November 2025 arXiv papers — page 40

Luohe Shi, Zuchao Li, Lefei Zhang, Baoyuan Qi

Speculative decoding accelerates LLM inference by utilizing otherwise idle computational resources during memory-to-chip data transfer. Current speculative decoding methods typically assume a considerable amount of available computing power, then generate a complex and massive draft tree using a small autoregressive language model to improve overall predicti

Roger D. Jones, Achille Giacometti, Alan M. Jones

Biological information processing often arises from mesoscopic molecular systems operating far from equilibrium, yet their complexity can make the underlying principles difficult to visualize. In this study, we introduce a macroscopic hydraulic model that serves as an intuitive analog for the molecular switching behavior exhibited by G protein-coupled recept

Duncan Adamson, George B. Mertzios, Paul G. Spirakis

Graph colouring is a fundamental problem for networks, serving as a tool for avoiding conflicts via symmetry breaking, for example, avoiding multiple computer processes simultaneously updating the same resource. This paper considers a generalisation of this problem to \emph{temporal graphs}, i.e., to graphs whose structure changes according to an ordered seq

Marco Rossazza, Andrea Mignone, Matteo Bugli, Stefano Truzzi

We present preliminary performance results of gPLUTO, the new GPU-optimized implementation of the PLUTO code for computational plasma astrophysics. Like its predecessor, gPLUTO employs a finite-volume formulation to numerically solve the equations of magnetohydrodynamics (MHD) in multiple spatial dimensions. Still, this new implementation is a complete rewri

N. Sahakyan, D. Bégué, P. Giommi, H. Dereli-Bégué

Understanding the impact of spectral coverage on parameter recovery is critical for accurate interpretation of blazar spectra. In this study, we examine how the data coverage influences the reliability of parameter estimation within the one-zone synchrotron self-Compton (SSC) framework. Using OJ 287, TXS 0506+056, and Mrk 421 as representative of the low-, i

Onur Berk Tore, Ibrahim Samil Yalciner, Server Calap

Estimating homography from a single image remains a challenging yet practically valuable task, particularly in domains like retail, where only one viewpoint is typically available for shelf monitoring and product alignment. In this paper, we present a deep learning framework that predicts a 4-point parameterized homography matrix to rectify shelf images capt

Salah Abdeljabar, Mohamed-Slim Alouini

Access to quality education remains unequal, particularly in rural areas where Internet connectivity is limited or nonexistent. This paper introduces a framework for a digital learning platform that uses Delay Tolerant Networking (DTN) to extend educational opportunities to underserved communities. Unlike conventional models that rely on continuous Internet

Ali Madad

InvisibleBench is a deployment gate for caregiving-relationship AI, evaluating 3-20+ turn interactions across five dimensions: Safety, Compliance, Trauma-Informed Design, Belonging/Cultural Fitness, and Memory. The benchmark includes autofail conditions for missed crises, medical advice (WOPR Act), harmful information, and attachment engineering. We evaluate

Zachary Gardner, Jeroen Hekking

We develop the basic theory of derived quasi-coherent ideals for stacks relative to a given derived algebraic context. We compare different notions of adic completeness with respect to derived ideals, define and compare formal spectra and formal completions along closed immersions, and connect the theory of derived ideals to that of derived extended Rees alg

Yuhan Wu, Tiantian Wei, Shuo Wang, ZhiChao Wang

Interactive articulated manipulation requires long-horizon, multi-step interactions with appliances while maintaining physical consistency. Existing vision-language and diffusion-based policies struggle to generalize across parts, instances, and categories. We first introduce ArtiBench, a five-level benchmark covering kitchen, storage, office, and tool envir

Max Potratzki, Manuel Adams, Timo Bröhl, Klaus Lehnertz

We introduce circulance, a scalar measure for classifying time series of dynamical systems. Circulance captures the extent of temporal regularity or irregularity that is encoded in the topology of a directed ordinal pattern transition network derived from a time series. We demonstrate numerically that circulance sensitively and robustly positions time series

Jacob Haqq-Misra, Ravi Kopparapu, George Profitiliotis

The search for technosignatures--remotely detectable evidence of extraterrestrial technology--draws upon examples from the recent history of Earth as well as projections of Earth's technosphere. Facilities like the Habitable Worlds Observatory (HWO) will significantly advance the feasibility of characterizing the atmospheres of habitable exoplanets at visibl

Stefan Lengauer, Sarah Annabelle Von Götz, Marie-Therese Hoesch, Florian Dieter Steinwidder

Interaction data is widely used in multiple domains such as cognitive science, visualization, human computer interaction, and cybersecurity, among others. Applications range from cognitive analyses over user/behavior modeling, adaptation, recommendations, to (user/bot) identification/verification. That is, research on these applications - in particular those

Michael Kilgour, Mark E. Tuckerman, Jutta Rogal

We present MXtalTools, a flexible Python package for the data-driven modelling of molecular crystals, facilitating machine learning studies of the molecular solid state. MXtalTools comprises several classes of utilities: (1) synthesis, collation, and curation of molecule and crystal datasets, (2) integrated workflows for model training and inference, (3) cry

Saptarshi Majumdar, Aleksandra Petković

We study the motion of an impurity under the action of a constant force through a one-dimensional system of weakly-interacting bosons. The interplay of the impurity-boson interaction, the boson-boson interaction, and the driving force gives rise to a rich dynamics. We focus on the influence of a finite external force. Under these far-from-equilibrium conditi

Tianyi Yan, Tao Tang, Xingtai Gui, Yongkang Li

End-to-end models for autonomous driving hold the promise of learning complex behaviors directly from sensor data, but face critical challenges in safety and handling long-tail events. Reinforcement Learning (RL) offers a promising path to overcome these limitations, yet its success in autonomous driving has been elusive. We identify a fundamental flaw hinde

Yulieth Prieto-Montañez, Ian Selvaggi

We prove that very general, dual Gushel-Mukai surfaces are not isomorphic, though derived and L-equivalent. We use this result to study two semiorthogonal decompositions for a family of Fano fourfolds of K3 type, answering a question by Bernardara-Fatighenti-Manivel-Tanturri.

Jules Tindzogho Ntsiri, Samuel Zamour

We prove the existence of Cartan subrings, i.e., self-normalizing nilpotent subrings in soluble ranked Lie rings.

Stefan Perko

In this thesis, we extend the recently introduced theory of stochastic modified equations (SMEs) for stochastic gradient optimization algorithms. In Ch. 3 we study time-inhomogeneous SDEs driven by Brownian motion. For certain SDEs we prove a 1st and 2nd-order weak approximation properties, and we compute their linear error terms explicitly, under certain re

Patrick Kenny

We seek to clarify the concept of active inference by disentangling it from the Free Energy Principle. We show how the optimizations that need to be carried out in order to implement active inference in discrete state spaces can be formulated as constrained divergence minimization problems which can be solved by standard mean field methods that do not appeal

Kevin A. Urquía-Calderón, Oleg Ruchayskiy

The minimal type-I seesaw model provides a simple explanation of neutrino flavor oscillations and induces charged lepton flavor violation (cLFV). Despite extensive previous studies, semileptonic cLFV channels remain underexplored. Using updated form factors, decay constants, and oscillation data, we revisit $\tau$ and meson decay channels, performing a syste

Xuelin Qian, Jiaming Lu, Zixuan Wang, Wenxuan Wang

Infrared Small Target Detection (IRSTD) faces significant challenges due to low signal-to-noise ratios, complex backgrounds, and the absence of discernible target features. While deep learning-based encoder-decoder frameworks have advanced the field, their static pattern learning suffers from pattern drift across diverse scenarios (\emph{e.g.}, day/night var

Shanshan Luo, Peng Wu, Zhi Geng

Online advertising aims to increase user engagement and maximize revenue, but users respond heterogeneously to ad exposure. Some users purchase only when exposed to ads, while others purchase regardless of exposure, and still others never purchase. This heterogeneity can be characterized by latent response types, commonly referred to as principal strata, def

A. I. Perminov

Matrix multiplication optimization remains a fundamental challenge in computational mathematics. This work introduces a novel approach that discovers matrix multiplication schemes whose coefficients are restricted to the set $\{-1, 0, 1\}$ (denoted $Z_T$), minimizing naive additive complexity for efficient hardware implementation. The core of the method is a

Rafael Froner Prando, Pietro Speziali

Function field lattices are an interesting example of algebraically constructed lattices. Their minimum distance is bounded below by a function of the gonality of the underlying function field. Known explicit examples--coming mostly from elliptic and Hermitian curves--typically meet this lower bound. In this paper, we construct, for every integer $n \geqslan

Abhinav Joshi, Divyanshu Bhatt, Ashutosh Modi

Large Language Models (LLMs) show strong generalization across diverse tasks, yet the internal decision-making processes behind their predictions remain opaque. In this work, we study the geometry of hidden representations in LLMs through the lens of \textit{intrinsic dimension} (ID), focusing specifically on decision-making dynamics in a multiple-choice que

Roshani Madurawala, Kerstin Meurisch, Louis Joswig, Anna Lina Wyschkon

The human brain, with its energy-efficient and massively parallel architecture seamlessly integrates memory and computation. Its topology and functionality serve as the inspiration for the field of neuromorphic computing. Realizing brain-like hardware requires the integration of fundamental properties such as synaptic plasticity, self-organization, hierarchi

Li Zhou, Marc Dacier, Charalambos Konstantinou

The Software Bill of Materials (SBOM) is a critical tool for securing the software supply chain (SSC), but its practical utility is undermined by inaccuracies in both its generation and its application in vulnerability scanning. This paper presents a large-scale empirical study on 2,414 open-source repositories to address these issues from a practical standp

Alexander Beiser, Flavio Martinelli, Wulfram Gerstner, Johanni Brea

Network weights can be reverse-engineered given enough informative samples of a network's input-output function. In a teacher-student setup, this translates into collecting a dataset of the teacher mapping -- querying the teacher -- and fitting a student to imitate such mapping. A sensible choice of queries is the dataset the teacher is trained on. But curre

Qiaoyun Ma, Hui Dou, Yiting Chen, Guangyi Jia

Landau level (LL) engineered photonic spin Hall effect (PSHE) holds great promise for nanoscale manipulation and steering of magneto-optical transport in two-dimensional atomic systems. Herein, we theoretically investigate PSHE modulated by LL transitions {\delta}n = n'-n =-2, 0, +2 (where n and n' indicate the LL indexes of valence and conduction bands, res

Pragyan Pratim Bordoloi, Shubh Mittal, Shabnam Iyyani

Gamma-ray burst (GRB) prompt emission is commonly attributed to non-thermal radiation processes operating in the optically thin regions of a relativistic outflow. Among these, optically thin inverse-Compton (IC) scattering remains an important yet under-tested mechanism. From an initial set of 41 bursts selected using empirical Band-function criteria that hi

Kawon Han, Christos Masouros, Taneli Riihonen, Moeness G. Amin

Integrated sensing and communications (ISAC), which is recognized as a key enabler for sixth generation (6G), has brought new opportunities for intelligent, sustainable, and connected wireless networks. Multiple-input multiple-output (MIMO) transceiver technology lies at the core of this paradigm, providing the degrees of freedom required for simultaneous da

Shengqian Li, Ming Gao, Yi Liu, Zuzeng Lin

Rectified Flow (RF) models have advanced high-quality image and video synthesis via optimal transport theory. However, when applied to image-to-image translation, they still depend on costly multi-step denoising, hindering real-time applications. Although the recent adversarial training paradigm, CycleGAN-Turbo, works in pretrained diffusion models for one-s

Han Guo, Chenyang Liu, Haotian Zhang, Bowen Chen

Remote sensing change detection (RSCD) aims to identify surface changes across bi-temporal satellite images. Most previous methods rely solely on mask supervision, which effectively guides spatial localization but provides limited constraints on the temporal semantic transitions. Consequently, they often produce spatially coherent predictions while still suf

Hongfeng Liu, Zizhao Han, Xinfang Nie, Zhenhuan Liu

Non-Markovian noise, arising from environmental memory effects, is the most general and challenging form of noise in quantum computing, and is typically difficult to characterize and suppress. Here, we analyze and experimentally demonstrate a non-Markovian noise suppression scheme inspired by quantum purification protocols. We theoretically prove that, even

Chengfeng Shen, Felix Kübler, Zhennan Zhou

In this paper we examine non-convex dynamic optimization problems with forward looking constraints. We prove that the recursive multiplier formulation in \cite{marcet2019recursive} gives the optimal value if one assumes that the planner has access to a public randomization device and forward looking constraints only have to hold in expectations. Whether one

Zexin Hu, Daniela D. Doneva, Stoytcho S. Yazadjiev, Lijing Shao

Observations of gravitational waves (GWs) generated by binary black hole (BBH) mergers provide us with a powerful way to explore the strong and highly dynamical regime of gravity theories. The ringdown of BBH merger, consisting of a series of quasi-normal modes (QNMs), is of particular interest for both the black hole (BH) spectroscopy and the inspiral-merge

Linjing Feng, Sihan Jiao, Fengwei Xu, Hauyu Baobab Liu

We characterize star-forming gas in six molecular clouds (Sgr B1-off, Sgr B2, Sgr C, the 20 km s$^{-1}$ and 50 km s$^{-1}$ molecular clouds, and the Brick) in the Galactic central molecular zone (CMZ), and compare their star-forming activities with those in molecular clouds outside the CMZ. Using multi-band continuum observations taken from ${\it Planck}$, $

Róisín Keenan, Joost C. Dessing

Recent advancements in robotics have increased the possibilities for integrating robotic systems into human-involved workplaces, highlighting the need to examine and optimize human-robot coordination in collaborative settings. This study explores human-robot interactions during handover tasks using Virtual Reality (VR) to investigate differences in human mot

Godfred Kumi Tenkorang, Michel Daoud Yacoub

This paper derives second-order statistics for diversity-combining techniques over Log-mu fading channels. Closed-form expressions for the level crossing rate (LCR) and average fading duration (AFD) are derived for pure selection combining (PSC), while exact multidimensional integral expressions are obtained for equal gain combining (EGC) and maximal ratio c

Baoshun Shi, Ke Jiang, Qiusheng Lian, Xinran Yu

Despite significant advancements in deep learning-based sparse-view computed tomography (SVCT) reconstruction algorithms, these methods still encounter two primary limitations: (i) It is challenging to explicitly prove that the prior networks of deep unfolding algorithms satisfy Lipschitz constraints due to their empirically designed nature. (ii) The substan

Dnyandeep Mandaokar, Bernhard Rinner

Tracking maneuvering targets requires estimators that are both responsive and robust. Interacting Multiple Model (IMM) filters are a standard tracking approach, but fusing models via Gaussian mixtures can lag during maneuvers. Recent winnertakes-all (WTA) approaches react quickly but may produce discontinuities. We propose SAFE-IMM, a lightweight IMM variant

Chaowei He, Yuanjun Liu, Qingzhi Ma, Shenyuan Ren

Machine unlearning in learned cardinality estimation (CE) systems presents unique challenges due to the complex distributional dependencies in multi-table relational data. Specifically, data deletion, a core component of machine unlearning, faces three critical challenges in learned CE models: attribute-level sensitivity, inter-table propagation and domain d

Dong Wang, Daniel Casado Herraez, Stefan May, Andreas Nüchter

Reliable odometry in highly dynamic environments remains challenging when it relies on ICP-based registration: ICP assumes near-static scenes and degrades in repetitive or low-texture geometry. We introduce Dynamic-ICP, a Doppler-aware registration framework. The method (i) estimates ego motion from per-point Doppler velocity via robust regression and builds

Yu Dian Lim, Chuan Seng Tan

Artificial Intelligence (AI) software based on transformer model is developed to automatically design gratings for possible integrations in ion traps to perform optical addressing on ions. From the user-defined (x,z) coordinates and full-width half-maximum (FWHM) values, the AI software can automatically generate the Graphic Design System (GDS) layout of the

Xuebo Qiu, Mingqi Lv, Yimei Zhang, Tieming Chen

Provenance-based threat hunting identifies Advanced Persistent Threats (APTs) on endpoints by correlating attack patterns described in Cyber Threat Intelligence (CTI) with provenance graphs derived from system audit logs. A fundamental challenge in this paradigm lies in the modality gap -- the structural and semantic disconnect between provenance graphs and

Kang Wang, Renzhe Xu, Bo Li

Understanding the bias-variance tradeoff in user representation learning is essential for improving recommendation quality in modern content platforms. While well studied in static settings, this tradeoff becomes significantly more complex when content creators strategically adapt to platform incentives. To analyze how such competition reshapes the tradeoff

Yongming Zhang

In this paper, we prove that for any smooth projective curve $C$ of genus $g\geq2$ over an algebraically closed field of positive characteristic, there exists a stable vector bundle over $C$ whose exterior power is not semi-stable.

Anton V. Hlushchenko, Mykhailo I. Bratchenko, Aleksei V. Chechkin

Current-driven motion of domain walls and skyrmions is central to the operation of non-volatile magnetic memory devices. Racetrack memory requires current densities high enough to generate velocities above 50 m/s, but such conditions also enhance spin-current noise. We develop a theoretical framework based on the stochastic Thiele equation to analyze the eff

Giorgos Anastasiou, Ignacio J. Araya, Avik Chakraborty, Cristóbal Corral

The asymptotic analysis for the metric of a generic solution of Einstein-Gauss-Bonnet AdS theory is provided by solving the field equations in the Fefferman-Graham frame. Using standard holographic renormalization, the counterterms that render the action finite are found up to seven spacetime dimensions. In the case of 6D, an equivalent formulation that perm

Mingyu Jeon, Jaeyoung Suh, Suwan Cho

Schema matching is a critical task in data integration, particularly in the medical domain where disparate Electronic Health Record (EHR) systems must be aligned to standard models like OMOP CDM. While Large Language Models (LLMs) have shown promise in schema matching, they suffer from hallucination and lack of up-to-date domain knowledge. Knowledge Graphs (

Friederike Groschupp, Daniele Lain, Aritra Dhar, Lara Magdalena Lazier

Precise access control decisions are crucial for the security of both traditional applications and emerging agent-based systems. Typically, these decisions are made by users during app installation or at runtime. However, due to the increasing complexity and automation of systems, making access control decisions can impose a significant cognitive burden on u

Marta Grzeskiewicz

Understanding household behaviour is essential for modelling macroeconomic dynamics and designing effective policy. While heterogeneous agent models offer a more realistic alternative to representative agent frameworks, their implementation poses significant computational challenges, particularly in continuous time. The Aiyagari-Bewley-Huggett (ABH) framewor

Yang Liu, Xilin Zhao, Peisong Wen, Siran Dai

Recent progress in video generation has led to impressive visual quality, yet current models still struggle to produce results that align with real-world physical principles. To this end, we propose an iterative self-refinement framework that leverages large language models and vision-language models to provide physics-aware guidance for video generation. Sp

Fabian Gülhan, Emil Mededovic, Yuli Wu, Johannes Stegmaier

End-to-end transformer architectures have driven significant progress in multi-object tracking by unifying detection and association into a single, heuristic-free framework. Despite these benefits, poor detection performance and the inherent conflict between detection and association in a joint architecture remain critical concerns. Recent approaches aim to

Yinghui Li, Qianyu Zhou, Di Shao, Hao Yang

Domain adaptive point cloud completion (DA PCC) aims to narrow the geometric and semantic discrepancies between the labeled source and unlabeled target domains. Existing methods either suffer from limited receptive fields or quadratic complexity due to using CNNs or vision Transformers. In this paper, we present the first work that studies the adaptability o

Yiheng Zhang, Shaowu Wu, Yuanzhuo Xu, Jiajun Wu

Adaptive optimizers such as Adam have achieved great success in training large-scale models like large language models and diffusion models. However, they often generalize worse than non-adaptive methods, such as SGD on classical architectures like CNNs. We identify a key cause of this performance gap: adaptivity in pre-conditioners, which limits the optimiz

Lianzhe Hu, Yu Wang, Bikash Pal

This paper presents an LLM-driven, end-to-end workflow that addresses the lack of automation and intelligence in power system transient stability assessment (TSA). The proposed agentic framework integrates large language models (LLMs) with a professional simulator (ANDES) to automatically generate and filter disturbance scenarios from natural language, and e

Chenhui Dong, Haozhe Xu, Wenhao Feng, Zhipeng Wang

Reinforcement learning (RL) controllers have made impressive progress in humanoid locomotion and light-weight object manipulation. However, achieving robust and precise motion control with intense force interaction remains a significant challenge. To address these limitations, this paper proposes HAFO, a dual-agent reinforcement learning framework that concu

Areeb Ahmad, Abhinav Joshi, Ashutosh Modi

Transformer-based language models exhibit complex and distributed behavior, yet their internal computations remain poorly understood. Existing mechanistic interpretability methods typically treat attention heads and multilayer perceptron layers (MLPs) (the building blocks of a transformer architecture) as indivisible units, overlooking possibilities of funct

Baptiste Devyver, Louis Dupaigne, Pierre-Damien Thizy

In the Euclidean space $\mathbb{R}^d$, the sharp classical Sobolev inequality is equivalent by conformal invariance to a Sobolev inequality on the hyperbolic space $\mathbb{H}^d$. This inequality is sharp in dimension $d\geq 4$, but it is not in dimension $d=3$ by results of Benguria, Frank and Loss, as well as Mancini and Sandeep. In this article, we invest

Amirhossein Khadivi Noghredeh, Abdollah Safari, Fatemeh Ziaeetabar, Firoozeh Haghighi

Anomaly detection in industrial visual inspection is challenging due to the scarcity of defective samples. Most existing methods rely on unsupervised reconstruction using only normal data, often resulting in overfitting and poor detection of subtle defects. We propose a semi-supervised deep reinforcement learning framework that integrates a neural batch samp

Siqi Ding, Xiaobo Jin, Fengchun Lei, Fengling Li

In this paper, we introduce the 0-smoothing invariant $\mathcal{F}$ of virtual knotoids constructed from local modification at classical crossings, which take values in a free $\mathbb Z$-module generated by non-oriented flat virtual knotoids. We prove that $\mathcal{F}$ is a Vassiliev invariant of order one. It was observed by Henrich that smoothing invaria

E. Salibur, A. Hallé, F. Combes

Galaxy disks in rotation are sometimes the site of radial flows, especially in their gas component. It is important to estimate the outflows, due to AGN or supernovae feedback, or inflows due to bar gravity torques. However, these radial flows may be confused with non-circular motions, which are quite frequent in the center of galaxy disks. We use a simulate

Ekta Bhatia, Jack Lombardi, Tuan Vo, Michael Senatore

Josephson junctions form the core circuit element in superconducting quantum computing circuits, single flux quantum digital logic circuits, and sensing devices such as SQUIDs. Aluminum oxide has typically been used as the tunnel barrier. Its formation by exposure to low oxygen pressures at room temperature for short periods of time makes it susceptible to a

Can Zheng, Jiguang He, Chung G. Kang, Guofa Cai

This paper proposes a vision-conditioned flow matching (FM) framework for beam prediction in millimeter-wave vehicle-to-infrastructure links. Instead of modeling discrete beam-index sequences, the proposed method learns the temporal evolution of normalized beam receive power vectors through a continuous vector field governed by an ordinary differential equat

Shuijin Zhang, Jialin Wang, Yu Zheng, Xiang Li

We apply the moving plane method in integral forms to classify the positive solutions of the critical Hartree equation on Heisenberg group \begin{equation}\label{0.1} -Δ_{\mathbb{H}}u=\left(\int_{\mathbb{H}^{n}}\frac{|u(ξ)|^{Q^{\ast}_μ}}{|ζ^{-1}ξ|^μ}\mathrm{d}ξ\right)|u|^{Q^{\ast}_μ-2}u,~~~ζ,ξ\in\mathbb{H}^{n}, \end{equation} where $Δ_{\mathbb{H}}$ denotes t

Omer Belhasin, Shelly Golan, Ran El-Yaniv, Michael Elad

Image classification is a well-studied task in computer vision, and yet it remains challenging under high-uncertainty conditions, such as when input images are corrupted or training data are limited. Conventional classification approaches typically train models to directly predict class labels from input images, but this might lead to suboptimal performance

Shah Fahad, Gao Xianlong

We propose a scheme to manipulate the Goos-H\"{a}nchen shift (GHS) of a reflected probe field in a non-Hermitian cavity magnomechanical system. The platform consists of a yttrium-iron-garnet sphere coupled to a microwave cavity, where a strong microwave drive pumps the magnon mode and a weak field probes the cavity. The traveling field's interaction with the

Xiang Li, Xiangjian Qian, Mingpu Qin

In this work, we investigate the Kitaev honeycomb model employing the recently developed Clifford Circuits Augmented Matrix Product States (CAMPS) method. While the model in the gapped phase is known to reduce to the toric code model - whose ground state is entirely constructible from Clifford circuits - we demonstrate that the very different gapless quantum

Lorenzo Brasco, Luca Briani, Francesca Prinari

We consider periodically perforated unbounded open sets and prove existence of extremals for the relevant sharp Poincar\'e-Sobolev embedding constant. The existence result holds no matter the shape or the regularity of the hole: it is sufficient that the latter is a compact set with positive capacity. We also show how to apply the main result in order to get

Diego Díaz, Aman Bhargava, Franziska Walz, Azadeh Sharifi

The wetting behavior of drops on natural and industrial surfaces is determined by the advancing and receding contact angles. They are commonly measured by the sessile drop technique, also called goniometry, which doses liquid through a solid needle. Consequently, this method requires substantial drop volumes, long contact times, tends to be user-dependent, a

Xiaohan Wang, Zhangtao Cheng, Ting Zhong, Leiting Chen

Weight Averaging (WA) has emerged as a powerful technique for enhancing generalization by promoting convergence to a flat loss landscape, which correlates with stronger out-of-distribution performance. However, applying WA directly to multi-modal domain generalization (MMDG) is challenging: differences in optimization speed across modalities lead WA to overf

Zhiguo Zhang, Xiaoliang Ma, Daniel Schlesinger

Accurate and interpretable air pollution forecasting is crucial for public health, but most models face a trade-off between performance and interpretability. This study proposes a physics-guided, interpretable-by-design spatiotemporal learning framework. The model decomposes the spatiotemporal behavior of air pollutant concentrations into two transparent, ad

Weijia Mao, Hao Chen, Zhenheng Yang, Mike Zheng Shou

A reliable reward function is essential for reinforcement learning (RL) in image generation. Most current RL approaches depend on pre-trained preference models that output scalar rewards to approximate human preferences. However, these rewards often fail to capture human perception and are vulnerable to reward hacking, where higher scores do not correspond t

Aamna Ahmed, Mónica Benito, Beatriz Pérez-González

Flat-band lattices hosting compact localized states are highly sensitive to external modulation, and the tailored design of a perturbation to imprint specific features becomes relevant. Here we show that periodic driving in the high-frequency regime transforms the all-flat-band diamond chain into one featuring two tunable quasi-flat bands and a residual flat

Alexander C. Jenke, Gregor Just, Claas de Boer, Martin Wagner

Purpose: Robot-assisted minimally invasive surgery relies on endoscopic video as the sole intraoperative visual feedback. The DaVinci Xi system overlays a graphical user interface (UI) that indicates the state of each robotic arm, including the activation of the endoscope arm. Detecting this activation provides valuable metadata such as camera movement infor

Andrey Lemeshko, Bulat Gabdullin, Nikita Drozdov, Anton Konushin

3D object detection is fundamental for spatial understanding. Real-world environments demand models capable of recognizing diverse, previously unseen objects, which remains a major limitation of closed-set methods. Existing open-vocabulary 3D detectors relax annotation requirements but still depend on training scenes, either as point clouds or images. We tak

Gabriel K. Gegenhuber, Philipp É. Frenzel, Maximilian Günther, Johanna Ullrich

WhatsApp, with 3.5 billion active accounts as of early 2025, is the world's largest instant messaging platform. Given its massive user base, WhatsApp plays a critical role in global communication. To initiate conversations, users must first discover whether their contacts are registered on the platform. This is achieved by querying WhatsApp's servers with mo

Bo-Kai Ruan, Teng-Fang Hsiao, Ling Lo, Yi-Lun Wu

Modern text-to-image models produce impressive visual results from richly specified prompts, yet their behavior under long prompts remains insufficiently understood. In this paper, we study a practical failure mode in which accumulated semantic constraints progressively suppress output variation, causing diversity to collapse even when many visual factors re

Daniel Kienzle, Katja Ludwig, Julian Lorenz, Shin'ichi Satoh

Obtaining the precise 3D motion of a table tennis ball from standard monocular videos is a challenging problem, as existing methods trained on synthetic data struggle to generalize to the noisy, imperfect ball and table detections of the real world. This is primarily due to the inherent lack of 3D ground truth trajectories and spin annotations for real-world

Sébastien Bonte, Gauvain Devillez, Valentin Dusollier, Alain Hertz

Topological indices are graph-theoretic descriptors that play a crucial role in mathematical chemistry, capturing the structural characteristics of molecules and enabling the prediction of their physicochemical properties. A widely studied category of topological indices, known as degree-based topological indices, are calculated as the sum of the weights of

Jan Majewski, Francesca Giardini

Gossip has been shown to be a relatively efficient solution to problems of cooperation in reputation-based systems of exchange, but many studies don't conceptualize gossiping in a realistic way, often assuming near-perfect information or broadcast-like dynamics of its spread. To solve this problem, we developed an agent-based model that pairs realistic gossi

Ziyuan Gao, Philippe Morel

Medical AI systems face catastrophic forgetting when deployed in clinical settings, where models must learn new imaging protocols while retaining prior diagnostic capabilities. This challenge is particularly acute for medical vision-language models that must preserve complex cross-modal alignments between medical images and clinical terminology across divers

Duy-Duc Dao, Frédéric Nowacki

We present recent developments of the Discrete Non-Orthogonal Shell Model (DNO-SM) for nuclear structure studies far from stability. Exact shell-model solutions are obtained for typical open-shell light sd and pf nuclei using non-orthogonal Slater determinants consistently derived from the variation after projection approach. The latter represents a powerful

Jørgen Bang-Jensen, Lucas Picasarri-Arrieta, Anders Yeo

The dichromatic number $\vec{\chi}(D)$ of a digraph $D=(V,A)$ is the minimum number of sets in a partition $V_1,\ldots{},V_k$ of $V$ into $k$ subsets so that the induced subdigraph $D[V_i]$ is acyclic for each $i\in [k]$. This is a generalization of the chromatic number for undirected graphs as a graph has chromatic number at most $k$ if and only if the comp

Jawaria Maqbool, M. Imran Cheema

Existing deep learning methods in multimode fiber (MMF) imaging often focus on simpler datasets, limiting their applicability to complex, real-world imaging tasks. These models are typically data-intensive, a challenge that becomes more pronounced when dealing with diverse and complex images. In this work, we propose HistoSpeckle-Net, a deep learning archite

João G. A. Caribé, Marcelo S. Guimarães, Itzhak Roditi, Silvio P. Sorella

We investigate the mass dependence of the Araki-Uhlmann relative entropy between a localized coherent excitation and the vacuum state of a free scalar quantum field on the $(1+d)$-dimensional Minkowski spacetime for $d = 1, 2, 3$. In this context, the relative entropy admits a closed expression in terms of the smeared Pauli-Jordan distribution, whose analyti

Stefan Marian Ludwig

We introduce the theory $\mathrm{PF}^{+,\times}$ of pseudofinite fields with generic additive and multiplicative character added as continuous logic predicates. Using the Weil bounds on character sums over finite fields as well as the Erd\H{o}s-Tur\`an-Koksma inequality we show that it is the asymptotic theory (in characteristic $0$) of finite fields with (s

Shi-Jie Gao, Xiang-Dong Li, Song Wang, Kareem El-Badry

Recent optical astrometric and spectroscopic surveys have identified numerous neutron star (NS) candidates in non-accreting detached binary systems, but their compact-object nature remains unconfirmed. In this work, we present targeted radio observations of 31 such candidates using the Five-hundred-meter Aperture Spherical radio Telescope (FAST), the Robert

Gregory F. Stock, Alexander Haberl, Juan A. Fraire, Holger Hermanns

Routing in Delay-Tolerant Networks (DTNs) is inherently challenging due to sparse connectivity, long delays, and frequent disruptions. While Markov Decision Processes (MDPs) have been used to model uncertainty, they assume full state observability - an assumption that breaks down in partitioned DTNs, where each node operates with inherently partial knowledge

Chun Song, Minfu Feng

This paper presents an enriched Galerkin (EG) finite element method for the incompressible Navier--Stokes equations. The method augments continuous piecewise linear velocity spaces with elementwise bubble functions, yielding a locally conservative velocity approximation while retaining the efficiency of low-order continuous elements. The viscous term is disc

Jan Krejčí, Oliver Kost, Yuxuan Xia, Lennart Svensson

This paper addresses multi-object systems, where objects may occlude one another relative to the sensor. The standard point-object model for detection-based sensors is enhanced so that the probability of detection considers the presence of all objects. A principled tracking method is derived, assigning each object an expected probability of detection, where

Ondřej Červinka, Vlastimil Weiss, Martin Hrtoň, Petr Bouchal

Digital holographic microscopy is a powerful tool for characterizing transparent and reflective phase objects. Its ability to reconstruct amplitude and phase can also offer great insight into wavefront shaping and design of all-dielectric optical metasurfaces. While metasurfaces have reached widespread popularity, their design is often based purely on the re

Zeynab Kaseb, Matthias Moller, Lindsay Spoor, Jerry J. Guo

The Newton-Raphson (NR) method is widely used for solving power flow (PF) equations due to its quadratic convergence. However, its performance deteriorates under poor initialization or extreme operating scenarios, e.g., high levels of renewable energy penetration. Traditional NR initialization strategies often fail to address these challenges, resulting in s

Liren Yu, Wenming Zhang, Silu Zhou, Tao Zhang

We propose HHFT (Hierarchical Heterogeneous Feature Transformer), a Transformer-based architecture tailored for industrial CTR prediction. HHFT addresses the limitations of DNN through three key designs: (1) Semantic Feature Partitioning: Grouping heterogeneous features (e.g. user profile, item information, behaviour sequennce) into semantically coherent blo

Olivier Moulin, Vincent Francois-lavet, Paul Elbers, Mark Hoogendoorn

Generalizability of Reinforcement Learning (RL) agents (ability to perform on environments different from the ones they have been trained on) is a key problem as agents have the tendency to overfit to their training environments. In order to address this problem and offer a solution to increase the generalizability of RL agents, we introduce a new methodolog

Mamatha Ramanjineyulu Maddur, Hemansh Shah, Praveen Pathak

In this paper we show how students can measure optical features of smartphone displays through three experiments. Observing diffraction patterns from smartphone displays allows students to determine the Pixels Per Inch (PPI). Observing reflections within a smartphone display provides information about touch glass thickness and pixel layer properties. Finally

Louisa A. Mason, Michael A. Garrett, Andrew P. V. Siemion

Searches for radio technosignatures place constraints on the prevalence of extraterrestrial transmitters in our Galaxy and beyond. It is important to account for the complete stellar population captured within a radio telescope's field of view, or stellar 'bycatch'. In recent years, catalogues from ESA's Gaia mission have enabled SETI surveys to place tighte

Sofiane Bouarroudj, Hamza El Ouali

A non-associative superalgebra is called pre-symplectic if it is equipped with a non-degenerate, anti-symmetric bilinear form. It is called quasi-Frobenius if, in addition, is a Lie superalgebra and the form is closed. We introduce the Levi-Civita product associated with pre-symplectic superalgebras and establish its existence and uniqueness. We then introdu