December 2025 arXiv papers — page 89

Gangxiong Zhang, Yongchao Long, Yuxi Zhou, Yong Zhang

Accurately predicting mortality risk in intensive care unit (ICU) patients is essential for clinical decision-making. Although large language models (LLMs) show strong potential in structured medical prediction tasks, their outputs may exhibit biases related to demographic attributes such as sex, age, and race, limiting their reliability in fairness-critical

Michael Amir, Manon Flageat, Amanda Prorok

The success of machine learning for real-world robotic systems has created a new form of intellectual property: the trained policy. This raises a critical need for novel methods that verify ownership and detect unauthorized, possibly unsafe misuse. While watermarking is established in other domains, physical policies present a unique challenge: remote detect

Honey Singh Chauhan, Zahraa S. Abdallah

Kernel-based methods such as Rocket are among the most effective default approaches for univariate time series classification (TSC), yet they do not perform equally well across all datasets. We revisit the long-standing intuition that different representations capture complementary structure and show that selectively fusing them can yield consistent improvem

Andrea Ciaccio, Francesco Moscone, Elisa Tosetti

In this paper we investigate the causal impact of the European Union Emissions Trading System, a cap-and-trade scheme limiting greenhouse gas emissions of firms, on their environmental performance. Although previous studies have focused primarily on the effect of the emission cap imposed by the policy, we argue that the trading mechanism creates complex inte

Michael Brandenbursky, Jarek Kedra, Michal Marcinkowski, Egor Shelukhin

Let $M$ be a smooth compact oriented connected manifold, and ${\rm Homeo}_0(M,\mu)$ the group of homeomorphisms of $M$ supported away from $\partial M,$ which preserve a Borel probability measure $\mu$ induced by a volume form on $M$, and are isotopic to the identity. In this paper, we identify those Gambaudo-Ghys and Polterovich quasimorphisms $\Psi\colon {

Sebastian Resch-Schopper, Romain Rumpler, Gerhard Müller

Parametric model order reduction (pMOR) is a powerful tool for accelerating finite element (FE) simulations while maintaining parametric dependencies. For geometric parameters, pMOR by matrix interpolation is a well-suited approach because it does not require an affine representation of the parametric dependency, which is often not available for geometric pa

Mikel Williams-Lekuona, Georgina Cosma

Vision transformers in vision-language models typically use the same amount of compute for every image, regardless of whether it is simple or complex. We propose ICAR (Image Complexity-Aware Retrieval), an adaptive computation approach that enables vision transformers to use less compute for simple images whilst processing complex images through their full n

Leyang Xue, Kai-Cheng Yang, Peng-Bi Cui, Zengru Di

The spread of ideas, behaviors, and technologies generally depends on feedback mechanisms operating across multiple scales. Previous studies have extensively examined pairwise transmission and local reinforcement. However, the role of macro-level social influence -- where widespread adoption enhances further adoption -- remains understudied. Here, we focus o

Tzveta Apostolova, Balint Kiss, Deborata Rajak, Katalin Pirisi

We theoretically and experimentally investigate orientation-dependent high-harmonic generation (HHG) in zinc oxide subjected to intense femtosecond mid-infrared (MIR) laser pulses. In agreement with past measurements from literature, we observe non-perturbative harmonic spectra with harmonics extending beyond the material band gap. The spectra depend sensiti

Maximilian Kellner, Mariana Ferrandon Cervantes, Yuandong Pan, Ruodan Lu

We propose a novel dataset that has been specifically designed for 3D semantic segmentation of bridges and the domain gap analysis caused by varying sensors. This addresses a critical need in the field of infrastructure inspection and maintenance, which is essential for modern society. The dataset comprises high-resolution 3D scans of a diverse range of brid

Ursula Mello, Martin Nybom, Jan Stuhler

Lacking lifetime income data, most intergenerational mobility estimates are subject to lifecycle bias. Using long income series from Sweden and the US, we illustrate that standard correction methods struggle to account for one important property of income processes: children from affluent families experience faster income growth, even conditional on their ow

Benoit Cadorel, Ya Deng, Katsutoshi Yamanoi

This is Part II of a series of three papers. We studies the hyperbolicity of complex quasi-projective varieties $X$ in the presence of a big and reductive representation $\varrho: \pi_1(X)\to {\rm GL}_N(\mathbb{C})$. For any Galois conjugate variety $X^\sigma$ with $\sigma \in {\rm Aut}(\mathbb{C}/\mathbb{Q})$, we prove the generalized Green-Griffiths-Lang c

Ángela Tajuelo-Valbuena, Jara Trujillo-Mulero, Juan J. Meléndez, Rodolfo Cuerno

We have simulated an automaton version of the quenched Kardar-Parisi-Zhang (qKPZ) equation in one and two dimensions in order to study the scaling properties of the interface at the depinning transition. Specifically, the $\alpha$, $\beta$, $\theta$, and $\delta$ critical exponents characterizing the surface kinetic roughening and depinning behaviors have be

Gajendra Doniparthi, Shashank Balu Pandhare, Stefan Deßloch, Timo Mühlhaus

Traditional search applications within Research Data Management (RDM) ecosystems are crucial in helping users discover and explore the structured metadata from the research datasets. Typically, text search engines require users to submit keyword-based queries rather than using natural language. However, using Large Language Models (LLMs) trained on domain-sp

Oren Becker, Emmanuel Breuillard

We show that random walks on semisimple algebraic groups do not concentrate on proper algebraic subvarieties with uniform exponential rate of anti-concentration. This is achieved by proving a uniform spectral gap for quasi-regular representations of countable linear groups. The method makes key use of Diophantine heights and the Height Gap theorem. We also d

Adrian Straker, Paul Magdon, Marco Zullich, Maximilian Freudenberg

Aiming to advance research in the field of interpretability of deep learning models for tree species classification using TLS 3D point clouds we present insights in the classification abilities of YOLOv8 through a new framework which enables systematic analysis of saliency maps derived from CAM (Class Activation Mapping). To investigate the contribution of s

Zixin Wei, Yucan Guo, Jinyang Li, Xiaolin Han

The search for suitable datasets is the critical "first step" in data-driven research, but it remains a great challenge. Researchers often need to search for datasets based on high-level task descriptions. However, existing search systems struggle with this task due to ambiguous user intent, task-to-dataset mapping and benchmark gaps, and entity ambiguity. T

Chunhao Cai

We consider estimation of the drift parameter $\vartheta>0$ in a \emph{partially observed} Ornstein--Uhlenbeck type model driven by a mixed fractional Brownian noise. Our framework extends the partially observed model of \cite{BrousteKleptsyna2010} to the \emph{mixed} case. We construct the canonical innovation representation, derive the associated Kalman fi

Pedro Garcia-Gomez, Paula Guerrero-Lopez, Silvia Hervas-Raluy, Jose Manuel Garcia-Aznar

Cancer research has shifted from a purely gene-centric view to a more holistic understanding that recognizes the critical role of the tumour microenvironment, where mechanics and metabolism are key drivers of disease progression. However, the intricate interplay between these multifactorial mechanisms remains poorly understood. To address this gap, we presen

K. Shih, D. Endrizzi, D. A. Sutherland, J. Anderson

Absolute extreme ultraviolet (AXUV) diode arrays are widely used in fusion experiments for time-resolved measurements of plasma radiation. We report the first implementation of an AXUV-based analysis framework on the Wisconsin High-Temperature Superconducting (HTS) Axisymmetric Mirror (WHAM). A single, precisely calibrated 20-channel AXUV assembly measures l

Carina Zell-Ziegler, Célia Burghardt, Kaya Dünzen, David Schöpf

The sixth assessment report of the Intergovernmental Panel on Climate Change (IPCC) emphasises the potential of demand-side measures, such as sufficiency, for mitigating climate change. Although quantified potentials of various sufficiency measures exist, policy advisors and energy modellers criticise the lack of findability and comparability of the relevant

Zhengyi Zhao, Shubo Zhang, Yuxi Zhang, Huimin Wang

Large Language Models (LLMs) have shown impressive capabilities in complex reasoning tasks. However, current approaches employ uniform language density for both intermediate reasoning and final answers, leading to computational inefficiency. Our observation found that reasoning process serves a computational function for the model itself, while answering ser

Xiaozhe Xi, Junzhe Cao, Kun Xu

Coherent Rayleigh-Brillouin scattering (CRBS) holds great promise for the characterization of gas properties and the investigation of gas kinetic processes. The CRBS spectrum exhibits a strong dependence on the Knudsen number (Kn), revealing its inherently multiscale nature. In the unified gas-kinetic scheme (UGKS), collisions are intrinsically coupled with

István Miklós, Miklós Ruszinkó, Bogdán Zavalnij

We prove a complete dichotomy theorem for the parameterized sparse $t$-uniform hypergraphic degree sequence problem, $\mathrm{sparse}\text{-}t\text{-}\mathrm{uni}\text{-}\mathrm{HDS}_{\alpha',\alpha}$. For any fixed $t \ge 3$, given parameters $0 \le \alpha' \le \alpha < t-1$, the input consists of degree sequences $D$ of length $n$ with degrees between $n^{

Andreas Buchinger, Christian Seifert, Sascha Trostorff, Marcus Waurick

We consider evolutionary equations as introduced by R.\ Picard in 2009 and develop a general theory for approximation which can be seen as a theoretical foundation for numerical analysis for evolutionary equations. To demonstrate the approximation result, we apply it to a spatial discretisation of the heat equation using spectral methods.

Joao Queiroz

Recent evidence shows that the versification of prompts constitutes a highly effective adversarial mechanism against aligned LLMs. The study 'Adversarial poetry as a universal single-turn jailbreak mechanism in large language models' demonstrates that instructions routinely refused in prose become executable when rewritten as verse, producing up to 18 x more

Clément Cornet, Romaric Besançon, Hervé Le Borgne

We propose an alternative to sparse autoencoders (SAEs) as a simple and effective unsupervised method for extracting interpretable concepts from neural networks. The core idea is to cluster differences in activations, which we formally justify within a discriminant analysis framework. To enhance the diversity of extracted concepts, we refine the approach by

Rhea Alexander, Michalis Skotiniotis, Daniel Manzano

The detection and characterization of quantum coherence is of fundamental importance both in the foundations of quantum theory as well as for the rapidly developing field of quantum technologies, where coherence has been linked to quantum advantage. Typical approaches for detecting coherence employ {\it coherence witnesses} -- observable quantities whose exp

Medet Jumadildayev

The duality theorem of Lass relates the matching polynomials of a simple graph $G$ with the matching polynomials of its complement $\bar G$. In particular, this relation gives rise to Godsil's result, which offers a nice interpretation of the Lebesgue-Stieltjes integral associated with the Hermite orthogonality measure. In this work, we introduce the concept

Hanzhang Yin

In this paper, we establish a priori estimates and existence results for solutions of a general class of fully non-linear equations on noncompact K\"{a}hler and Hermitian manifolds. As geometric applications, we construct complete K\"{a}hler metrics with prescribed volume forms on strictly pseudoconvex domains, as well as find Einstein metrics on complete no

Nan-Hong Kuo, Renata Wong

We propose a quantum analogue of Bluestein's algorithm (QBA) that implements an exact $N$-point Quantum Fourier Transform (QFT) for arbitrary $N$. Our construction factors the $N$-dimensional QFT unitary into three diagonal quadratic-phase gates and two standard radix-2 QFT subcircuits of size $M = 2^m$ (with $M \ge 2N - 1$). This achieves asymptotic gate co

M. V. Kondrin, Y. B. Lebed, V. V. Brazhkin

The hard sphere model is widely used in description of fluids and solid media as a zero approximation to real systems. Despite the uniqueness of the model, few analytical results are known for it, both for the 2D and 3D cases. In present research we have investigated melting of the hard disk system by considering accumulation of extended defects of a certain

Shiran Ge, Chenyi Huang, Yuang Ai, Qihang Fan

Group Relative Policy Optimization (GRPO) is a powerful technique for aligning generative models, but its effectiveness is bottlenecked by the conflict between large group sizes and prohibitive computational costs. In this work, we investigate the trade-off through empirical studies, yielding two key observations. First, we discover the reward clustering phe

N. J. Lopez-Alamilla, A. L. Nord, F. Pedaci, J. Palmeri

The bacterial flagellar motor (BFM) converts transmembrane ion flux into directed mechanical rotation, driving bacterial motility. Despite extensive study, the frictional forces and energetics governing its torque generation remain poorly understood. Here, we combine single-molecule rotation measurements with stochastic thermodynamics to quantitatively estim

V. A. Checha, A. Aret, I. Kolka, T. Liimets

Context. The post-main-sequence evolution of massive stars remains poorly understood, particularly for blue supergiants. These objects play a crucial role in the dynamical and chemical evolution of galaxies and exhibit pronounced photometric and spectroscopic variability, often quasi-periodic rather than strictly periodic. Aims. We investigate the variabilit

Hiroyoshi Nagahama, Katsufumi Inoue, Masayoshi Todorokihara, Michifumi Yoshioka

Predictive maintenance of rotating machinery increasingly relies on vibration signals, yet most learning-based approaches either discard phase during spectral feature extraction or use raw time-waveforms without explicitly leveraging phase information. This paper introduces two phase-aware preprocessing strategies to address random phase variations in multi-

Efe Bozkir, Enkelejda Kasneci

The rapid development of generative artificial intelligence (AI) and large language models (LLMs), and the availability of services that make them accessible, have led the general public to begin incorporating them into everyday life. The extended reality (XR) community has also sought to integrate LLMs, particularly in the form of conversational agents, to

Zhentian Zhang, Jian Dang, David Morales-Jimenez, Hao Jiang

The fluid antenna system (FAS) refers to a family of reconfigurable antenna technologies that provide substantial spatial gains within a compact, predefined small space, thereby offering extensive degrees of freedom in the physical layer for future communication networks. The acquisition of channel state information (CSI) is critical, as it determines the pl

Gregorio Curello, Ludvig Sinander, Mark Whitmeyer

In finite problems comprising objects, cases, and an object- and case-contingent payoff function, we study the comparative statics of the set of undominated objects, meaning those for which there exists no mixture over objects that is superior in every case. We consider both weak and strict dominance (corresponding to different degrees of &#39;strictness&#39

Junjie Chen, Fei Wang, Zhihao Huang, Qing Zhou

Human conversation involves continuous exchanges of speech and nonverbal cues such as head nods, gaze shifts, and facial expressions that convey attention and emotion. Modeling these bidirectional dynamics in 3D is essential for building expressive avatars and interactive robots. However, existing frameworks often treat talking and listening as independent p

Peili Mao, Joseph Billingsley, Wang Miao, Geyong Mi

Modern data centers contain thousands of servers making them major consumers of electricity. To minimize their environmental impact, it is critical that we use their resources efficiently. In this paper we study how to discover the optimal placement of virtual network functions in large scale data centers. We propose a novel parallel metaheuristic, fast heur

Chloe E. Fisher, Matthew J. Hooton, Amélie Gressier, Merlin Zgraggen

Determining the prevalence of atmospheres on terrestrial planets is a core objective in exoplanetary science. While M dwarf systems offer a promising opportunity, conclusive observations of terrestrial atmospheres have remained elusive, with many yielding flat transmission spectra. We observe four transits of the hot terrestrial planet TOI-1685 b using JWST'

Pietro Metuh, Paweł Wyborski, Athanasios Paralikis, Frederik Schröder

Scalable photonic quantum information technologies require a platform combining quantum light sources, waveguides, and detectors on a single chip. Here, we introduce a van der Waals platform comprising strain-engineered bilayer WSe$_2$ quantum emitters, integrated on multimode WS$_2$ waveguides with optimized grating couplers, enabling efficient on-chip quan

Xingwu Chen, Jiahao Li, Tao Li

In this paper we investigate the crossing-sliding bifurcations of planar Filippov systems with $\mathbb{Z}_2$-symmetry. Such bifurcations are triggered by the perturbations of a critical crossing cycle and constitute an important class of discontinuity-induced bifurcations. By constructing transition maps and developing a decomposition theorem of functions t

Chenxiang Zhang, Tongxi Qu, Zhong Li, Tian Zhang

Deep neural networks are widely deployed with quantization techniques to reduce memory and computational costs by lowering the numerical precision of their parameters. While quantization alters model parameters and their outputs, existing privacy analyses primarily focus on full-precision models, leaving a gap in understanding how bit-width reduction can aff

Xiao-Jie Zhu

We provide explicit formulas for quadratic Gauss sums over $\mathbb{Z}^n/c\mathbb{Z}^n$, which generalize some of the existing formulas, e.g., Skoruppa and Zagier's (for $n=2$), and Iwaniec and Kowalski's (for arbitrary $n$). We then give four main applications. As the first application, we prove a duality theorem, which relates a sum over a subgroup of $\ma

Nishant Gaurav, Adit Akarsh, Tejas Ravishankar, Manoj Bajaj

Current tool-using AI agents suffer from limited action space, context inefficiency, and probabilistic instability that makes them unsuitable for handling repetitive tasks which are otherwise reliably and efficiently tackled by agentic workflows built on platforms like n8n and Zapier. Earlier works like CodeAct, DynaSaur, Code Mode have tried to tackle the f

Alon Beck, Moshe Goldstein

This paper investigates wave-packet dynamics in non-Hermitian lattice systems and reveals a surprising phenomenon: The simultaneous propagation of two distinct wavefronts, one traveling at the non-Hermitian velocity and the other at the Hermitian velocity. We show that this dual-front behavior arises naturally in systems governed by a pseudo-Hermitian Hamilt

Julio Careaga, Qian Huang, Julian Koellermeier

In this work, we develop a modelling framework for granular flows based on the shallow water moment equations on inclined planes. Under the assumption of a polynomial expansion of the velocity field, the model extends the classical shallow water equations to vertically variable velocity profiles. The friction effects, which are captured through the strain-ra

Wei Li, Zhenyu Bai, Heru Wang, Pranav Dangi

Spatial dataflow accelerators are a promising direction for next-generation computer systems because they can reduce the memory bottlenecks of traditional von Neumann machines such as CPUs and GPUs. They organize computation around explicit, compiler-managed data movement over on-chip networks, allowing operands to be forwarded directly between processing el

Fei Zhao, Mengxi Guo, Shijie Zhao, Junlin Li

There has been a growing trend in compressing and transmitting videos from terminals for machine vision tasks. Nevertheless, most video coding optimization method focus on minimizing distortion according to human perceptual metrics, overlooking the heightened demands posed by machine vision systems. In this paper, we propose a video preprocessing framework t

Paul Bagourd, Julian Jang-Jaccard, Vincent Lenders, Alain Mermoud

Quantum computers pose a fundamental threat to widely deployed public-key cryptosystems, such as RSA and ECC, by enabling efficient integer factorization using Shor's algorithm. Theoretical resource estimates suggest that 2048-bit RSA keys could be broken using Shor's algorithm with fewer than a million noisy qubits. Although such machines do not yet exist,

Juliane Krautz

Starting from pointwise gradient estimates for the heat semigroup, we study three characterizations of weak lower curvature bounds on metric graphs. More precisely, we prove the equivalence between a weak notion of the Bakry-\'Emery curvature condition, a weak Evolutionary Variational Inequality and a weak form of geodesic convexity. The proof is based on a

Ayman Eddakrouri, Amani Ramadan

This study applies Brookes' Measure of Categorical Dispersion ({\Delta}) to analyze the thematic structure of contemporary Arabic Applied Linguistics research. Using a comprehensive, real-world dataset of 1,564 publications from 2019 to 2025, classified into eight core sub-disciplines, we calculate a dispersion index of {\Delta} = 0.194. This remarkably low

Parvesh Saini, Soumyadipta Maiti, Beena Rai

Capabilities and the number of vision-based models are increasing rapidly. And these vision models are now able to do more tasks like object detection, image classification, instance segmentation etc. with great accuracy. But models which can take accurate quantitative measurements form an image, as a human can do by just looking at it, are rare. For a robot

Yuxin Jiang, Yunkang Can, Weiming Shen

Knowledge distillation is an effective image anomaly detection and localization scheme. However, a major drawback of this scheme is its tendency to overly generalize, primarily due to the similarities between input and supervisory signals. In order to address this issue, this paper introduces a novel technique called masked reverse knowledge distillation (MR

Agnieszka Bienkowska, Jacek Malecki, Alexander Mathiesen-Ohman, Katarzyna Tworek

Organizations increasingly operate in environments characterized by volatility, uncertainty, complexity, and ambiguity (VUCA), where early indicators of change often emerge as weak, fragmented signals. Although artificial intelligence (AI) is widely used to support managerial decision-making, most AI-based systems remain optimized for prediction and resoluti

Amit Hogadi, Anand Sawant

We show that the sheaf of $\mathbb A^1$-connected components of a quasi-split group over a perfect field is a strictly $\mathbb A^1$-invariant sheaf with (Voevodsky) transfers. As a consequence, we show that the norm principle holds for any quasi-split group over a perfect field.

Malihe Dahmardeh, Francesco Setti

In this paper we propose MECAD, a novel approach for continual anomaly detection using a multi-expert architecture. Our system dynamically assigns experts to object classes based on feature similarity and employs efficient memory management to preserve the knowledge of previously seen classes. By leveraging an optimized coreset selection and a specialized re

Ming Lu, Zhuoyi Zhao

A quantum symmetric pair consists of a quantum group $\widetilde{\mathbf{U}}$ and its coideal subalgebra $\widetilde{\mathbf{U}}^\imath$. The Hall algebra constructions of $\widetilde{\mathbf{U}}$ and $\widetilde{\mathbf{U}}^\imath$ are given by Bridgeland and Lu--Wang, respectively. In this paper, we construct a Hall algebra framework for the coideal subalg

Stanislav Dubnička, Anna Zuzana Dubničková, Mikhail A. Ivanov, Andrej Liptaj

We present a study of various $B_{s}$ meson decays, including hadronic and semileptonic final states with different spins and diagram topologies. The covariant confined quark model is employed to describe hadronic effects, and our analysis serves as a broad test of the current understanding of the underlying dynamics within the Standard Model. The level of a

Nazerke T. Tleukhanova, Makhpal Manarbek

The main aim of this paper is to obtain Bochkarev-type inequalities for the anisotropic grand Lorentz spaces. In the classical setting, Bochkarev obtained inequalities of the Hardy--Littlewood type, which reveal the connection between the integral properties of functions and the summability of their Fourier coefficients. His results describe the behavior of

Yuxin Jiang, Yunkang Cao, Weiming Shen

Few-shot anomaly detection (FSAD) denotes the identification of anomalies within a target category with a limited number of normal samples. Existing FSAD methods largely rely on pre-trained feature representations to detect anomalies, but the inherent domain gap between pre-trained representations and target FSAD scenarios is often overlooked. This study pro

Jan Schwientek, Katrin Teichert, Jan Schröder, Johannes Höller

Model-based process simulation can be used to derive designs and operating conditions of chemical processes that optimally balance multiple objectives, such as quality, costs, or environmental impacts. This work focuses on identifying designs that hedge against uncertainties in model parameters to ensure feasibility, taking the possibility to adjust operatin

M. Buhmann, J. Jódar, M. Rodríguez

In a previous paper we have introduced a new class of radial basis functions that are powerful means to approximate functions by quasi-interpolation. In this article we extend the results to create new ways of approximating functions by prewavelets that are constructed from spaced spanned of the new hyperbolic radial basis functions. They consist of highly l

Francesco Xotta, Nina Bavdaž, Christopher Eckner, Dmitry Malyshev

In 2010, the Fermi Gamma-ray Space Telescope observed two gamma-ray emitting structures, the Fermi Bubbles (FBs), that extend up to 55{\deg} above and below the Galactic plane and that seem to emanate from the Galactic center region. Although the spectrum at latitudes |b| > 10{\deg} has a softening or a cutoff around 100 GeV, the one at the base of the FBs,

Antony Jerald, Dattesh Shanbhag, Sudhanya Chatterjee

Motion artifacts degrade MRI image quality and increase patient recalls. Existing automated quality assessment methods are largely limited to binary decisions and provide little interpretability. We introduce AutoMAC-MRI, an explainable framework for grading motion artifacts across heterogeneous MR contrasts and orientations. The approach uses supervised con

Feiyi Liao, Haochen Liu, Hehu Xie

With the regular decomposition technique, we decompose the space $\mathbf{H}_0^s(\mathbf{curl}; \Omega)$ into the sum of a vector potential space and the gradient of a scalar space, both possessing higher regularity. Based on this new high order regular decomposition, a novel numerical method using standard high order Lagrange finite elements is designed for

Charan Prakash Rathore, Saumi Ray, Dhruv Kumar

Extracting structured information from zeolite synthesis experimental procedures is critical for materials discovery, yet existing methods have not systematically evaluated Large Language Models (LLMs) for this domain-specific task. This work addresses a fundamental question: what is the efficacy of different prompting strategies when applying LLMs to scient

Wenke E, Yixin Sun, Jiaxu Liu, Hubert P. H. Shum

We present the first cross-modality distillation framework specifically tailored for single-panoramic-camera Bird's-Eye-View (BEV) segmentation. Our approach leverages a novel LiDAR image representation fused from range, intensity and ambient channels, together with a voxel-aligned view transformer that preserves spatial fidelity while enabling efficient BEV

Wangyu Wu, Zhenhong Chen, Xiaowei Huang, Fei Ma

Weakly Supervised Semantic Segmentation (WSSS) with image level labels aims to produce pixel level predictions without requiring dense annotations. While recent approaches have leveraged generative models to augment existing data, they remain dependent on real world training samples. In this paper, we introduce a novel direction, Zero Shot Weakly Supervised

Kai Zhang, Shoubin Chen, Dong Li, Baiyang Zhang

Autonomous exploration is a fundamental capability that tightly integrates perception, planning, control, and motion execution. It plays a critical role in a wide range of applications, including indoor target search, mapping of extreme environments, resource exploration, etc. Despite significant progress in individual components, a holistic and practical de

Basil Ell

We introduce graph pattern-based association rules (GPARs) for directed labeled multigraphs such as RDF graphs. GPARs support both generative tasks, where a graph is extended, and evaluative tasks, where the plausibility of a graph is assessed. The framework goes beyond related formalisms such as graph functional dependencies, graph entity dependencies, rela

Roberto de A. Capistrano Filho, Hugo Parada, Jandeilson Santos da Silva

This paper investigates a boundary-value problem for the Korteweg-de Vries (KdV) equation on a star-graph structure. We develop a unified framework introducing the notion of $s$-compatibility, which generalizes classical compatibility conditions to star-shaped and more complex graph configurations, inspired by the works of Bona, Sun, and Zhang [14]. By combi

Erik Schultheis, Dan Alistarh

We present LLMQ, an end-to-end CUDA/C++ implementation for medium-sized language-model training, e.g. 3B to 32B parameters, on affordable, commodity GPUs. These devices are characterized by low memory availability and slow communication compared to datacentre-grade GPUs. Consequently, we showcase a range of optimizations that target these bottlenecks, includ

M. Bergemann, G. Kordopatis, G. Casali, S. Khoperskov

The formation and evolution of the Milky Way's disc, bar, and bulge remain fundamentally limited by the lack of a contiguous, Galaxy-wide, high-precision chemo-dynamical map. Key open questions - including the survival or destruction of the primitive discs, the origin of the bulge's multi-component structure, the role of mergers and secular processes, and th

Laurent Navoret, Roxana Sublet, Marcela Szopos

The goal of the present work is to propose an agent-based model that originally combines classical Vicsek-like polarity alignments and contact forces, as implemented in the framework developed by Maury and Venel in [Maury, Venel, 2011]. The description additionally incorporates velocity feedback on polarity and soft attraction-repulsion interactions. After c

Axel J. M. Deenen, Dirk Grundler

The superconducting diode effect (SDE) is a key nonreciprocal phenomenon with broad relevance for superconducting electronics. Using time-dependent Ginzburg-Landau simulations, we predict and quantify a superconducting diode effect arising solely from geometric chirality imposed to a conventional superconductor. The helical geometry and magnetic-field-induce

Elena Patyukova, Junwen Yin, Susmita Basak, Samuel Pinilla Sanchez

Performing density functional theory (DFT) calculations requires a careful choice of computational parameters to ensure convergence and obtain meaningful results. This represents a particularly important problem for high-throughput and agentic workflows, where due to computational cost, any additional convergence studies are preferably to be avoided. So, the

Xiaotian Zhang, Yuan Wang, Ruizhe Chen, Zeya Wang

The deployment of Large Language Models (LLMs) in interactive systems necessitates a deep alignment with the nuanced and dynamic preferences of individual users. Current alignment techniques predominantly address universal human values or static, single-turn preferences, thereby failing to address the critical needs of long-term personalization and the initi

Yanhao Sun, Jiayu Ma, Xiangyu Wang, Song Yu

Continuous-variable quantum key distribution (CV-QKD) enables two remote parties to establish information-theoretically secure keys and offers high practical feasibility due to its compatibility with mature coherent optical communication technologies. However, as CV-QKD systems progress toward digital implementations, device nonidealities drive the optical f

Shiki Ogata, Shunsaku Kitagawa, Kenji Ishida, Manuel Brando

Superconductor LaRh$_2$As$_2$ has the same crystal structures as CeRh$_2$As$_2$, which exhibits superconducting (SC) multiphase in the $c$-axis magnetic field. Although the SC transition temperatures $T_c$ are similar, around 0.3 K, LaRh$_2$As$_2$ shows conventional type-II superconductivity with a small upper critical field $H_{c2}\sim$ 10 mT. At present, t

Seok-Hyun Ga, Chun-Yen Chang

The rapid development of Generative AI is bringing innovative changes to education and assessment. As the prevalence of students utilizing AI for assignments increases, concerns regarding academic integrity and the validity of assessments are growing. This study utilizes the Earth Science I section of the 2025 Korean College Scholastic Ability Test (CSAT) to

Yue-Hong Wu, Ning-Hua Tong

In this paper, we investigate the decoherence of qubit due to its coupling to a Hermitian or a non-Hermitian bath within the pure dephasing spin-boson model. First, using this model, we analytically establish the previously anticipated similarity between the non-equilibrium and the equilibrium correlation functions $P_x(t)$ and $C_x(t)$. Then, in the short/l

Claudia Ceci, Luca Semerari

We analyze the problem of optimal reduction of the debt-to-GDP ratio in a stochastic control setting. The debt-to-GDP dynamics are modeled through a stochastic differential equation in which fiscal policy simultaneously affects both debt accumulation and GDP growth. A key feature of the framework is the introduction of a cost functional that captures the dis

Toshihide Ubukata, Enhong Mu, Takuto Yamauchi, Mingyue Zhang

Controller synthesis is a formal method approach for automatically generating Labeled Transition System (LTS) controllers that satisfy specified properties. The efficiency of the synthesis process, however, is critically dependent on exploration policies. These policies often rely on fixed rules or strategies learned through reinforcement learning (RL) that

M. Guidi, M. Moresco, H. K. Herrera-Alcantar, G. Aricò

Stage IV galaxy surveys (DESI, 4MOST, MOONS, Euclid) are establishing precision constraints on cosmological parameters through baryon acoustic oscillations and redshift-space distortions, yet fundamental questions on neutrino masses, inflationary physics, and the nature of gravity remain beyond their reach. We present a science case for next-generation wide-

Riku Shintani

About a century ago, P. A. MacMahon introduced a class of $q$-series, which are nowadays referred to as MacMahon series. More recently, in 2013, G. E. Andrews and S. C. F. Rose revealed the quasimodular property of these series. In this paper, we introduce a generalization of MacMahon series. Specifically, for any positive integers $t, k, N$ and a polynomial

Shreyas Tiruvaskar, Russell Boey, Richard Easther, Chris Gordon

We investigate the impact of ultralight dark matter (ULDM) on the mergers of supermassive black holes (SMBH) and the resulting stochastic gravitational wave background. ULDM is based on exceptionally light particles and yields galactic halos with dense central solitons. This increases the drag experienced by binary SMBH, decreasing merger times and potential

Prasanta Malik, Anirban Paul

In this paper, we introduce the notions of I and I*-soft convergence of sequences of soft points in soft topological spaces and study some basic properties of these notions. Also we introduce the notions of I-soft limit points and I-soft cluster points of a sequence of soft points in a soft topological space and study their interrelationship

Jie Zhou, Junhao Xie

This paper addresses two critical limitations in diagonally loaded (DL) adaptive matched filter (AMF) detector: (1) the lack of CFAR property with respect to arbitrary covariance matrices, and (2) the absence of selection criteria for optimal loading factor from the perspective of maximizing the detection probability (Pd). We provide solutions to both challe

C. Pantouvakis, M. Rignanese, T. Zenger, S. Ciarlantini

Monolithic Active Pixel Sensors (MAPS) achieved widespread use in several scientific applications, thanks to their properties, such as low material budget and high granularity. The ARCADIA INFN project developed a Fully-Depleted MAPS (FD-MAPS), using a modified LFoundry 110 nm CIS process. This work presents the first laboratory characterization of the ARCAD

M. A. Reiter, E. O. Lazarev, D. A. Glazov, A. V. Malyshev

The energy correction associated with the self-energy diagram is the leading (in magnitude) and fundamental (in significance) contribution to the Lamb shift in highly charged ions. Conventional approaches to this correction rely on partial-wave expansions, which is a stumbling block limiting accuracy. To elucidate the issue, we perform a comprehensive compar

Emmanuel Fricain, Muath Karaki, Javad Mashreghi, Maëva Ostermann

In this paper, we characterize the boundedness and the compactness of weighted composition operators acting on a de Branges-Rovnyak space $\mathcal H(b)$, where the symbol $b$ is a rational function in the unit ball of $H^\infty$ that is not a finite Blaschke product. Our results extend those of [2] by exploiting a close relationship between weighted composi

Siva Sai, Ishika Goyal, Shubham Sharma, Sri Harshita Manuri

The increasing number of cyber threats and rapidly evolving tactics, as well as the high volume of data in recent years, have caused classical machine learning, rules, and signature-based defence strategies to fail, rendering them unable to keep up. An alternative, Quantum Machine Learning (QML), has recently emerged, making use of computations based on quan

Aleksei Shestov, Anton Klenitskiy, Daria Denisova, Amurkhan Dzagkoev

Modern representation learning increasingly relies on unsupervised and self-supervised methods trained on large-scale unlabeled data. While these approaches achieve impressive generalization across tasks and domains, evaluating embedding quality without labels remains an open challenge. In this work, we propose Persistence, a topology-aware metric based on p

Yoav Green

No. Eighty years ago, the two seminal works by Goldman [J. Gen. Phys. 27, 37 (1943)] and by Hodgkin-Katz [J. of Physio 108, 37 (1949)] derived the foundational framework for interpreting electro-physiological measurements in what is commonly termed the Goldman-Hodgkin-Katz (GHK) theory for the membrane potential. Both seminal papers postulate a constant/unif

Colin Cros, Laurent Ferro-Famil

Direction of Arrival (DOA) estimation is a fundamental problem in signal processing. Diffuse sources, whose power density cannot be represented with a single angular coordinate, are usually characterized based on prior assumptions, which associate the source angular density with a specific set of functions. However, these assumptions can lead to significant

Martijn IJtsma, Salvatore Hargis

Studies of human-robot interaction in dynamic and unstructured environments show that as more advanced robotic capabilities are deployed, the need for cooperative competencies to support collaboration with human problem-holders increases. Designing human-robot systems to meet these demands requires an explicit understanding of the work functions and constrai

Feifei Zhang, Zhenhong Jia, Sensen Song, Fei Shi

Medical image segmentation is constrained by sparse pathological annotations. Existing augmentation strategies, from conventional transforms to random masking for self-supervision, are feature-agnostic: they often corrupt critical diagnostic semantics or fail to prioritize essential features. We introduce "Keep the Core," a novel data-centric paradigm that u