November 2025 arXiv papers — page 121

Jian Gao, Richeng Xuan, Zhaolu Kang, Dingshi Liao

The rapid advancement of large language models (LLMs) has not been matched by their evaluation in low-resource languages, especially Southeast Asian languages like Lao. To fill this gap, we introduce \textbf{LaoBench}, the first large-scale, high-quality, and multidimensional benchmark for assessing LLM language understanding and reasoning in Lao. LaoBench c

Teresa Cortadellas Benitez, Carlos D'Andrea, Ana Belen de Felipe, Joel Hurtado Moreno

With the aid of Hensel Lemma, we refine the 2-adic Newton polygon algorithm proposed by Magron, Koprowski, and Vaccon at ISSAC 2023 to express computationally a given positive univariate polynomial with rational coefficients as a sum of five squares of rational polynomials -the effective Pourchet's Theorem- and extend it to cover almost all the possible inpu

Merlin Füllgraf, Jochen Gemmer, Jiaozi Wang

The Eigenstate Thermalization Hypothesis (ETH) has been established as a cornerstone for understanding thermalization in quantum many-body systems. Recently, there has been growing interest in the full ETH, which extends the framework of the conventional ETH and postulates a smooth function to describe the multi-point correlations among matrix elements. With

E. Lima, L. L. Braseth, A. H. Mjøs, M. Hjorth-Jensen

Reconstructing gamma-ray spectra from detector measurements is an ill-posed inverse problem. Standard methods, such as Folding Iteration with Compton Subtraction (FICS), provide point estimates but lack calibrated uncertainties and may bias the spectrum. We introduce an unfolding framework based on regularized maximum-likelihood estimation (RMLE) that enforc

Christopher Mitchell, Tracey Barnett, Meredith Newton, Erika Thompson

Novel oral nicotine products, particularly nicotine pouches, have rapidly gained popularity among adolescents. Among U.S. high school students, nicotine pouch use has doubled since 2021, with 2.4% reporting current use in 2024. We analyzed Florida Youth Tobacco Survey data from 2022-2024 to assess prevalence trends and developed a grade-structured compartmen

Sascha Desmettre, Agnes Mallinger, Amira Meddah, Irene Tubikanec

Piecewise diffusion Markov processes (PDifMPs) form a versatile class of stochastic hybrid systems that combine continuous diffusion processes with discrete event-driven dynamics, enabling flexible modelling of complex real-world hybrid phenomena. The practical utility of PDifMP models, however, depends critically on accurate estimation of their underlying p

Chaoyun Zhang, Liqun Li, He Huang, Chiming Ni

Large language model (LLM)-powered agents are transforming digital devices from passive tools into proactive intelligent collaborators. However, most existing frameworks remain confined to a single OS or device, making cross-device workflows brittle and largely manual. We present UFO$^3$, a system that unifies heterogeneous endpoints, desktops, servers, mobi

Shuai Su, Xiurong Yan, Qian Zhang

In this paper, we develop a novel enriched Galerkin (EG) method for the steady incompressible Navier-Stokes equations in rotational form, which is both pressure-robust and parameter-free. The EG space employed here, originally proposed in [1], differs from traditional EG methods: it enriches the first-order continuous Galerkin (CG) space with piecewise const

Vlad Rakhlin, Amir Jevnisek, Shai Avidan

ReLU activations are the main bottleneck in Private Inference that is based on ResNet networks. This is because they incur significant inference latency. Reducing ReLU count is a discrete optimization problem, and there are two common ways to approach it. Most current state-of-the-art methods are based on a smooth approximation that jointly optimizes network

Max Goldberg, Antoine C. Petit

Planet formation simulations consistently predict compact systems of numerous small planets in chains of mean motion resonances formed by planet-disk interaction, but transiting planet surveys have found most systems to be non-resonant and somewhat dynamically excited. A scenario in which nearly all of the primordial resonant chains undergo dynamical instabi

Masahiro Ogihara, Masanobu Kunitomo

Recent observations have revealed the distribution of orbital period ratios of adjacent planets in multiple super-Earth systems and how these distributions change with time. The aim of this study is to clarify under what conditions the observed features of orbital period ratios of super-Earths can be explained, and to identify what causes the dynamical insta

Jon Miles

We describe how to glue prime-to-$p$ torsion sheaves along Harder-Narasimhan strata of Fargues-Scholze's $\mathrm{Bun}_G$ in terms of the cohomology of locally closed strata inside the smooth charts $\mathcal{M}_b \to \mathrm{Bun}_G$ constructed in [FS21], which are moduli of certain split parabolic bundles. Our computations for $G=\operatorname{GL}_2$ expli

Anuj Dawar, Lauri Hella, Benedikt Pago

We investigate the expressive power of generalized quantifiers closed under partial polymorphism conditions motivated by the study of constraint satisfaction problems. We answer a number of questions arising from the work of Dawar and Hella (CSL 2024) where such quantifiers were introduced. For quantifiers closed under partial near-unanimity polymorphisms, w

Tobias Nadolny, Christoph Bruder

Limit-cycle oscillators are the basic building blocks for synchronization; yet, the notion of a quantum limit cycle has remained unclear. Here, we study quantum limit cycles and synchronization in the presence of continuous heterodyne measurement. The resulting quantum trajectories, i.e., time evolutions of the quantum state conditioned on the measurement ou

Anurag J. Vaidya, Felix Meissen, Daniel C. Castro, Shruthi Bannur

Digitized histopathology analysis involves complex, time-intensive workflows and specialized expertise, limiting its accessibility. We introduce NOVA, an agentic framework that translates scientific queries into executable analysis pipelines by iteratively generating and running Python code. NOVA integrates 49 domain-specific tools (e.g., nuclei segmentation

Yitian Kou, Yihe Gu, Chen Zhou, DanDan Zhu

Navigating human-populated environments without causing discomfort is a critical capability for socially-aware agents. While rule-based approaches offer interpretability through predefined psychological principles, they often lack generalizability and flexibility. Conversely, data-driven methods can learn complex behaviors from large-scale datasets, but are

Kristina Bätz, Frank Werner

It is well-known in practice, that L^1 data fitting leads to improved robustness compared to standard L^2 data fitting. However, it is unclear whether resulting algorithms will perform as well in case of regular data without outliers. In this paper, we therefore analyze generalized Tikhonov regularization with L^1 data fidelity for Inverse Problems F(u) = g

Jiaqi Lin, Yi Jiang, Abhronil Sengupta

Spiking Neural Networks (SNNs) promise energy-efficient, sparse, biologically inspired computation. Training them with Backpropagation Through Time (BPTT) and surrogate gradients achieves strong performance but remains biologically implausible. Equilibrium Propagation (EP) provides a more local and biologically grounded alternative. However, existing EP fram

Quentin Hillebrand, Pasin Manurangsi, Vorapong Suppakitpaisarn, Phanu Vajanopath

Rank aggregation is a task of combining the rankings of items from multiple users into a single ranking that best represents the users' rankings. Alabi et al. (AAAI'22) presents differentially-private (DP) polynomial-time approximation schemes (PTASes) and $5$-approximation algorithms with certain additive errors for the Kemeny rank aggregation problem in bo

Derun Zhou, Keisuke Yano, Mahito Sugiyama

In probabilistic modeling, parameter estimation is commonly formulated as a minimization problem on a parameter manifold. Optimization in such spaces requires geometry-aware methods that respect the underlying information structure. While the natural gradient leverages the Fisher information metric as a form of Riemannian gradient descent, it remains a first

Vidar Skogvoll, Tor Ole Odden

Generative AI is rapidly reshaping how physicists teach, learn, and conduct research, yet little is known about how physics faculty are responding to these changes. We interviewed 12 physics professors at a major Scandinavian research university to explore their uses and perceptions of Generative AI (GenAI) in both teaching and research. Using the theoretica

Vincenzo Amato, Rosa Barbato, Simone Cito, Alba Lia Masiello

The purpose of this paper is to establish a quantitative version of the Talenti comparison principle for solutions to the Poisson equation with Robin boundary conditions. This quantitative enhancement is proved in terms of the asymmetry of domain. The key role is played by a careful analysis of the propagation of asymmetry for the level sets of the solutions

Jawad Ibn Ahad, Muhammad Rafsan Kabir, Robin Krambroeckers, Sifat Momen

Natural Language Processing (NLP) has transformed the financial industry, enabling advancements in areas such as textual analysis, risk management, and forecasting. Large language models (LLMs) like BloombergGPT and FinMA have set new benchmarks across various financial NLP tasks, including sentiment analysis, stock movement prediction, and credit risk asses

Lisanne van Veenen, Piyush Sharda, Serena Viti, Shyam H. Menon

Cosmological simulations find that pockets of star-forming gas could remain pristine up until the Epoch of Reionization (EoR) due to the inhomogeneous nature of metal mixing and enrichment in the early Universe. Such pristine clouds could have formed Population III stars, which could have distinct properties compared to their very high redshift ($z \geq 20$)

Tanveer Hannan, Dimitrios Mallios, Parth Pathak, Faegheh Sardari

Large Vision-Language Models (LVLMs) have demonstrated strong multimodal reasoning capabilities on long and complex documents. However, their high memory footprint makes them impractical for deployment on resource-constrained edge devices. We present DocSLM, an efficient Small Vision-Language Model designed for long-document understanding under constrained m

Zifei Yu, Baode Li

We obtain the result of approximating \( f \) in the \( H^1(\mathbb{R}) \) norm using partial Hausdorff integrals. Specifically, by leveraging the homogeneous multiplier theory of \( H^1(\mathbb{R}) \) and the \( K \) functional theory, one result from Pinos and Liflyand [CMB,~2021,~64,~no.3] is extended from \( L^p(\mathbb{R}) \) ( \( 1 \leq p \leq \infty \

Petros Koutsouvelis, Matej Gazda, Leroy Volmer, Sina Amirrajab

The field of computer vision is undergoing a paradigm shift toward large-scale foundation model pre-training via self-supervised learning (SSL). Leveraging large volumes of unlabeled brain MRI data, such models can learn anatomical priors that improve few-shot performance in diverse neuroimaging tasks. However, most SSL frameworks are tailored to natural ima

Joseph Abdo, Aditya Shibu, Moaiz Saeed, Abdul Maajid Aga

Autonomous racing offers a rigorous setting to stress test perception, planning, and control under high speed and uncertainty. This paper proposes an approach to design and evaluate a software stack for an autonomous race car in CARLA: Car Learning to Act simulator, targeting competitive driving performance in the Formula Student UK Driverless (FS-AI) 2025 c

Riccardo Zuliani, Efe C. Balta, John Lygeros

Model-based policy optimization often struggles with inaccurate system dynamics models, leading to suboptimal closed-loop performance. This challenge is especially evident in Model Predictive Control (MPC) policies, which rely on the model for real-time trajectory planning and optimization. We introduce a novel policy optimization framework for MPC-based pol

Saptarshi Neil Sinha, Julius Kühn, Mika Silvan Goschke, Michael Weinmann

Automated and selective harvesting of fruits has become an important area of research, particularly due to challenges such as high costs and a shortage of seasonal labor in advanced economies. This paper focuses on 6D pose estimation of strawberries using purely synthetic data generated through a procedural pipeline for photorealistic rendering. We employ th

Wei Fan, JinYi Yoon, Bo Ji

Large Language Model (LLM) agent systems have advanced rapidly, driven by their strong generalization in zero-shot settings. To further enhance reasoning and accuracy on complex tasks, Multi-Agent Debate (MAD) has emerged as a promising framework that engages multiple LLM agents in structured debates to encourage diverse reasoning. However, triggering MAD fo

Chenghan Fu, Daoze Zhang, Yukang Lin, Zhanheng Nie

We introduce MOON, our comprehensive set of sustainable iterative practices for multimodal representation learning for e-commerce applications. MOON has already been fully deployed across all stages of Taobao search advertising system, including retrieval, relevance, ranking, and so on. The performance gains are particularly significant on click-through rate

Sabir Ahammed, Molla Basir Ahamed, Ming-Sheng Liu

Slice regular functions are a generalization of holomorphic functions to the setting of quaternions (and more generally, Clifford algebras). In this paper, we first establish the Bohr inequality for slice starlike functions and slice close-to-convex functions over quaternions $\mathbb{H}$. Next, we present a generalization of the Bohr inequality, and improve

Katayoun Eshkofti, Henrik Sandberg, Mikael Nilsson, Matthieu Barreau

Monitoring wastewater pump stations is essential because they are critical infrastructure. However, monitoring is still often performed manually due to the lack of suitable algorithmic methods and data. This paper introduces a high-fidelity, physics-enhanced simulator of a three-pump wastewater station that captures transient hydro-mechanical dynamics at a o

Konan A. Allaly, Jozef Urban, Karol Mikula

Segmentation of the aorta is crucial for various medical analyses, such as the diagnosis and treatment of cardiovascular diseases. This work presents mathematical models and methods yielding a semi-automatic segmentation of the aorta from non-enhanced CT data. Our framework consists of three steps. First, using the minimal path approach, we extract a path wi

Junqing Cai, Guanghui Wang, Yun Wang, Zhiwei Zhang

An antidirected cycle in a digraph $G$ is a subdigraph whose underlying graph is a cycle, and in which no two consecutive edges form a directed path in $G$. Let $\sigma_{+-}(G)$ be the minimum value of $d^+(x)+d^-(y)$ over all pairs of vertices $x, y$ such that there is no edge from $x$ to $y$, that is, $$\sigma_{+-}(G)=\min\{d^+(x)+d^-(y): \{x,y\}\subseteq

Ruoxi Cheng, Haoxuan Ma, Teng Ma, Hongyi Zhang

Large Vision-Language Models (LVLMs) exhibit powerful reasoning capabilities but suffer sophisticated jailbreak vulnerabilities. Fundamentally, aligning LVLMs is not just a safety challenge but a problem of economic efficiency. Current alignment methods struggle with the trade-off between safety, utility, and operational costs. Critically, a focus solely on

Abrar Ahmed Naqash, Fardeen Ahmad Sofi, Mohammad Haris Khan, Sundus Abdi

Nonlinear eigenvalue equations arise naturally in quantum information theory, particularly in the variational quantification of entanglement. In this work, we present a hybrid analytical and numerical framework for evaluating the geometric measure of entanglement. The method combines a Gauss Seidel fixed point iteration with a controlled perturbative correct

Haokun Chen, Jianing Li, Yao Zhang, Jinhe Bi

Multimodal Large Language Models (MLLMs) achieve impressive performance once optimized on massive datasets. Such datasets often contain sensitive or copyrighted content, raising significant data privacy concerns. Regulatory frameworks mandating the 'right to be forgotten' drive the need for machine unlearning. This technique allows for the removal of target

Yihao Zhang, Yuankai Qi, Xi Zheng

Foundation models applied in robotics, particularly \textbf{Vision--Language--Action (VLA)} models, hold great promise for achieving general-purpose manipulation. Yet, systematic real-world evaluations and cross-model comparisons remain scarce. This paper reports our \textbf{empirical experiences} from benchmarking four representative VLAs -- \textbf{ACT}, \

Gabriele Buriola, Andreas Weiermann

Higman's lemma and Kruskal's theorem are two of the most celebrated results in the theory of well quasi-orders. In his seminal paper G. Higman obtained what is known as Higman's lemma as a corollary of a more general theorem, dubbed here Higman's theorem. While the lemma deals with finite sequences over a well quasi-order, the theorem is abou

Cyriana M. A. Roelofs, Edison Guevara Bastidas, Thomas Hugo, Stefan Faulstich

Early detection of faults in district heating substations is imperative to reduce return temperatures and enhance efficiency. However, progress in this domain has been hindered by the limited availability of public, labelled datasets. We present an open-source framework combining a service report validated public dataset, an evaluation method based on accura

Shahriar Hasnat Kazi, Niall Adams, Edward A. K. Cohen

This paper develops the first online algorithms for estimating the spectral density function -- a fundamental object of interest in time series analysis -- that satisfies the three core requirements of streaming inference: fixed memory, fixed computational complexity, and temporal adaptivity. Our method builds on the concept of forgetting factors, allowing t

Yichao Tang, Mingyang Li, Di Miao, Sheng Li

The advancement of artificial intelligence generated content (AIGC) has created a pressing need for robust image watermarking that can withstand both conventional signal processing and novel semantic editing attacks. Current deep learning-based methods rely on training with hand-crafted noise simulation layers, which inherently limit their generalization to

Bertille Tierny, Arthur Charpentier, François Hu

Linear models are widely used in high-stakes decision-making due to their simplicity and interpretability. Yet when fairness constraints such as demographic parity are introduced, their effects on model coefficients, and thus on how predictive bias is distributed across features, remain opaque. Existing approaches on linear models often rely on strong and un

Jiheum Park, Chao Pang, Tristan Y. Lee, Jeong Yun Yang

Current cancer screening guidelines cover only a few cancer types and rely on narrowly defined criteria such as age or a single risk factor like smoking history, to identify high-risk individuals. Predictive models using electronic health records (EHRs), which capture large-scale longitudinal patient-level health information, may provide a more effective too

Santiago Arranz-Olmos, Gilles Barthe, Lionel Blatter, Benjamin Grégoire

High-assurance cryptography provides strong guarantees that source implementations are functionally correct and provably secure. In this paper, we demonstrate that the Jasmin compiler preserves functional correctness and KEM-IND-CCA security (which were established in prior work) of a highly optimized Jasmin implementation of ML-KEM used in the popular messe

Jean-Christophe Aval, Sébastien Labbé

We consider the $q$-deformation of rational numbers introduced recently by Morier-Genoud and Ovsienko. We propose three enumerative interpretations of these $q$-rationals: in terms of a new version of Ostrowski's numeration system for integers, in terms of order ideals of fence posets and in terms of perfect matchings of snake graphs. Contrary to previous re

Hengfei Wang, Zhongqun Zhang, Yihua Cheng, Hyung Jin Chang

Gaze redirection methods aim to generate realistic human face images with controllable eye movement. However, recent methods often struggle with 3D consistency, efficiency, or quality, limiting their practical applications. In this work, we propose RTGaze, a real-time and high-quality gaze redirection method. Our approach learns a gaze-controllable facial re

Mann Khatri, Mirza Yusuf, Rajiv Ratn Shah, Ponnurangam Kumaraguru

Large Language Models (LLMs), trained on extensive datasets from the web, exhibit remarkable general reasoning skills. Despite this, they often struggle in specialized areas like law, mainly because they lack domain-specific pretraining. The legal field presents unique challenges, as legal documents are generally long and intricate, making it hard for models

Ruslan R. Boyko

It is known that the price of call options in the Heston model is determined in a non-unique way. In this paper, this problem is analyzed from the point of view of the existing mathematical theory of uniqueness classes for degenerate parabolic equations. For the special case of degeneracy, a new example is constructed demonstrating the accuracy of the unique

Sven Schultze, Meike Verena Kietzmann, Nils-Lucas Schönfeld, Ruth Stock-Homburg

The increasing deployment of autonomous AI agents on the web is hampered by a fundamental misalignment: agents must infer affordances from human-oriented user interfaces, leading to brittle, inefficient, and insecure interactions. To address this, we introduce VOIX, a web-native framework that enables websites to expose reliable, auditable, and privacy-prese

Ruoqi Wang, Haitao Wang, Shaojie Guo, Qiong Luo

Out-of-domain (OOD) robustness is challenging to achieve in real-world computer vision applications, where shifts in image background, style, and acquisition instruments always degrade model performance. Generic augmentations show inconsistent gains under such shifts, whereas dataset-specific augmentations require expert knowledge and prior analysis. Moreove

Yuki Miyoshi, Masaki Inoue, Yusuke Fujimoto

Natural language data, such as text and speech, have become readily available through social networking services and chat platforms. By leveraging human observations expressed in natural language, this paper addresses the problem of state estimation for physical systems, in which humans act as sensing agents. To this end, we propose a Language-Aided Particle

Jiawei Chen, Junhao Peng, Yanwei Liang, Renhai Wang

Room temperature superconductivity remains elusive, and hydrogen-base compounds despite remarkable transition temperatures(Tc) typically require extreme pressures that hinder application. To accelerate discovery under moderate pressures, an interpretable framework based on symbolic regression is developed to predict Tc in hydrogen-based superconductors. A ke

Joris Roos, Andreas Seeger

We survey old and new conjectures and results on various types of spherical maximal functions, emphasizing problems with a fractal dilation set.

Erwan Christian Escudie, Matthia Sabatelli, Olivier Buffet, Jilles Steeve Dibangoye

We present a novel framework for {\epsilon}-optimally solving two-player zero-sum partially observable stochastic games (zs-POSGs). These games pose a major challenge due to the absence of a principled connection with dynamic programming (DP) techniques developed for two-player zero-sum stochastic games (zs-SGs). Prior attempts at transferring solution metho

Byoungjun Park, Pedro Porto Buarque de Gusmão, Dongjin Ji, Minhoe Kim

Federated learning is a promising paradigm that utilizes distributed client resources while preserving data privacy. Most existing FL approaches assume clients possess labeled data, however, in real-world scenarios, client-side labels are often unavailable. Semi-supervised Federated learning, where only the server holds labeled data, addresses this issue. Ho

Patrick Koopmann, Yasir Mahmood, Axel-Cyrille Ngonga Ngomo, Balram Tiwari

We introduce the notion of contrastive ABox explanations to answer questions of the type "Why is a an instance of C, but b is not?". While there are various approaches for explaining positive entailments (why is C(a) entailed by the knowledge base) as well as missing entailments (why is C(b) not entailed) in isolation, contrastive explanations consider both

K K Pathak, S Bhattacharya, T Das

The leptonic and semileptonic decay propertis of Bc mesons are studied in a QCD potential model. The CKM element for the leptonic decay is calculated by evaluating its decay constant. With the help of the Isgur-Wise function, the form factors and the CKM element for semileptonic decay are also calculated. The variation in four-momentum square in the two deca

Minoru Kitahara, Hiroshi Uno

This paper studies one-sided matching under a complete exchange (CE) requirement, where each agent must be assigned an object different from its initial endowment. We introduce assignment partition -- a partition of agents and choice sets that builds CE into feasibility -- and, within this structure, propose two new mechanisms. Chain Serial Dictatorship (C-S

Haoyi Wang

Volumetric medical image segmentation presents unique challenges due to the inherent anatomical structure and limited availability of annotations. While recent methods have shown promise by contrasting spatial relationships between slices, they rely on hard binary thresholds to define positive and negative samples, thereby discarding valuable continuous info

Julius Wenzel, Syeda Umaima Alam, Andreas Schmidt, Hanwei Zhang

An ever increasing number of high-stake decisions are made or assisted by automated systems employing brittle artificial intelligence technology. There is a substantial risk that some of these decision induce harm to people, by infringing their well-being or their fundamental human rights. The state-of-the-art in AI systems makes little effort with respect t

Jin-Yang Shen, Xin-Yi Liu, Jin-Rui Wu, Yue-Liang Wu

We present a fully back-reacted Einstein--Maxwell--Dilaton--flavor model with dynamical light and strange sectors, calibrated to lattice QCD using a machine-learning--assisted spectral method. The model reproduces the 2+1-flavor equation of state and chiral dynamics with quantitative accuracy, and maps the Columbia plot with a tri-critical point at $m_s^{\ma

Yang Xiao, Yuxiao Xie

We generalize the construction of Compactified Imaginary Liouville Theory (CILT), a non-unitary logarithmic Conformal Field Theory (CFT) defined on closed surfaces, to surfaces with boundary. Starting from a compactified Gaussian Free Field (GFF) with Neumann boundary condition, we perturb it by adding in curvature terms and exponential potentials on both th

R. G. G. Amorim, A. F. Santos, K. V. S. Araújo, S. C. Ulhoa

This article deals with a nonrelativistic cosmological model based on Galilean covariance, formulated within a five-dimensional Galilean manifold. Within this framework, we construct an isotropic and homogeneous metric analogous to the Friedmann--Robertson--Walker metric but without a universal speed limit. Two distinct solutions of the Einstein-like field e

Bruno Grenet

The works presented in this habilitation concern the algorithmics of polynomials. This is a central topic in computer algebra, with numerous applications both within and outside the field - cryptography, error-correcting codes, etc. For many problems, extremely efficient algorithms have been developed since the 1960s. Here, we are interested in how this effi

Mikel Robredo, Matteo Esposito, Davide Taibi, Rafael Peñaloza

Software quality research increasingly relies on large-scale datasets that measure both the product and process aspects of software systems. However, existing resources often focus on limited dimensions, such as code smells, technical debt, or refactoring activity, thereby restricting comprehensive analyses across time and quality dimensions. To address this

Tobias Kappé, Alexandra Silva, Jana Wagemaker

This booklet serves as an introduction to Kleene Algebra (KA), a set of laws that can be used to study general equivalences between programs. It discusses how general programs can be modeled using regular expressions, how those expressions correspond to automata, and how this correspondence can be exploited to obtain the central result of KA, namely that an

Melika Behjati, James Henderson

Fine-grained knowledge is crucial for vision-language models to obtain a better understanding of the real world. While there has been work trying to acquire this kind of knowledge in the space of vision and language, it has mostly focused on aligning the image patches with the tokens on the language side. However, image patches do not have any meaning to the

Vilde G. S. Lunde, Benedicte S. Ofstad, Øystein S. Fjellvåg, Philippe Ohresser

This work presents electronic and magnetic properties of selected members in the cubic Laves phase series Nd1-xPrxCoNi and Ce0.25Pr0.75CoNi, together with the corresponding binary compositions (NdCo2, NdNi2, PrCo2, PrNi2, CeCo2, CeNi2), using soft x-ray absorption spectroscopy, x-ray magnetic circular dichroism (XMCD), density-functional theory, and crystal

Anish Kumar, Vivek Semwal, Shradha Mishra

We study the collective behavior of a polar flock in an inhomogeneous environment in two-dimensions. The inhomogeneity is modelled by introducing regions at random locations on the substrate with higher noise but accessible for the flock to move. Hence inside such regions the particles orientation get randomised. Such inhomogeneities are different from the p

Sania Nayab, Marco Simoni, Giulio Rossolini, Andrea Saracino

The generation of questions and answers (QA) from knowledge graphs (KG) plays a crucial role in the development and testing of educational platforms, dissemination tools, and large language models (LLM). However, existing approaches often struggle with scalability, linguistic quality, and factual consistency. This paper presents a scalable and deterministic

Yuqi Yin, Yibo Fu, Siyuan Wang, Peng Sun

The discovery of novel Ionic Liquids (ILs) is hindered by critical challenges in property prediction, including limited data, poor model accuracy, and fragmented workflows. Leveraging the power of Large Language Models (LLMs), we introduce AIonopedia, to the best of our knowledge, the first LLM agent for IL discovery. Powered by an LLM-augmented multimodal d

Jingyu Lin, Li Chen, Xiaoqian Ye

Non-binary linear block codes (NB-LBCs) are an important class of error-correcting codes that are especially competent in correcting burst errors. They have broad applications in modern communications and storage systems. However, efficient soft-decision decoding of these codes remains to be further developed. This paper proposes successive cancellation list

Wencai Ye, Mingjie Sun, Shuhang Chen, Wenjin Wu

Large Language Models (LLMs) demonstrate significant advantages in leveraging structured world knowledge and multi-step reasoning capabilities. However, fundamental challenges arise when transforming LLMs into real-world recommender systems due to semantic and behavioral misalignment. To bridge this gap, we propose Align$^3$GR, a novel framework that unifies

Jing Yu, Xiangjun Zhen

The aim of this paper is to study coquasitriangular structures on a class of cosemisimple Hopf algebras of the form $\Bbbk^G {}^\tau \#_{\sigma} \Bbbk F$, constructed as abelian extensions of $\Bbbk F$ by $\Bbbk^G$ for a finite group $G$ and an arbitrary group $F$. We investigate when a coquasitriangular structure exists on $\Bbbk^G{}^\tau\#_{\sigma}\Bbbk F$

Hyemin Boo, Hyoryung Kim, Myungjin Lee, Seunghyeon Lee

Text-to-image diffusion models generate realistic and coherent images but often fail to follow numerical instructions in text, revealing a gap between language and visual representation. Interestingly, we found that these models are not entirely blind to numbers-they are implicitly aware of their own counting accuracy, as their internal signals shift in cons

Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah

Autonomous aerial systems increasingly rely on large language models (LLMs) for mission planning, perception, and decision-making, yet the lack of standardized and physically grounded benchmarks limits systematic evaluation of their reasoning capabilities. To address this gap, we introduce UAVBench, an open benchmark dataset comprising 50,000 validated UAV f

Tianzheng Miao, Thomas Feys, Gilles Callebaut, Jarne Van Mulders

Distributed MIMO (D-MIMO) has emerged as a key architecture for future sixth-generation (6G) networks, enabling cooperative transmission across spatially distributed access points (APs). However, most existing studies rely on idealized channel models and lack hardware validation, leaving a gap between algorithmic design and practical deployment. Meanwhile, r

Biagio Boi, Christian Esposito

Smart contracts have emerged as key components within decentralized environments, enabling the automation of transactions through self-executing programs. While these innovations offer significant advantages, they also present potential drawbacks if the smart contract code is not carefully designed and implemented. This paper investigates the capability of l

Melih Coşğun, Mert Gençtürk, Sinem Sav

Data normalization is a crucial preprocessing step for enhancing model performance and training stability. In federated learning (FL), where data remains distributed across multiple parties during collaborative model training, normalization presents unique challenges due to the decentralized and often heterogeneous nature of the data. Traditional methods rel

Jianyu Wei, Qingtao Li, Shijie Cao, Lingxiao Ma

Large language models (LLMs) are increasingly deployed on customer devices. To support them, current devices are adopting SoCs (System on Chip) with NPUs (Neural Processing Unit) installed. Although high performance is expected, LLM inference on NPUs is slower than its CPU counterpart. The reason is that NPUs have poor performance on computations other than

Mats Esseldeurs, Leen Decin, Joris De Ridder, Yoshiya Mori

Close companions influence stellar evolution through tidal interactions, mass transfer, and mass loss effects. While such companions are detected around young stellar objects, main-sequence stars, red giants, and compact objects, direct observational evidence of close-in companions around asymptotic giant branch (AGB) stars has remained elusive. Here, we pre

Tin Nwe Aye, Linus Carlsson

This article explores the properties of an SEIR$^{\text{T}}$R$^{\text{P}}$D endemic model expressed through delay-differential equations with distributed delays for latency and temporary immunity. Our research delves into the variability of latent periods and immunity durations across diseases, in particular, we introduce a class of delays defined by continu

Hong Huang, Chengyu Yao, Haiming Chen, Hang Gao

Attributed graphs, typically characterized by irregular topologies and a mix of numerical and categorical attributes, are ubiquitous in diverse domains such as social networks, bioinformatics, and cheminformatics. While graph kernels provide a principled framework for measuring graph similarity, existing kernel methods often struggle to simultaneously captur

Shijian Deng, Erin E. Kosloski, Siva Sai Nagender Vasireddy, Jia Li

The automatic detection of gaze targets in autistic children through artificial intelligence can be impactful, especially for those who lack access to a sufficient number of professionals to improve their quality of life. This paper introduces a new, real-world AI application for gaze target detection in autistic children, which predicts a child's point of g

Jitesh Chavan, Rohit Lal, Anand Kamat, Mengjia Xu

State-space models (SSMs), Mamba in particular, are increasingly adopted for long-context sequence modeling, providing linear-time aggregation via an input-dependent, causal selective-scan operation. Along this line, recent "Mamba-for-vision" variants largely explore multiple scan orders to relax strict causality for non-sequential signals (e.g., images). Ra

Shengbang Zhou, Qing Li, Yi Ren, Jingyan Xu

The use of nuclear spins as physical sensing systems is disadvantaged by their low signal responsivity, particularly when compared to sensing techniques based on electron spins. This primarily results from the small nuclear gyromagnetic ratio and the difficulties in achieving high spin polarization. Here we develop a new approach to investigating the respons

J. T. Horne-Jones, D. J. Glinnan, A. G. W. Lawrie, R. J. R. Williams

The dynamics of turbulent mixing induced by Rayleigh-Taylor instability are heavily dependent on the acceleration experienced by the fluids and the frequency content of the initial interface between them. Both are readily controllable in numerical simulations, but in experimental studies are difficult to influence and adequately diagnose. In this paper we pr

Yuhan Xie, Chen Lyu

Split Federated Learning (SFL) is an emerging paradigm for privacy-preserving distributed learning. However, it remains vulnerable to sophisticated data poisoning attacks targeting local features, labels, smashed data, and model weights. Existing defenses, primarily adapted from traditional Federated Learning (FL), are less effective under SFL due to limited

Zhongbin Guo, Jiahe Liu, Yushan Li, Wenyu Gao

Existing Vision Language Models (VLMs) architecturally rooted in "flatland" perception, fundamentally struggle to comprehend real-world 3D spatial intelligence. This failure stems from a dual-bottleneck: input-stage conflict between computationally exorbitant geometric-aware encoders and superficial 2D-only features, and output-stage misalignment where discr

Seed, Baisheng Li, Banggu Wu, Bole Ma

We introduce Virtual Width Networks (VWN), a framework that delivers the benefits of wider representations without incurring the quadratic cost of increasing the hidden size. VWN decouples representational width from backbone width, expanding the embedding space while keeping backbone compute nearly constant. In our large-scale experiment, an 8-times expansi

Daniel Blankenburg, Antonia Ellerbrock, Thomas Kesselheim, Jens Vygen

We study the problem of minimizing an ordered norm of a load vector (indexed by a set of $d$ resources), where a finite number $n$ of customers $c$ contribute to the load of each resource by choosing a solution $x_c$ in a convex set $X_c \subseteq \mathbb{R}^d_{\geq 0}$; so we minimize $||\sum_{c}x_c||$ for some fixed ordered norm $||\cdot||$. We devise a ra

Cong Cao, Huanjing Yue, Yujie Xu, Xiaodong Xu

In recent years, image editing has garnered growing attention. However, general image editing models often fail to produce satisfactory results when confronted with new styles. The challenge lies in how to effectively fine-tune general image editing models to new styles using only a limited amount of paired data and a minimum number of parameters. To address

Maan Al Balkhi, Kordian Gontarska, Marko Harasic, Adrian Paschke

This paper investigates neural network-based biometric authentication using finger-drawn digits on touchscreen devices. We evaluated CNN and autoencoder architectures for user authentication through simple digit patterns (0-9) traced with finger input. Twenty participants contributed 2,000 finger-drawn digits each on personal touchscreen devices. We compared

Jader Martins Camboim de Sá, Jooyoung Lee, Cédric Pruski, Marcos Da Silveira

Fine-grained word meaning resolution remains a critical challenge for neural language models (NLMs) as they often overfit to global sentence representations, failing to capture local semantic details. We propose a novel adversarial training strategy, called LANE, to address this limitation by deliberately shifting the model's learning focus to the target wor

Huajian Zhang, Mingyue Cheng, Yucong Luo, Xiaoyu Tao

Table reasoning with large language models (LLMs) plays a critical role in building intelligent systems capable of understanding and analyzing tabular data. Despite recent progress, existing methods still face key limitations: their reasoning processes lacks depth and explicit multi-step reasoning, often relying solely on implicit language model understandin

Mingwei Xing, Xinliang Wang, Yifeng Shi

Constructing a unified 3D scene understanding model has long been hindered by the significant topological discrepancies across different sensor modalities. While applying the Mixture-of-Experts (MoE) architecture is an effective approach to achieving universal understanding, we observe that existing 3D MoE networks often suffer from semantics-driven routing

Abiram Panchalingam, Indu Bodala, Stuart Middleton

High-fidelity gaze redirection is critical for generating augmented data to improve the generalization of gaze estimators. 3D Gaussian Splatting (3DGS) models like GazeGaussian represent the state-of-the-art but can struggle with rendering subtle, continuous gaze shifts. In this paper, we propose DiT-Gaze, a framework that enhances 3D gaze redirection models