March 2026 arXiv papers — page 8

Anmin Liu, Ruixuan Yang, Huiqiang Jiang, Bin Lin

Long-context video understanding and generation pose a significant computational challenge for Transformer-based video models due to the quadratic complexity of self-attention. While existing sparse attention methods employ coarse-grained patterns to improve efficiency, they typically incur redundant computation and suboptimal performance. To address this is

Ziliang Guo, Ziheng Li, Bo Tang, Feiyu Xiong

Memory-augmented Large Language Models (LLMs) are essential for developing capable, long-term AI agents. Recently, applying Reinforcement Learning (RL) to optimize memory operations, such as extraction, updating, and retrieval, has emerged as a highly promising research direction. However, existing implementations remain highly fragmented and task-specific,

David Bani-Harouni, Chantal Pellegrini, Julian Lüers, Su Hwan Kim

Safe deployment of Large Vision-Language Models (LVLMs) in radiology report generation requires not only accurate predictions but also clinically interpretable indicators of when outputs should be thoroughly reviewed, enabling selective radiologist verification and reducing the risk of hallucinated findings influencing clinical decisions. One intuitive appro

Mohammad Mahdi Mesgari, Jackie Ma, Wojciech Samek, Sebastian Lapuschkin

In the evaluation of attribution quality, the quantitative assessment of explanation legibility is particularly difficult, as it is influenced by varying shapes and internal organization of attributions not captured by simple statistics. To address this issue, we introduce Minimum Spanning Tree Compactness (MST-C), a graph-based structural metric that captur

Simon Schmidt, Nicole Gehring, Abdurrahman Irscheid

The paper presents an approach to flatness-based control design for hyperbolic multi-input systems, building upon the hyperbolic controller form (HCF). The transformation into HCF yields a simplified system representation that considerably facilitates the design of state feedback controllers for trajectory tracking. The proposed concept is demonstrated for a

Yu Su, Yao Wang

Open quantum systems are traditionally described by decomposing the total Hilbert space into a system and an external environment, linked by an explicit interaction Hamiltonian. We propose an alternative framework in which the environment is not introduced as an independent sector a priori, but instead emerges from the dynamical activation of constraints in

Aakanksha Tewari, Jai Mangal, Sumit J Darak, Shobha Sundar Ram

Integrated Sensing and Communication (ISAC) is a key enabler of high speed, ultra low latency vehicular communication in 6G. ISAC leverages radar signal processing (RSP) to localize multiple unknown targets amid static clutter by jointly estimating range, azimuth, and Doppler velocity (3D), thereby enabling highly directional beamforming toward intended mobi

M. Hooft, A. Nikolakopoulos, J. García-Marcos, Y. De Backer

Single-pion production is an important contribution to the total neutrino-nucleus interaction cross section in accelerator-based neutrino oscillation experiments. The goal of this paper is to improve the Ghent model in the Delta resonance region by incorporating as many physical constraints as possible while keeping the number of fitted parameters as low as

Wu Zuhui, Wang Qiuping, Yan Ting

In this paper, we introduce a general model for jointly modelling the nodal heterogeneity and covariates in weighted or unweighted bipartite networks, which contains two different types of nodes. The model has a degree heterogeneity parameter for each node and a fixed-dimensional regression coefficient for the covariates. We use the method of moments to esti

Karan Pathak, David Atienza, Marina Zapater

The growing demands in the training and inference of Large Language Models (LLMs) are accelerating the adoption of scale-up systems that extend server shared memory through the use of Compute Express Link (CXL)-based load/store interconnects. Accurate full-system simulation of such architectures remains challenging, as existing tools (all very recent) rely o

Berthold-Georg Englert

Majorana found a way to exploit the scaling properties of the Thomas-Fermi equation for converting this second-order differential equation into one of first order. We explore his method for the familiar neutral-atom solution and extend it to the solution that is relevant for weakly ionized atoms. Various integrals and other quantities with importance for ato

Thomas Franosch

In the first part of these short lecture notes, we will present an introduction on (auto-)correlation functions and linear-response functions in the language of a physicist. In particular, the fluctuation-dissipation theorem in classical physics is presented underlining the central role of correlation functions. The fundamental importance of (auto-)correlati

Huanfeng Shen, Chan Li, Menghui Jiang, Penghai Wu

Land surface temperature (LST) is a critical parameter for characterizing surface energy balance and hydrothermal processes. While Landsat provides invaluable LST observations at medium spatial resolution for over 40 years, its native spatial resolution of thermal bands (e.g., 100 m) remains insufficient compared to its 30 m optical bands, failing to meet th

Laurent Feuilloley, Soumyadeep Paul, Ami Paz

We consider three classification systems for distributed decision tasks: With unbounded computation and certificates, defined by Balliu, D'Angelo, Fraigniaud, and Olivetti [JCSS'18], and with (two flavors of) polynomially bounded local computation and certificates, defined in recent works by Aldema Tshuva and Oshman [OPODIS'23], and by Reiter [PODC'24]. The

Veronica T. C. Lai, Christopher E. Patrick

Nd-Fe-B magnets are the most widely used high performance magnets in the world today, and remain the subject of both experimental and computational research aimed at understanding and optimizing them. Atomistic spin dynamics (ASD) is one technique which has been used in recent years to provide insight into magnetic properties relevant to coercivity, such as

Philipp Reis, Jacqueline Henle, Stefan Otten, Eric Sax

The increasing capabilities of machine learning models, such as vision-language and multimodal language models, are placing growing demands on data in automotive systems engineering, making the quality and relevance of collected data enablers for the development and validation of such systems. Traditional Big Data approaches focus on large-scale data collect

Benny Avelin, Gerardo Barrera

In this manuscript, we establish the existence/non-existence of the cut-off phenomenon for the Langevin--Kolmogorov random dynamics with monomial convex potentials, possible singular, and driven by a Brownian motion with small strength. We consider a truncated $\chi^2$-distance, that is, a distance based on Galerkin projections of the eigensystem, and show t

N. Z. van Gils, E. Belli, M. Bergamaschi, A. Clairembaud

This manuscript proposes a method to enable controlled high-gradient particle acceleration when requiring self-modulation of the drive bunch. While electron bunch seeding of self-modulation (eSSM) has been realised at a plasma electron density $n_\mathrm{pe}\cong10^{14}\mathrm{cm}^{-3}$, it has not been demonstrated at higher plasma densities due to limitati

Yuki Tochitani

We determine the generators of the autoequivalence group of the derived category of coherent sheaves on a bielliptic surface over an algebraically closed field of arbitrary characteristic. As a consequence, we prove that any algebraic variety derived equivalent to such a surface is isomorphic to the surface itself.

Andrew G. Ross, Alan M. Ross

Socio-technical scenarios for net-zero and other transformation pathways combine qualitative storylines with quantitative models, embedding them in plausible societal contexts for model assessment. Conventional scenario generation is resource-intensive, can be limited in internal consistency and diversity of expert and stakeholder perspectives, and is rarely

Xudong Zhou, Jinyuan Liang, Qiuyi Guo, Guozheng Li

We present iPoster, an interactive layout generation framework that empowers users to guide content-aware poster layout design by specifying flexible constraints. iPoster enables users to specify partial intentions within the intention module, such as element categories, sizes, positions, or coarse initial drafts. Then, the generation module instantly genera

Nuno Freitas, Filip Gawron

We investigate the splitting behavior of $\ell$ in the coefficient field of a newform $f$ of level $N$, under the assumption that $f$ is congruent modulo a prime above $\ell$ to another newform $g$ whose level divides $N/p^2$ for some prime $p\mid N$. In particular, we show that the maximal real subfield of the $\ell$-th cyclotomic field, $\mathbb{Q}(\zeta_\

Seung Hun Han, Youssef Mohamed, Mohamed Elhoseiny

This paper presents a Multilingual Vision Large Language Model, named M-MiniGPT4. Our model exhibits strong vision-language understanding (VLU) capabilities across 11 languages. We utilize a mixture of native multilingual and translated data to push the multilingual VLU performance of the MiniGPT4 architecture. In addition, we propose a multilingual alignmen

Davide Aruta, Francesca Prinari, Francesco Solombrino

We consider the homogenization of random integral functionals which are possibly unbounded, that is, the domain of the integrand is not the whole space and may depend on the space-variable. In the vectorial case, we develop a complete stochastic homogenization theory for nonconvex unbounded functionals with convex growth of generalized Orlicz-type, under a s

Simon Girel, Quentin Richard

We investigate an infection-age structured competitive epidemiological model involving multiple strains. While classical results establish competitive exclusion when a unique maximal basic reproduction number exists, we provide here a complete characterization of the asymptotic behavior for an arbitrary number of populations without assuming uniqueness of th

Timothy M. Craig, Adrien Moncomble, Ajinkya A. Kadu, Gail A. Vinnacombe-Willson

Direct observation of nanoscale transformations in three dimensions (3D) is essential for understanding materials evolution under operating conditions, yet dynamic electron tomography remains limited by slow tilt series acquisition and by reconstruction methods that assume static specimens. These constraints prevent continuous 3D imaging of evolving structur

Shuyin Xia, Meng Yang, Dawei Dai, Fan Chen

Superpixels provide a compact region-based representation that preserves object boundaries and local structures, and have therefore been widely used in a variety of vision tasks to reduce computational cost. However, most existing superpixel algorithms produce irregularly shaped regions, which are not well aligned with regular operators such as convolutions.

Pierre Andreoletti

We study trajectory forecasting under squared loss for time series with weak conditional structure, using highly expressive prediction models. Building on the classical characterization of squared-loss risk minimization, we emphasize regimes in which the conditional expectation of future trajectories is effectively degenerate, leading to trivial Bayes-optima

Dirk Dubbers

I study electron movement in electromagnetic fields beyond the adiabatic approximation, using so-called Stormer theory. Some of the electron orbits are regular or integrable, but their measure is zero. Other orbits, called quasiperiodic, are unstable, but only for infinite times. All other orbits are chaotic or hyperchaotic, or simple scattering states. Exam

Hao-Yang Qi, Zi-Fan Zhu, Yao Wang, Rui-Xue Xu

We study a system consisting of an infinite one-dimensional molecular chain and a locally coupled probe. Starting from the Hamiltonian of the chain-probe composite and the corresponding spectral densities, we evaluate the heat current between the probe and the chain. For this purpose, we develop a dissipaton-based quantum approach that is fully nonperturbati

Ewen Lallinec, Antoine Levitt

We present a comparative study of numerical methods for computingelectronic densities of states (DOS) in periodic systems. We provide a detailed analysis of the domain of validity of the Brillouincomplex deformation (BCD), a recently-proposed method promising exponential convergence without need for smearing. We compare on a range of systems the BCD with sev

Nianjheng Wu, Jules Lefeuvre, Andrew Mayne, Stéphane Campidelli

The low temperature magnetoresistance of graphene functionalized by an array of magnetic Terbium Phthalocyanines molecules is found to exhibit a magnetic field-dependent 1/f noise, along with universal conductance fluctuations (UCFs) typical of a mesoscopic phase-coherent sample. A thorough analysis of the magnetic field, temperature and chemical potential d

Ningzhi Gao, Siquan Huang, Leyu Shi, Ying Gao

Federated prototype learning (FPL), as a solution to heterogeneous federated learning (HFL), effectively alleviates the challenges of data and model heterogeneity.However, existing FPL methods fail to balance the fidelity and discriminability of the feature, and are limited by a single global prototype. In this paper, we propose FedDBP, a novel FPL method to

Baoyi Zeng, Andrea Nini

Authorship verification (AV), the task of determining whether a questioned text was written by a specific individual, is a critical part of forensic linguistics. While manual authorial impersonation by perpetrators has long been a recognized threat in historical forensic cases, recent advances in large language models (LLMs) raise new challenges, as adversar

Alexandros I. Papadopoulos, Sotiris Kopsinos, Dimitrios Tyrovolas, Antonios Lalas

Reconfigurable Intelligent Surfaces (RISs) are emerging as a key enabler of Programmable Wireless Environments for 6G, but their practical integration into operational networks still lacks orchestration mechanisms that can jointly support resource allocation, energy efficiency, and admission control with low online complexity. This paper presents SCROOGE, a

Alexander Armbruster, Fabrizio Grandoni, Antoine Tinguely, Andreas Wiese

The (Non-Preemptive) Throughput Maximization problem is a natural and fundamental scheduling problem. We are given $n$ jobs, where each job $j$ is characterized by a processing time and a time window, contained in a global interval $[0,T)$, during which~$j$ can be scheduled. Our goal is to schedule the maximum possible number of jobs non-preemptively on a si

Tom Simon, Stephane Nicolas, Pierrick Tranouez, Clement Chatelain

While state-of-the-art Handwritten Text Recognition (HTR) models perform well on standard benchmarks, they frequently struggle with writers exhibiting highly specific styles that are underrepresented in the training data. To handle unseen and atypical writers, writer adaptation techniques personalize HTR models to individual handwriting styles. Leading write

Youngung Han, Minkyung Cha, Kyeonghun Kim, Induk Um

Minimizing invasive diagnostic procedures to reduce the risk of patient injury and infection is a central goal in medical imaging. And yet, noninvasive diagnosis of perineural invasion (PNI), a critical prognostic factor involving infiltration of tumor cells along the surrounding nerve, still remains challenging, due to the lack of clear and consistent imagi

Avelino Vicente

A tentative approach to explain the flavor puzzle consists of embedding the Standard Model in a larger gauge symmetry that contains a separate gauge group for each fermion family. In such gauge non-universal (or flavor-deconstructed) theories, neutrinos pose some challenges. I will discuss existing ideas in the literature and present a simple model in which

Dmitri Panyushev, Oksana Yakimova

E.B. Vinberg's theory of quasi-derivations of algebras is extended to a broader framework of near-derivations. This deepens connections between Poisson geometry and Lie theory. Although basic results apply to arbitrary algebras, our substantial applications concern the Poisson algebra $(\mathcal S(\mathfrak q),\{\ ,\,\})$ of a Lie algebra $\mathfrak q$. We d

Arnaud Debussche, Baptiste Huguet

We study a stochastic multiscale spatial gene network. These naturally arise in molecular biology. In our model, the reactants are subject to on-site reactions on both scales and diffusion on the continuous scale only, although diffusion on both scales could easily be handled. We obtain, under a light condition on the scales between the total population size

Sampath Kumar Mulagaleti, Alberto Bemporad

We present a dual Model Predictive Control (MPC) framework for the simultaneous identification and control of quasi-Linear Parameter Varying (qLPV) systems. The framework is composed of an online estimator for the states and parameters of the qLPV system, and a controller that leverages the estimated model to compute inputs with a dual purpose: tracking a re

Stelios Negrepontis, Athanase Papadopoulos

We provide some historical context to the study of solid angles carried out by Euler in his memoir \emph{De mensura angulorum solidorum} (On the measure of solid angles). We extend our study to the general notion of angle (not only solid). While doing so, we explore some works by Ancient Greek mathematicians and others by Arabs mathematicians of the Middle-A

José M. Begines, Suman Das, Hyun-Sik Jeong, Juan F. Pedraza

Originally proposed by 't Hooft, the brick wall model has recently reemerged as a useful framework for probing quantum aspects of horizon physics, particularly in the context of holography. In this paper, we apply it to asymptotically de Sitter spacetimes. We compute the normal modes of a massless scalar field in pure de Sitter space and in the Schwarzschild

Yijie Zheng, Weijie Wu, Bingyue Wu, Long Zhao

While the Earth observation community has witnessed a surge in high-impact foundation models and global Earth embedding datasets, a significant barrier remains in translating these academic assets into freely accessible tools. This tutorial introduces EarthEmbeddingExplorer, an interactive web application designed to bridge this gap, transforming static rese

Lisandro Fermin, Ricardo Rios, Luis-Ángel Rodríguez

This paper is the second part of our study on the non-parametric estimation of MS-NAR processes started with [L. Fermin et al. 2017]. We consider the Nadaraya-Watson type regression function estimator for non-linear autoregressive Markov switching processes. In this context the regression function estimator is interpreted as a solution of a local weighted We

Songhuan He, Yifei Cui, Bo Liu, Kai Guo

Superconducting quantum processor units (QPUs) are incapable of producing massive datasets for quantum error correction (QEC) because of hardware limitations. Thus, QEC decoders heavily depend on synthetic data from qubit error models. Classic depolarizing error models with polynomial complexity present limited accuracy. Coherent density matrix methods suffe

Antoine Bottenmuller, Etienne Decencière, Petr Dokládal

Semantic segmentation and hyperspectral unmixing are two central problems in spectral image analysis. The former assigns each pixel a discrete label corresponding to its material class, whereas the latter estimates pure material spectra, called endmembers, and, for each pixel, a vector representing material abundances in the observed scene. Despite their com

Wenli Li, Kai Zhao, Haoran Jiang, Enquan Yang

Vision-language models (VLMs) have been widely adopted for 3D question answering (3D QA). In typical pipelines, visual tokens extracted from multiple viewpoints are concatenated with language tokens and jointly processed by a large language model (LLM) for inference. However, aggregating multi-view observations inevitably introduces severe token redundancy,

Laura Bazahica, Alejandra Avalos-Pacheco, Matthew Moores, Lassi Roininen

Bayesian inference for models with intractable likelihoods, such as Markov random fields, poses a fundamental computational challenge due to the tradeoff between inferential accuracy and computational cost. Various MCMC methods have been developed to address this challenge. The exchange algorithm targets the exact posterior, but requires an expensive perfect

Davide Accadia, Danilo Lewański, Sergej Monavari

We provide a new proof of a result of Bessenrodt on the relation among the generating series of reversed plane partitions and skew plane partitions, motivated by the geometric DT/PT wallcrossing formula for local curves recently proved by the third author. This also recovers a result of Sagan. We moreover establish various new closed formulas for the weighte

Hideki Murakawa, Florian Salin

This paper proposes a novel reaction-diffusion system approximation tailored for singular diffusion problems, typified by the fast diffusion equation. While such approximation methods have been successfully applied to degenerate parabolic equations, their extension to singular diffusion-where the diffusion coefficient diverges at low densities-has remained u

Esperanza Maya-Barbecho, José-Enrique García-Ramos

The study of the structure of odd-mass nuclei in regions characterized by the interplay of multiple particle-hole configurations represents a major challenge in nuclear structure physics. The odd-mass niobium isotopes ($Z = 41$), located near the $N = 60$ region, are of particular interest due to shape coexistence and quantum phase transitions. This work inv

Zhongheng Jiang, Yuechao Zhao, Donglin Xie, Chenxi Sun

Medical time-series data captures the dynamic progression of patient conditions, playing a vital role in modern clinical decision support systems. However, real-world clinical data is highly heterogeneous and inconsistently formatted. Furthermore, existing machine learning tools often have steep learning curves and fragmented workflows. Consequently, a signi

Laura González, Francesc Planas-Vilanova

This paper is intended to give closed formulae for binomial determinants with consecutive or almost consecutive rows or columns, as well as calculating the generator of left nullspaces defined by some binomial matrices. In the meantime, we reprove, by different means, the positivity of binomial determinants shown by Gessel and Viennot.

Federico M. Bandi, Nicola Fusari, Guido Gazzani, Roberto Renò

Options with maturities below one week, hereafter "ultra-short-term" options, have seen a sharp increase in trading activity in recent years. Yet, these instruments are difficult to price jointly using classical pricing models due to the pronounced oscillations observed in the at-the-money implied-volatility term structure across ultra-short-term tenors. We

Yahan Li, Chaohao Du, Zeyang Li, Christopher Chun Kuizon

Mental-health support is increasingly mediated by conversational systems (e.g., LLM-based tools), but users often lack structured ways to audit the quality and potential risks of the support they receive. We introduce CounselReflect, an end-to-end toolkit for auditing mental-health support dialogues. Rather than producing a single opaque quality score, Couns

Xuesong Wang, Harry Wang

Vision-language models (VLMs) exhibit a systematic bias when confronted with classic optical illusions: they overwhelmingly predict the illusion as "real" regardless of whether the image has been counterfactually modified. We present a tool-guided inference framework for the DataCV 2026 Challenge (Tasks I and II) that addresses this failure mode without any

Tillmann Miltzow

We introduce a lightweight and accessible approach to computation over the real numbers, with the aim of clarifying both the underlying concepts and their relevance in modern research. The material is intended for a broad audience, including instructors who wish to incorporate real computation into algorithms courses, their students, and PhD students encount

Jiaao Ma, Chuan Lin, Guangjie Han, Shengchao Zhu

Multi-agent reinforcement learning (MARL) provides a promising solution for cooperative target tracking in networks of autonomous underwater vehicles (AUVs). However, existing methods still face three major challenges: 1) policy non-stationarity caused by concurrent updates among multiple agents; 2) inefficient policy learning caused by the heterogeneous qua

Andrew Baker

The well known Joker $\mathcal{A}(1)$-module of Adams and Priddy is known to be realisable as the cohomology of a $1$-connected space. By attaching an extra cell we obtain an $8$-dimensional Poincar\'e duality space whose mod~$2$ cohomology realising is an unstable $\mathcal{A}$-algebra. We use obstruction theory to show that this admits a $PL$-structure. Al

Tim S. Lyon

This paper develops a novel nested sequent proof-search methodology for intuitionistic tense logics (ITLs), supporting finite counter-model extraction. We introduce a new loop-checking method that detects repeating nested sequents using homomorphisms, thereby bounding the height of derivations during proof-search. Due to the non-invertibility of some inferen

Chenxin Zhu, Yushun Fang, Lu Liu, Shibo Yin

Blind face restoration (BFR) aims to recover high-quality facial images from degraded inputs, yet its inherently ill-posed nature leads to ambiguous and uncontrollable solutions. Recent diffusion-based BFR methods improve perceptual quality but remain uncontrollable, whereas text-guided face editing enables attribute manipulation without reliable restoration

Marina Villanueva, Juan M. Espin, Juan E. Tapia

The integration of multimodal models into Presentation Attack Detection (PAD) for ID Documents represents a significant advancement in biometric security. Traditional PAD systems rely solely on visual features, which often fail to detect sophisticated spoofing attacks. This study explores the combination of visual and textual modalities by utilizing pre-trai

Christoforos Panagiotis, Alexandre Stauffer

We consider a broad class of dependent site-percolation models on $\mathbb{Z}^d$ obtained by applying a monotone automaton to a random initial particle configuration drawn from a stochastically increasing family of measures. We prove that whenever the underlying particle configuration is sampled from an insertion-tolerant measure and the avalanches generated

Qiyuan Zhuang, He-Yang Xu, Yijun Wang, Xin-Yang Zhao

Understanding object affordances is essential for enabling robots to perform purposeful and fine-grained interactions in diverse and unstructured environments. However, existing approaches either rely on retrieval, which is fragile due to sparsity and coverage gaps, or on large-scale models, which frequently mislocalize contact points and mispredict post-con

Meiwen Ding, Song Xia, Chenqi Kong, Xudong Jiang

Although multimodal large language models (MLLMs) are increasingly deployed in real-world applications, their instruction-following behavior leaves them vulnerable to prompt injection attacks. Existing prompt injection methods predominantly rely on textual prompts or perceptible visual prompts that are observable by human users. In this work, we study imperc

Guillaume Maillard, Adrien Saumard

Considering a general framework of bootstrap with exchangeable weights, we show some concentration inequalities for the supremum of the bootstrap empirical process. On the one hand, we discuss the concentration of the bootstrap empirical process around its conditional expectation with respect to the original data, and on the other hand, the concentration of

Ni Ou, Zhuo Chen, Xinru Zhang, Junzheng Wang

Accurate camera-LiDAR fusion relies on precise extrinsic calibration, which fundamentally depends on establishing reliable cross-modal correspondences under potentially large misalignments. Existing learning-based methods typically project LiDAR points into depth maps for feature fusion, which distorts 3D geometry and degrades performance when the extrinsic

Sujoy Majumder, Abhijit Banerjee, Shantanu Panja

This paper investigates certain classes of entire functions in C^n that, together with their partial derivatives, share a finite set consisting of three elements. By employing normality criteria, we study the behaviour of such functions and derive the necessary conditions governing their existence. Our results extend those of [4], originally established for

Zeyu Chen, Junde Wu

A basic question in the study of measure-once quantum finite automata is whether two distinct input words can be separated with certainty. The exact separation problem reduces to a trace-vanishing question in \(SU(2)\). The main difficulty lies in the genuinely nonabelian regime, where \(u\) and \(v\) have the same abelianization. This paper develops a slice

Yubo Cui, Xianchao Guan, Zijun Xiong, Zheng Zhang

Pre-trained vision-language models (VLMs) exhibit strong zero-shot generalization but remain vulnerable to adversarial perturbations. Existing classification-guided adversarial fine-tuning methods often disrupt pre-trained cross-modal alignment, weakening visual-textual correspondence and degrading zero-shot performance. In this paper, we propose an Alignmen

Andrew Jeong, Jaemin Kim, Sebin Lee, Sung-Eui Yoon

Robotic manipulation involves kinematic and semantic transitions that are inherently coupled via underlying actions. However, existing approaches plan within either semantic or latent space without explicitly aligning these cross-modal transitions. To address this, we propose CLaD, a framework that models how proprioceptive and semantic states jointly evolve

I-Chun Cho, Chien-Hsu Chen, Huan Niu, Cheng-Ya Pan

A fast-reading current integrator is developed for high time-resolution and low-noise ion beam diagnostics under both continuous-wave and pulsed operating conditions. The system combines a low-leakage transimpedance front-end with a hybrid digitization architecture based on charge-balancing and voltage-to-frequency conversion. The input current is converted

Fabrizio Pittorino, Manuel Roveri

Edge AI is often framed as model compression and deployment under tight constraints. We argue a stronger operational thesis: Edge AI in realistic deployments is necessarily adaptive. In long-horizon operation, a fixed (non-adaptive) configuration faces a fundamental failure mode: as data and operating conditions evolve and change in time, it must either (i)

Fu Wang, Qifeng Lu, Xinyu Long, Meng Zhang

Accurate forecasting of three-dimensional (3D) cloud fields is important for atmospheric analysis and short-range numerical weather prediction, yet it remains challenging because cloud evolution involves cross-layer interactions, nonlocal dependencies, and multiscale spatiotemporal dynamics. Existing spatiotemporal prediction models based on convolutions, re

Tal Ishon, Yoav Goldberg, Uri Shaham

Topic modeling seeks to uncover latent semantic structure in text, with LDA providing a foundational probabilistic framework. While recent methods often incorporate external knowledge (e.g., pre-trained embeddings), such reliance limits applicability in emerging or underexplored domains. We introduce \textbf{PRISM}, a corpus-intrinsic method that derives a D

Wei Suo, Hanzu Zhang, Lijun Zhang, Ji Ma

Large Vision-Language Models have demonstrated exceptional performance in multimodal reasoning and complex scene understanding. However, these models still face significant hallucination issues, where outputs contradict visual facts. Recent research on hallucination mitigation has focused on retraining methods and Contrastive Decoding (CD) methods. While bot

Zhuoyi Fang, Kexuan Shi, Jiajia Liu, Qiang Han

Medical image segmentation is of great significance in analysis of illness. The use of deep neural networks in medical image segmentation can help doctors extract regions of interest from complex medical images, thereby improving diagnostic accuracy and enabling better assessment of the condition to formulate treatment plans. However, most current medical im

Aiman Al Masoud, Antony Anju, Marco Arazzi, Mert Cihangiroglu

LLM-as-a-Judge (LaaJ) is a novel paradigm in which powerful language models are used to assess the quality, safety, or correctness of generated outputs. While this paradigm has significantly improved the scalability and efficiency of evaluation processes, it also introduces novel security risks and reliability concerns that remain largely unexplored. In part

Angelo Raffaele Fazio, Adam Smetana

We present a novel proposal for the effective Lagrangian of the low-energy Yang--Mills quantum field theory. The proposed effective Lagrangian exhibits the spontaneous BRST symmetry breaking. We built the Fujikawa model that we couple to the Yang--Mills elementary field sector, motivated by the analogy with Chiral Quark Model. We interpret the Fujikawa field

Christopher Zanoli, Andrea Giovannini, Tengjun Jin, Ana Klimovic

Constructing Extract-Load-Transform (ELT) pipelines is a labor-intensive data engineering task and a high-impact target for AI automation. On ELT-Bench, the first benchmark for end-to-end ELT pipeline construction, AI agents initially showed low success rates, suggesting they lacked practical utility. We revisit these results and identify two factors causing

Davide Zanchetta, Vittoria Bettio, Sandro Azaele, Manlio De Domenico

The spread of infectious disease and the evolution of antigenically distinct strains are often modeled separately, despite strong feedbacks mediated by host immune memory and heterogeneous contacts. To tackle this challenging problem, we introduce a coevolutionary framework in which transmission occurs on a metapopulation network while mutational exploration

Renaud Detcherry

A conjecture of Andersen, Masbaum and Ueno states that for any compact oriented surface $\Sigma_{g,n}$ and any pseudo-Anosov $f\in \mathrm{Mod}(\Sigma_{g,n}),$ the matrix $\rho_r(f)$ has infinite order for any large $r,$ where $\rho_r$ is the $\mathrm{SO}(3)$-WRT quantum representation of the mapping class group $\mathrm{Mod}(\Sigma_{g,n})$ at a primitive $r

Zoë Prins, Samuele Punzo, Frank Wildenburg, Giovanni Cinà

Standard evaluations of Large language models (LLMs) focus on task performance, offering limited insight into whether correct behavior reflects appropriate underlying mechanisms and risking confirmation bias. We introduce a simple, principled interpretability framework based on token-level perplexity to test whether models rely on linguistically relevant cue

Aman Desai, Kristin Lohwasser, Mohamed Ouchemhou, Tania Robens

We study the production of Beyond the Standard Model light scalar states in association with a vector boson ($Vh_2$, with $V = W^\pm, Z$) at the LHC. We consider the scenario where the Standard Model scalar sector is extended by two real scalar singlets, where these additional scalars have mass $ M_i \leq M_{h_{125}}$. In this work, the scalar boson $h_2$ de

Taewoo Suh, Sungpyo Kim, Jongmin Park, Munchurl Kim

Feed-forward 3D Gaussian Splatting (FF-3DGS) emerges as a fast and robust solution for sparse-view 3D reconstruction and novel view synthesis (NVS). However, existing FF-3DGS methods are built on incorrect screen-space dilation filters, causing severe rendering artifacts when rendering at out-of-distribution sampling rates. We firstly propose an FF-3DGS mode

Yihan Liu, Meiqi Tian, Teng Yan, Bingzhuo Zhong

Networked control systems (NCS) are widely used in safety-critical applications, but they are often analyzed under the assumption of ideal communication channels. This work focuses on the synthesis of safety controllers for discrete-time linear systems affected by unknown disturbances operating in imperfect communication channels. The proposed method guarant

Max Lodel, Nils Wilde, Robert Babuška, Javier Alonso-Mora

The use of semantic features can improve the efficiency of target search in unknown environments for robotic search and rescue missions. Current target search methods rely on training with large datasets of similar domains, which limits the adaptability to diverse environments. However, human experts possess high-level knowledge about semantic relationships

Hiroki Isobe

The quantum Hall effect realizes a quantized Hall resistance $R_{xy} = h/(\nu e^2)$ whereas the longitudinal resistance vanishes. The quantized value consists of the fundamental physical quantities, the elementary charge $e$ and the Planck constant $h$, along with an integer or fractional constant $\nu$. High precision measurements of $R_{xy}$ allude to a li

Renaud Baillou, Matthew Terkel, Cristiano Nisoli, Pietro Tierno

Artificial spin ices have transcended their origins in frustrated rare-earth pyrochlores to become a versatile platform for engineering exotic states of matter. Across diverse implementations, from nanomagnets and superconducting vortices to colloids, quantum annealers, liquid crystals, and metamaterials, they are unified by the ice rule, which often leads t

Jose Antonio Rebollo, Rafael Vazquez, Claudio Bombardelli

Non-Gaussian tails dominate collision probability estimates in conjunction assessment, yet capturing them without Monte Carlo sampling is challenging, especially when process noise is included. We present a closed-form, grid-free solution to the Fokker-Planck equation by proving that an exponential-of-quadratic-form ansatz is structurally preserved under adv

Seungwoo Yoon, Jinmo Kim, Jaesik Park

In this paper, we propose Extend3D, a training-free pipeline for 3D scene generation from a single image, built upon an object-centric 3D generative model. To overcome the limitations of fixed-size latent spaces in object-centric models for representing wide scenes, we extend the latent space in the $x$ and $y$ directions. Then, by dividing the extended late

Jianpeng Wang, Haoyu Wang, Baoying Chen, Jishen Zeng

The rapid democratization of prompt-based AI image editing has recently exacerbated the risks associated with malicious content fabrication and misinformation. However, forgery localization methods targeting these emerging editing techniques remain significantly under-explored. To bridge this gap, we first introduce a fully automated mask annotating framewor

Sina Jorjani, Caglar Tunc, Ozgur Gurbuz, Akhtar Saeed

Accurate characterization of Terahertz (THz) band path loss is critical for reliable high-frequency communication, especially in aerial networks where transceivers may operate at different altitudes. Existing THz-band path loss models for aerial networks focus on horizontal or vertical transceiver deployments, and fall short at modeling the random 3D geometr

Yuxuan Liu, Wenchao Xu, Haozhao Wang, Zhiming He

Federated Graph Learning (FGL) has emerged as a powerful paradigm for decentralized training of graph neural networks while preserving data privacy. However, existing FGL methods are predominantly designed for static graphs and rely on parameter averaging or distribution alignment, which implicitly assume that all features are equally transferable across cli

Nils Lid Hjort, Stephen G. Walker

We show that the cumulative distribution function corresponding to a kernel density estimator with optimal bandwidth lies outside any confidence interval, around the empirical distribution function, with probability tending to 1 as the sample size increases.

Tom-Lukas Breitkopf, Vincent Froese, Anton Herrmann, André Nichterlein

We consider the well-studied problem of finding a spanning tree with minimum average distance between vertex pairs (called a MAD tree). This is a classic network design problem which is known to be NP-hard. While approximation algorithms and polynomial-time algorithms for some graph classes are known, the parameterized complexity of the problem has not been

Kaustubh Kartikey, Shalabh Bhatnagar

We present a finite-time analysis of two smoothed functional stochastic approximation algorithms for simulation-based optimization. The first is a two time-scale gradient-based method, while the second is a three time-scale Newton-based algorithm that estimates both the gradient and the Hessian of the objective function $J$. Both algorithms involve zeroth or

Jaroslav Kysela, Katharina Ludwig, Nitica Sakharwade, Anette Messinger

We define the class of register-logic graphs and prove that any uniformly deterministic measurement-based quantum computation (MBQC) where the inputs coincide with the outputs must be driven on such graphs by measurements in the $YZ$ plane of the Bloch sphere. This observation is revisited in the context that goes beyond uniform determinism, where we present