we introduce a novel appr

Khandakar Md Asif Elahi, Samir Choudhuri, Nirupam Roy, Md Rashid

Radio-interferometric arrays require very precise calibration to detect the Epoch of Reionization 21-cm signal. A remarkably complete and accurate sky model is therefore needed in the patches of the sky used to perform the calibration. Instruments such as HERA, which use a redundant calibration strategy, also require a reference sky model to fix degenerate g

Debarun Bhattacharjya, Balaji Ganesan, Junkyu Lee, Radu Marinescu

When does a large language model (LLM) know what it does not know? Uncertainty quantification (UQ) provides measures of uncertainty, such as an estimate of the confidence in an LLM's generated output, and is therefore increasingly recognized as a crucial component of trusted AI systems. Black-box UQ methods do not require access to internal model information

Nishant Gautam, Somya Sharma, Peter Corcoran, Kaspar Althoefer

Pseudo-haptics exploit carefully crafted visual or auditory cues to trick the brain into "feeling" forces that are never physically applied, offering a low-cost alternative to traditional haptic hardware. Here, we present a comparative psychophysical study that quantifies how visual and auditory stimuli combine to evoke pseudo-haptic pressure sensations on a

F. Poidevin, S. L. West, C. M. B. Omand, R. Könyves-Tóth

Current large-scale, high-cadence surveys, such as the ZTF, provide detections of new and rare types of transients and supernovae whose physical origins are not well understood. We investigate the nature of SN 2021lwz at a redshift $z=0.065$, an overluminous supernova (SN) of absolute magnitude, $M_{g} \sim -20.1$ AB, falling in the lower range of superlumin

Yongda Zhu, George D. Becker, Anson D'Aloisio, Ryan Endsley

How galaxies drive reionization and what governs its geometry remain fundamental questions. We present JWST/NIRCam wide-field slitless spectroscopy (WFSS) observations toward two of the most Ly$\alpha$-transmissive QSO sightlines near the end of reionization. We find that regions at $z \sim 5.7$ along both sightlines previously found to be low-density in Ly$

Jacopo Tagliabue, Ciro Greco

Data lakehouses run sensitive workloads, where AI-driven automation raises concerns about trust, correctness, and governance. We argue that API-first, programmable lakehouses provide the right abstractions for safe-by-design, agentic workflows. Using Bauplan as a case study, we show how data branching and declarative environments extend naturally to agents,

Ilia Revin, Leon Strelkov, Vadim A. Potemkin, Ivan Kireev

There are many critical challenges in optimizing neural network models, including distributed computing, compression techniques, and efficient training, regardless of their application to specific tasks. Solving such problems is crucial because the need for scalable and resource-efficient models is increasing. To address these challenges, we have developed a

Rajath Radhakrishnan, Adar Sharon, Nathanan Tantivasadakarn

Dynamical stabilizer codes (DSCs) have recently emerged as a powerful generalization of static stabilizer codes for quantum error correction, replacing a fixed stabilizer group with a sequence of non-commuting measurements. This dynamical structure unlocks new possibilities for fault tolerance but also introduces new challenges, as errors must now be tracked

Jiajie Zhao, Tao Luo, Yaoyu Zhang

While architecture is recognized as key to the performance of deep neural networks, its precise effect on training dynamics has been unclear due to the confounding influence of data and loss functions. This paper proposed an analytic framework based on the geometric control theory to characterize the dynamical properties intrinsic to a model's parameterizati

Pok Him Cheng, Joel E. Cohen, Hok Kan Ling, Sheung Chi Phillip Yam

Taylor's law, also known as fluctuation scaling in physics and the power-law variance function in statistics, is an empirical pattern widely observed across fields including ecology, physics, finance, and epidemiology. It states that the variance of a sample scales as a power function of the mean of the sample. We study generalizations of Taylor's law in the

Minkyoung Cho, Ruben Ohana, Christian Jacobsen, Adityan Jothi

Current controllable diffusion models typically rely on fixed architectures that modify intermediate activations to inject guidance conditioned on a new modality. This approach uses a static conditioning strategy for a dynamic, multi-stage denoising process, limiting the model's ability to adapt its response as the generation evolves from coarse structure to

River Beard, Kyle W. Martin, John D. Elgin, Brian L. Kasch

The optical atomic clock based on the $5S_{1/2} \rightarrow 5D_{5/2}$ two-photon transition in rubidium is a candidate for a next generation, manufacturable, portable clock that fits in a small size, weight, and power (SWaP) envelope. Here, we report the first two-photon rubidium clock stabilized by detecting 776~nm fluorescence. We also demonstrate the use

Adrien Picquenot, Fabio Acero, Valeria Olivares, Michela Negro

Deep Chandra observations of the Perseus galaxy cluster have allowed for the discovery of X-ray counterparts to the H{\alpha} filamentary structures and of a sloshing spiral. However, both components are extremely faint, and their study is largely hindered by the volume-filling hot intracluster medium (ICM). Using the Poisson General Morphological Component

Qiguang Chen, Zheng Yan, Mingda Yang, Libo Qin

As the volume of peer-reviewed research surges, scholars increasingly rely on social platforms for discovery, while authors invest considerable effort in promoting their work to ensure visibility and citations. To streamline this process and reduce the reliance on human effort, we introduce Automatic Promotion (AutoPR), a novel task that transforms research

Tzu-Lin Kuo, Wei-Ning Chiu, Wei-Yun Ma, Pu-Jen Cheng

Document expansion (DE) via query generation tackles vocabulary mismatch in sparse retrieval, yet faces limitations: uncontrolled generation producing hallucinated or redundant queries with low diversity; poor generalization from in-domain training (e.g., MS MARCO) to out-of-domain data like BEIR; and noise from concatenation harming dense retrieval. While L

Daniel Brubaker, William Sheffield, Junyi Jessy Li, Kanishka Misra

The role of world knowledge has been particularly crucial to predict the discourse connective that marks the discourse relation between two arguments, with language models (LMs) being generally successful at this task. We flip this premise in our work, and instead study the inverse problem of understanding whether discourse connectives can inform LMs about t

Raoyuan Zhao, Yihong Liu, Hinrich Schütze, Michael A. Hedderich

Large reasoning models (LRMs) increasingly rely on step-by-step Chain-of-Thought (CoT) reasoning to improve task performance, particularly in high-resource languages such as English. While recent work has examined final-answer accuracy in multilingual settings, the thinking traces themselves, i.e., the intermediate steps that lead to the final answer, remain

Thomas C. Smits, Nikolay Akhmetov, Tiffany S. Liaw, Mark S. Keller

Summary: Cell population plots are visualizations showing cell population distributions in biological samples with single-cell data, traditionally shown with stacked bar charts. Here, we address issues with this approach, particularly its limited scalability with increasing number of cell types and samples, and present scellop, a novel interactive cell popul

Yu Wang, Tianhao Tan, Yifei Wang

Retrieving relevant instructional videos from multilingual medical archives is crucial for answering complex, multi-hop questions across language boundaries. However, existing systems either compress hour-long videos into coarse embeddings or incur prohibitive costs for fine-grained matching. We tackle the Multilingual Video Corpus Retrieval (mVCR) task in t

Cyril Demarche, Hanqing Long

In this paper, we determine the motive of the classifying torsor of an algebraic torus. As a result, we give an exact sequence describing the degree 4 cohomological invariants of algebraic tori. Using results by Blinstein and Merkurjev, this provides a formula for the degree 4 unramified cohomology group of an algebraic torus, via a flasque resolution.

Gavriel Di Nepi, Federico Siciliano, Fabrizio Silvestri

By the end of 2024, Google researchers introduced Titans: Learning at Test Time, a neural memory model achieving strong empirical results across multiple tasks. However, the lack of publicly available code and ambiguities in the original description hinder reproducibility. In this work, we present a lightweight reimplementation of Titans and conduct a compre

Behzad Mortezapour, Sebastian Hamer, Rainer Herges, Roberto Robles

We investigated trioxatriangulenium functionalized with phenyl (phenyl-TOTA) on the (111) surfaces of Ag and Au using low-temperature scanning tunneling microscopy (STM) and density functional theory (DFT). On Ag(111), the molecules form hexagonal arrays, and on Au(111), honeycomb patterns are also observed. The orientations of the phenyl moieties are resolv

Andrew Stasiuk, Garrett Heller, Lance Berkey, Bo Xing

Emergent hydrodynamics (EHD) bridges short-time unitarity with late-time thermodynamics, universal transport phenomena characterize the manner and speed of transport and thermalization. Typical non-integrable systems with few conserved local quantities are expected to be diffusive. In contrast, strongly disordered systems which admit phases such as many-body

Eric Seewald, Sanat Ghosh, Nishchhal Verma, John Cenker

Rhombohedral graphene (rG) aligned with hexagonal boron nitride (hBN) has been shown to host flat bands that stabilize various strongly correlated quantum phases, including Mott insulators, integer, and fractional quantum anomalous Hall phases. In this work, we use scanning tunneling microscopy/spectroscopy (STM/STS) to visualize the dispersion of flat bands

Athanasios Bakopoulos, Christos Charmousis, Nikos Chatzifotis, Theodoros Nakas

We investigate the structure of nontrivial maximally symmetric vacua and compact-object solutions in shift-symmetric scalar-tensor theories. Focusing on Horndeski gravity, we derive consistency conditions directly from the field equations to identify the subclasses that admit Minkowski and de Sitter vacua with a nontrivial scalar field. In doing so, we obtai

Julia Gehrlein, Joachim Kopp, Margot MacMahon, George A. Parker

Upcoming precision long-baseline neutrino oscillation experiments will be severely limited by the large systematic uncertainties associated with neutrino flux predictions and neutrino--nucleus cross sections. A promising remedy is the PRISM (Precision Reaction Independent Spectrum Measurement) technique, whereby the near detector measures the neutrino spectr

Vincent N. Novellino, Dmitriy Y. Anistratov

This paper presents multilevel hybrid transport (MLHT) methods for solving the neutral-particle Boltzmann transport equation. The proposed MLHT methods are formulated on a sequence of spatial grids using a multilevel Monte Carlo (MLMC) approach. The general MLMC algorithm is defined by recursively estimating the expected value of the correction to a solution

Qiguang Chen, Hanjing Li, Libo Qin, Dengyun Peng

Recently, Diffusion Large Language Models (DLLMs) have offered high throughput and effective sequential reasoning, making them a competitive alternative to autoregressive LLMs (ALLMs). However, parallel decoding, which enables simultaneous token updates, conflicts with the causal order often required for rigorous reasoning. We first identify this conflict as

Chenghao Wang, Arjun Viswanathan, Eric Sihite, Alireza Ramezani

Animals achieve energy-efficient locomotion by their implicit passive dynamics, a marvel that has captivated roboticists for decades.Recently, methods incorporated Adversarial Motion Prior (AMP) and Reinforcement learning (RL) shows promising progress to replicate Animals' naturalistic motion. However, such imitation learning approaches predominantly capture

M. Aa. Solberg

We apply Lie symmetry analysis of partial differential equations (PDEs) to the Euler-Lagrange equations of the two-Higgs-doublet model (2HDM), to determine its scalar Lie point symmetries. A Lie point symmetry is a structure-preserving transformation of the spacetime variables and the fields of the model, which is also continuous and connected to the identit

Stef Cuyckens, Ryan Antonio, Chao Fang, Marian Verhelst

Implantable devices for reliable intracranial electroencephalography (iEEG) require efficient, accurate, and real-time detection of seizures. Dense hyperdimensional computing (HDC) proves to be efficient over neural networks; however, it still consumes considerable switching power for an ultra-low energy application. Sparse HDC, on the other hand, has the po

Chenyu Wang, Paria Rashidinejad, DiJia Su, Song Jiang

Diffusion large language models (dLLMs) are emerging as an efficient alternative to autoregressive models due to their ability to decode multiple tokens in parallel. However, aligning dLLMs with human preferences or task-specific rewards via reinforcement learning (RL) is challenging because their intractable log-likelihood precludes the direct application o

Jacob Van Grinsven

Let $H$ be a coradically graded Hopf algebra. For every Loewy-graded exact $H$-comodule algebra $A=\oplus_{n\geq 0} A(n)$ and $H_0$-equivariant Morita equivalence $A(0)\simeq_{H_0} X$, there exists a Loewy-graded $H$-comodule algebra $B$ (isomorphic to $X$ in degree zero) realizing an $H$-equivariant Morita equivalence $A\simeq_H B$. In addition, if every ex

Patrick Ferreira, Paula Costa

We present IF-D, a large-scale inertial dataset designed to enable self-supervised and foundational learning for IMU time series. IF-D comprises continuous, long-duration multichannel recordings (accelerometer, gyroscope, magnetometer) sampled at 200Hz using a UM7 IMU mounted inside a 3D-printed spherical enclosure that promotes diverse, free rotations durin

Chhatra Bahadur Subba, Bhanu Chettri, Amel Laref, Zeesham Abbas

The development of efficient and cost-effective catalysts for clean energy conversion remains a central challenge in materials science. Although platinum serves as the benchmark catalyst, its scarcity and high cost hinder large-scale deployment. In this study, we propose a newly predicted SnSiGeN4 MXene-family monolayer as a promising candidate for the oxyge

Arthur Bizzi, Matias Grynberg, Vitor Matias, Daniel Perazzo

Morphing is a long-standing problem in vision and computer graphics, requiring a time-dependent warping for feature alignment and a blending for smooth interpolation. Recently, multilayer perceptrons (MLPs) have been explored as implicit neural representations (INRs) for modeling such deformations, due to their meshlessness and differentiability; however, ex

Raoyuan Zhao, Yihong Liu, Lena Altinger, Hinrich Schütze

Large language models (LLMs) are increasingly deployed in multilingual, real-world applications with user inputs -- naturally introducing \emph{typographical errors} (typos). Yet most benchmarks assume clean input, leaving the robustness of LLMs to typos across languages largely underexplored. To address this gap, we introduce MulTypo, a multilingual typo ge

Feifan Song, Shaohang Wei, Bofei Gao, Yejie Wang

Large reasoning models (LRMs) boosted by Reinforcement Learning from Verifier Reward (RLVR) have shown great power in problem solving, yet they often cause overthinking: excessive, meandering reasoning that inflates computational cost. Prior designs of penalization in RLVR manage to reduce token consumption while often harming model performance, which arises

Percy S. Zhai, So Won Jeong, Veronika Ročková

We propose a generative multivariate posterior sampler via flow matching. It offers a simple training objective, and does not require access to likelihood evaluation. The method learns a dynamic, block-triangular velocity field in the joint space of data and parameters, which results in a deterministic transport map from a source distribution to the desired

Fabrizio Giordano, Yago Ascasibar, Luca Cortese, Ivan Valtchanov

Context. In the current era of multi-wavelength and multi-messenger astronomy, international organisations are actively working on the definition of new standards for the publication of astronomical data, and substantial effort is devoted to make them available through public archives. Aims. We present a set of tools that allow user-friendly access and basic

Hung T. Vuong, Ankit Mahajan, John L. Weber, James Shee

In this study, we evaluate multi-configurational trial wave function protocols for phaseless auxiliary field quantum Monte Carlo (ph-AFQMC) on transition metal containing systems. First, we benchmark vertical ionization potentials for 22 3d transition metal complexes against published high-accuracy ph-AFQMC values in a double zeta basis set. We then compute

PeiHuang Zheng, Yunlong Zhao, Zheng Cui, Yang Li

Small object detection in aerial images suffers from severe information degradation during feature extraction due to limited pixel representations, where shallow spatial details fail to align effectively with semantic information, leading to frequent misses and false positives. Existing FPN-based methods attempt to mitigate these losses through post-processi

Generich H. Capuli

The environments of severe thunderstorms that produced hail were examined using 171 proximity soundings (2005-2024) archived in the 3rd Data Release of Project SWAP. These soundings were categorized based on their geographical occurrence into three hail-prone environments across Luzon, Philippines. Key parameters describing instability, vertical wind shear,

Fangjie Qi, Caizhi Zhou, Haitao Qing, Haoze Sun

Jumping is an efficient locomotion strategy to traverse cluttered, uneven, or unstable environments in nature, yet replicating continuous, autonomous leaping in soft robots remains challenging due to limited energy storage and reliance on human intervention or latches. Here, we report a millimeter-scale, self-resetting soft ring that achieves repeated vertic

Mohammad Hossein Sameti, Sepehr Harfi Moridani, Ali Zarean, Hossein Sameti

Pre-trained transformer-based models have significantly advanced automatic speech recognition (ASR), yet they remain sensitive to accent and dialectal variations, resulting in elevated word error rates (WER) in linguistically diverse languages such as English and Persian. To address this challenge, we propose an accent-invariant ASR framework that integrates

Hossein Larki, Najmeh Rajabzadeh-Hasiri

The notion of a self-similar ultragraph $(G,\mathcal{U},\varphi)$ and its $C^*$-algebra $\mathcal{O}_{G,\mathcal{U}}$ were introduced in our recent work, where we proposed inverse semigroup and groupoid models for such $C^*$-algebras as well. In this paper, we investigate minimality and effectiveness of the groupoid of a self-similar ultragraph $(G,\mathcal{

Chenghao Wang, Kaushik Venkatesh Krishnamurthy, Shreyansh Pitroda, Adarsh Salagame

Multi-modal ground-aerial robots have been extensively studied, with a significant challenge lying in the integration of conflicting requirements across different modes of operation. The Husky robot family, developed at Northeastern University, and specifically the Husky v.2 discussed in this study, addresses this challenge by incorporating posture manipulat

Francesco Scali, Marco Finazzi, Federico Bottegoni, Carlo Zucchetti

Optical orientation has been proven as a powerful tool to inject spin-polarized electron and hole populations in III-V and group-IV semiconductors. In particular, the absorption of circularly-polarized light in bulk Ge generates a spin-oriented population of electrons in the conduction band with a spin-polarization up to 50%, whereas the hole spin-polarizati

Luka Hansen, Görkem Bilgin, Jan Benedikt

Vacuum ultraviolet (VUV) radiation produced by an atmospheric pressure plasma was successfully measured down to wavelengths of 58.4nm utilizing a 20nm thin Si3N4 membrane to transfer the VUV radiation into a vacuum monochromator. This method allows measurements without disturbing the plasma or the spectra. He2 absorption could be observed by filling the mono

Ian Gill, Ryan J. Hamilton, Helen Caines

Considerable evidence now exists for partonic energy loss due to interaction with the hot, dense medium created in ultra-relativistic heavy-ion collisions. A primary signal of this energy loss is the suppression of high transverse momentum $p_{\mathrm{T}}$ hadron yields in A-A collisions relative to appropriately scaled $pp$ collisions at the same energy. Me

Aparna Saksena, Yujun Zhao, J. Manoj Prabhakar, Dierk Raabe

Platinum, to date, is the most widely applied electrocatalyst for hydrogen evolution reaction (HER) in acidic media. It is assumed to be a proton-blocking catalyst with only surface-limited adsorption of the reaction intermediates. Here, we critically evaluate the bulk interaction of Pt with hydrogen (H), and its heavier isotope deuterium (D), by monitoring

Anthony J. Brady, Zihao Gong, Alexey V. Gorshkov, Saikat Guha

Incoherent imaging, including fluorescence and absorption microscopy, is often limited by weak signals and resolution constraints -- notoriously, Rayleigh's curse. We investigate how spatially structured quantum probes, combined with quantum detection strategies like spatial mode demultiplexing and photon counting, overcome these limitations. We propose a no

Ali Javadi-Abhari, Simon Martiel, Alireza Seif, Maika Takita

Entanglement is the quintessential quantum phenomenon and a key enabler of quantum algorithms. The ability to faithfully entangle many distinct particles is often used as a benchmark for the quality of hardware and control in a quantum computer. Greenberger-Horne-Zeilinger (GHZ) states, also known as Schr\"odinger cat states, are useful for this task. They a

Veronica Rammouz, Aaron Gonzalez, Carlos Cruzportillo, Adrian Tan

Estimating model performance without labels is an important goal for understanding how NLP models generalize. While prior work has proposed measures based on dataset similarity or predicted correctness, it remains unclear when these estimates produce reliable performance rankings across domains. In this paper, we analyze the factors that affect ranking relia

François Monard, Zhengyi Qi

In the literature on X-ray transform and Transport Twistor (TT) spaces, blow-down maps (or maps with holomorphic blow-down structure as defined in [BMP24]) are maps that desingularize the degenerate complex structure of the TT space of an oriented Riemannian surface, while collapsing (yet separating) geodesics of the unit tangent bundle of that surface. Such

Rafael A. Calvo, Dorian Peters

Conversational AI (CAI) systems offer opportunities to scale service provision to unprecedented levels and governments and corporations are already beginning to deploy them across services. The economic argument is similar across domains: use CAI to automate the time-consuming conversations required for customer, client or patient support. Herein we draw on

Oleksandr Tsyplyatyev, Yiqing Jin, María Moreno, Wooi Kiat Tan

The fate of spin-charge separation beyond the low energy remains elusive up to now. Here we develop a microscopic theory of the correlation functions using the strong coupling expansion of the Hubbard model and demonstrate its validity down to the experimentally relevant $r_{\rm s}>1$. Evaluating the spectral function, we show the general stability of the no

Albert Belenguer-Llorens, Carlos Sevilla-Salcedo, Janaina Mourao-Miranda, Vanessa Gómez-Verdejo

Real-world clinical problems are often characterized by multimodal data, usually associated with incomplete views and limited sample sizes in their cohorts, posing significant limitations for machine learning algorithms. In this work, we propose a Bayesian approach designed to efficiently handle these challenges while providing interpretable solutions. Our a

Mark Jones, Jannik Schestag

For a phylogenetic tree, the phylogenetic diversity of a set A of taxa is the total weight of edges on paths to A. Finding small sets of maximal diversity is crucial for conservation planning, as it indicates where limited resources can be invested most efficiently. In recent years, efficient algorithms have been developed to find sets of taxa that maximize

Hugo de Souza Oliveira, Michele Curatolo, Renate Sachse, Edoardo Milana

Variable stiffness is a key capability in biological and robotic systems, enabling adaptive interaction across tasks and environments. Mechanical metamaterials offer an alternative to conventional mechatronic solutions by encoding stiffness variation directly into monolithic structural architectures, reducing the need for discrete assemblies. Here, we introd

Siyue Zhang, Yuan Gao, Xiao Zhou, Yilun Zhao

We introduce MRMR, the first expert-level multidisciplinary multimodal retrieval benchmark requiring intensive reasoning. MRMR contains 1,502 queries spanning 23 domains, with positive documents carefully verified by human experts. Compared to prior benchmarks, MRMR introduces three key advancements. First, it challenges retrieval systems across diverse area

David Vázquez-Padín, Fernando Pérez-González, Alejandro Martín-Del-Río

We investigate diagonal artifacts present in images captured by several Samsung smartphones and their impact on PRNU-based camera source verification. We first show that certain Galaxy S series models share a common pattern causing fingerprint collisions, with a similar issue also found in some Galaxy A models. Next, we demonstrate that reliable PRNU verific

R. T. Sutherland, A. C. Hughes, J. P. Marceaux, H. M. Knaack

We demonstrate a new technique that adapts single-qubit randomized benchmarking to two-qubit M{\o}lmer-S{\o}rensen gates. We use the controllable gate phase to generate Cliffords that act on a two-state subspace, enabling benchmarking of two-qubit gates without single-qubit operations. In addition to quantifying the gate infidelity, the protocol provides val

Zixin Zhang, Kanghao Chen, Xingwang Lin, Lutao Jiang

The ability to use, understand, and create tools is a hallmark of human intelligence, enabling sophisticated interaction with the physical world. For any general-purpose intelligent agent to achieve true versatility, it must also master these fundamental skills. While modern Multimodal Large Language Models (MLLMs) leverage their extensive common knowledge f

Bob Holdom

One or two negative mass singularities are found to occur in static inhomogeneous spatially closed solutions to the Einstein equations. The singularities produce a positive Komar mass, and this decreases the size of the cosmological constant relative to normal matter. The energy density of a perfect fluid vanishes at the singularities and is finite elsewhere

Li Li, Ming Cheng, Juan Liu, Ming Li

This paper proposes a Spatially-Augmented Sequence-to-Sequence Neural Diarization (SA-S2SND) framework, which integrates direction-of-arrival (DOA) cues estimated by SRP-DNN into the S2SND backbone. A two-stage training strategy is adopted: the model is first trained with single-channel audio and DOA features, and then further optimized with multi-channel in

Liping Chen, Chenyang Guo, Kong Aik Lee, Zhen-Hua Ling

Recent advancements in adversarial attacks have demonstrated their effectiveness in misleading speaker recognition models, making wrong predictions about speaker identities. On the other hand, defense techniques against speaker-adversarial attacks focus on reducing the effects of speaker-adversarial perturbations on speaker attribute extraction. These techni

Trevor DePodesta, Johanna Beyer

Existing digital book management platforms often fail to capture the rich spatial and visual cues inherent to physical bookshelves, hindering users' ability to fully engage with their collections. We present LibraryLens, a novel visualization tool that addresses these shortcomings by enabling users to create, explore, and interact with immersive, two-dimensi

Sen-Peng Eu, Yong-Siang Lin, Wei-Liang Sun

Idempotent elements play a fundamental role in ring theory, as they encode significant information about the underlying algebraic structure. In this paper, we study idempotent matrices from two perspectives. First, we analyze the partially ordered set of idempotents in matrix rings over a division ring. We characterize the partial order relation explicitly i

Shiyuan Luo, Runlong Yu, Shengyu Chen, Yingda Fan

Understanding environmental ecosystems is vital for the sustainable management of our planet. However,existing physics-based and data-driven models often fail to generalize to varying spatial regions and scales due to the inherent data heterogeneity presented in real environmental ecosystems. This generalization issue is further exacerbated by the limited ob

Parhom Esmaeili, Virginia Fernandez, Pedro Borges, Eli Gibson

Interactive segmentation is a promising strategy for building robust, generalisable algorithms for volumetric medical image segmentation. However, inconsistent and clinically unrealistic evaluation hinders fair comparison and misrepresents real-world performance. We propose a clinically grounded methodology for defining evaluation tasks and metrics, and buil

Agata Azzolin, Gaia Giovannetti, Oliviero Cannelli, Sabine Rockenstein

Attosecond light sources based on high-order harmonic generation (HHG) constitute to date the only table-top solution for producing coherent broadband radiation covering the spectral range from the extreme ultraviolet to the soft X-rays. The so-called emission cutoff can be extended towards higher photon energies by increasing the driving wavelength at the e

Srikar Allaparapu, Michael Baur, Benedikt Böck, Michael Joham

Robust precoding is efficiently feasible in frequency division duplex (FDD) systems by incorporating the learnt statistics of the propagation environment through a generative model. We build on previous work that successfully designed site-specific precoders based on a combination of Gaussian mixture models (GMMs) and graph neural networks (GNNs). In this pa

Nico Bistolfi, Andreea Georgescu, Dave Hodson

As cloud infrastructure evolves to support dynamic and distributed workflows, accelerated now by AI-driven processes, the outdated model of standing permissions has become a critical vulnerability. Based on the Cloud Security Alliance (CSA) Top Threats to Cloud Computing Deep Dive 2025 Report, our analysis details how standing permissions cause catastrophic

Avik Dutta, Priyanshu Gupta, Hosein Hasanbeig, Rahul Pratap Singh

Real-world data analysis tasks often come with under-specified goals and unclean data. User interaction is necessary to understand and disambiguate a user's intent, and hence, essential to solving these complex tasks. Existing benchmarks for evaluating LLMs on data analysis tasks do not capture these complexities or provide first-class support for interactiv

Iftekhar Ahmed, Tanzil Ebad Chowdhury, Biggo Bushon Routh, Nafisa Tasmiya

Kidneys are the filter of the human body. About 10% of the global population is thought to be affected by Chronic Kidney Disease (CKD), which causes kidney function to decline. To protect in danger patients from additional kidney damage, effective risk evaluation of CKD and appropriate CKD monitoring are crucial. Due to quick and precise detection capabiliti

Sam Lau, Kianoosh Boroojeni, Harry Keeling, Jenn Marroquin

Generative AI (GenAI) tools are increasingly pervasive, pushing instructors to redesign how students use GenAI tools in coursework. We conceptualize this work as emergency pedagogical design: reactive, indirect efforts by instructors to shape student-AI interactions without control over commercial interfaces. To understand practices of lead users conducting

Siddhartha Mangamuri, Lénaïc Couëdel, Surabhi Jaiswal

This article presents a detailed investigation of solid-fluid phase coexistence in a bilayer dusty plasma crystal subjected to varying confinement ring bias voltages in a DC glow discharge argon plasma. Melamine formaldehyde particles were employed to form a stable, hexagonally ordered bilayer crystal within a confinement ring electrically isolated from the

Mehrdad M. Sourki, Wisdom Boinde, Ali N. Amiri, Mahdi Hosseini

Macroscopic systems, when governed by nonlinear interactions, can display rich behavior from persistent oscillations to signatures of ergodicity breaking. Nonlinearity, long regarded as a nuisance in precision systems, is increasingly recognized as a gateway to new physical regimes. While such dynamics have been extensively studied in optics and atomic physi

Deborah Pintani, Ariel Caputo, Noah Lewis, Marc Stamminger

Outdoor scene reconstruction remains challenging due to the stark contrast between well-textured, nearby regions and distant backgrounds dominated by low detail, uneven illumination, and sky effects. We introduce a two-stage Gaussian Splatting framework that explicitly separates and optimizes these regions, yielding higher-fidelity novel view synthesis. In s

Tom Braden, Nicholas Proudfoot

We survey three settings in which dimensions of intersection cohomology groups of algebraic varieties provide deep combinatorial and representation-theoretic information, and computations of the groups themselves have been made using combinatorial sheaves on finite posets. These settings are (1) intersection cohomology of Schubert varieties, the associated K

Shangzhe Li, Dongruo Zhou, Weitong Zhang

We study online adversarial imitation learning (AIL), where an agent learns from offline expert demonstrations and interacts with the environment online without access to rewards. Despite strong empirical results, the benefits of online interaction and the impact of stochasticity remain poorly understood. We address these gaps by introducing a model-based AI

Martin Roelfs, David Eelbode

This paper investigates how the spinor space of the electroweak gauge group $\text{SU}_{I}(2) \times \text{U}_{Y}(1)$ can be derived using recent geometric techniques within the real Clifford Algebra $\mathbb{R}_4 = \text{Cl}_4(\mathbb{R})$. Central to this approach is a novel procedure for constructing the spinor space of $\mathbb{R}_4$ directly, without co

Arnaud Tourin, Yamil Abraham, Marie Palla, Arthur Le Ber

We demonstrate the existence of a frequency band exhibiting acoustic transparency in 2D and 3D dense granular suspensions, enabling the transmission of a low-frequency ballistic wave excited by a high-frequency broadband ultrasound pulse. This phenomenon is attributed to spatial correlations in the structural disorder of the medium. To support this interpret

Hrad Ghoukasian, Bonwoo Lee, Shahab Asoodeh

We study the problem of sampling from a distribution under local differential privacy (LDP). Given a private distribution $P \in \mathcal{P}$, the goal is to generate a single sample from a distribution that remains close to $P$ in $f$-divergence while satisfying the constraints of LDP. This task captures the fundamental challenge of producing realistic-look

Lars Ohnemus, Nils Hantke, Max Weißer, Kai Furmans

Dynamic Scene Graphs (DSGs) provide a structured representation of hierarchical, interconnected environments, but current approaches struggle to capture stochastic dynamics, partial observability, and multi-agent activity. These aspects are critical for embodied AI, where agents must act under uncertainty and delayed perception. We introduce FOGMACHINE , an

Kay Joerg Wiese

We locate the phase-transition line for the Ising model on the fuzzy sphere from a finite-size scaling analysis of its ground-state energy. Our strategy is to write the latter as $E_{GS}(N_m)/N_m = E_{0} + E_1 /N_m + E_{3/2}/N_m^{3/2}+ ...$, and to search for a minimum of $ \chi:=E_{3/2}/E_0$ as a function of the couplings. Conformal perturbation theory pred

Walter D. van Suijlekom, Michael F. Wondrak, Heino Falcke

We consider a gravitational analogue of the Schwinger effect in a cosmological context. While the Schwinger effect is usually attributed to a static electric background, its derivation is actually based on a switching on/off of the electric field in the infinite past/future. Motivated by this, and our previous work on particle production in a gravitational b

Noé Brucy, Enrique Vázquez-Semadeni, Tine Colman, Jérémy Fensch

Star formation in galaxies is a complex phenomenon occurring on a very wide range of scales, and molecular clouds are at the heart of this process. The formation of these structures and the subsequent collapse of the gas within them to form new stars remain unresolved scientific questions. In particular, the role and importance of gravity at between the disk

Sara Merino-Aceituno, Carmela Moschella, Shotaro Otsuka, Christian Schmeiser

The endoplasmic reticulum (ER) is the largest continuous membrane-bound organelle in the cell and plays a central role in the synthesis and turnover of many lipids and proteins. It connects directly to the nucleus through specialized contact points known as ER-nuclear envelope (NE) junctions. In our recent study, we found that these ER-NE junctions are both

Sina Beyraghi, Javad Shabanpour, Giovanni Geraci, Paul Almasan

This work introduces a fully-automated RIS deployment strategy validated through a digital twin, powered by Sionna ray tracing, of a UK city. On a scene calibrated with measured data, the method jointly optimizes RIS placement, orientation, configuration, and BS beamforming across 4G, 5G, and hypothetical 6G frequencies. Candidate RIS sites are identified vi

Conor Hassan, Nasrulloh Loka, Cen-You Li, Daolang Huang

Set-based transformer models for amortized probabilistic inference and meta-learning, such as neural processes, prior-fitted networks, and tabular foundation models, excel at single-pass marginal prediction. However, many applications require joint distributions over multiple predictions. Purely autoregressive architectures generate these efficiently but sac

Paride Crisafulli, Tobias Galla, Antti Karlsson, Salvatore Miccichè

We obtain comorbidity networks starting from medical information stored in electronic health records collected by the Wellbeing Services County of Southwest Finland (Varha). Based on the data, we associate each patient to one or more diseases and construct complex comorbidity networks associated with large patient cohorts characterized by an age interval and