March 2026 arXiv papers — page 122

Luís Freire, Fernanda A. Andaló, Nicki Skafte Detlefsen

Large language models (LLMs) have demonstrated strong capabilities in generating executable code from natural language descriptions. However, general-purpose models often struggle in specialized programming contexts where domain-specific libraries, APIs, or conventions must be used. Customizing smaller open-source models offers a cost-effective alternative t

Ruo-Yu Shang, Karl Kosack, Brian Humensky

Analyses of Imaging Atmospheric Cherenkov Telescope (IACT) data for extended $\gamma$-ray sources face the issue that the field of view does not offer sufficient regions for background estimations. In cases where the source angular size exceeds or occupies a significant part of the field of view, an independent background estimation method is necessary to ca

Shokoufeh Naderi, Maude Blondin, Sébastien Roy

This paper analyzes consensus in multi-agent systems under uniform and nonuniform communication delays, a key challenge in distributed coordination with applications to robotic swarms. It investigates the convergence of a consensus algorithm accounting for delays across communication links in a connected, undirected graph. Novel convergence results are deriv

Rudranarayan Padhy, Sudhansu Sekhar Rout

Let $K$ be a number field with algebraic closure $\overline{K}$ and let $S$ be a finite set of places of $K$ that contain all the archimedean places. For an integer $d \ge 2$, consider the unicritical polynomial family $f_{d,c}(z) = z^d + c$. Recently, Benedetto and Ih studied the distribution of post-critically finite parameters $c$ that are $S$-integral re

Colin Rogers, Pablo Amster

Here, a novel 2+1-dimensional nonlinear evolution equation with temporal modulation is introduced which admits integrable Ermakov-Painlev\'e II symmetry reduction. Application is made to obtain exact solution to a class of Stefan-type moving boundary problems for this 2+1-dimensional nonlinear evolution equation. Involutory transformations with origin in aut

Jaroslaw Wojtun, Cezary Ziolkowski, Jan M. Kelner, Pawel Skokowski

Empirical path loss models are defined for a specific antenna system used during measurements and characterized by a particular radiation pattern and main lobe beam width. In this paper, we propose a novel approach to modifying such a model to estimate path loss for antenna systems with different radiation patterns and beam widths. This method is based on a

Doyon Kim, Youngmin Lee

We prove the quantum ergodicity of Eisenstein series on the arithmetic hyperbolic 3-manifold $\operatorname{PSL}_2(\mathcal{O}_F)\backslash \mathbb{H}^3$, where $F$ is an imaginary quadratic field with ring of integers $\mathcal{O}_F$ and class number $h_F\geq 1$. This extends the work of Koyama, who proved the result in the case $h_F=1$, and establishes the

Gabriel Knotz, Till M. Muenker, Timo Betz, Matthias Krüger

We investigate signatures of broken time reversal symmetry in stochastic trajectory data, employing the previously introduced three point correlation called mean back relaxation. We specifically investigate data from a simple driven model, as well as from colloidal particles within living or passivated biological cells. Both in the model as well as in cell d

Otmar Scherzer, Cong Shi, Thi Lan Nhi Vu

Chan-Vese algorithms have proven to be a first-class method for image segmentation. Early implementations used level set methods with a pixelwise representation of the level set function. Later, parametrized level set approximations, such as splines, have been studied and computationally developed to improve efficiency. In this paper, we use neural networks

Huamin Chen, Xunzhuo Liu, Yuhan Liu, Junchen Jiang

Modern LLM GPU fleets are provisioned for worst-case context lengths that the vast majority of requests never approach, wasting GPU capacity on idle KV-cache slots. We present FleetOpt, a framework that starts from first principles: given a workload's prompt-length CDF and a P99 TTFT target, derive the minimum-cost fleet analytically, then deploy it in pract

Dario Fasone, Rita Veilande, Luigi Giannelli, Giuseppe A. Falci

We investigate the role of the global driving phase $\Phi$ in the dynamics of driven few-level quantum systems, a central setting in coherent control of atomic, molecular, and solid-state platforms. In particular, we focus on systems with closed-loop couplings, where external driving fields induce interference effects that strongly influence population trans

A. V. Tutukov, A. V. Sobolev

The paper is devoted to the consideration of the role of the donor stellar wind in the matter exchange between the components of detached binary systems. A classification of close binary systems with interacting components is proposed. A list of potential donors and accretors of such systems, including X-ray binary and symbiotic stars, is given. Analytical t

Sarita de Berg, Joachim Gudmundsson, Peter Kramer, Christian Rieck

We investigate multiple fundamental variants of the classic coordinated motion planning (CMP) problem for unit square robots in the plane under the $L_1$ metric. In coordinated motion planning, we are given two arrangements of $k$ robots and are tasked with finding a movement schedule that minimizes a certain objective function. The two most prominent object

Jacek Dziarmaga

Quantum annealing with the D-Wave Advantage system in the random Ising model on a cubic lattice is simulated using a three-dimensional (3D) tensor network. The Hamiltonian is driven across a quantum phase transition from a paramagnetic phase to a spin-glass phase. The network is represented as a tensor product state, also known-particularly in two dimensions

Shuo Xiao, Shuang-Nan Zhang

A nearby primordial-black-hole (PBH) evaporation burst would produce a curved gamma-ray wavefront, leading to detectable departures from plane-wave inter-satellite delays. We introduce a purely geometric method that combines imaging localizations with multi-spacecraft timing to determine the distance of a gamma-ray transient. Applied to \textit{Swift}-locali

Nicoletta Cantarini, Fabrizio Caselli, Victor Kac

In this paper we classify degenerate Verma modules over the linearly compact Lie superalgebra $E(4,4)$. This completes the description of Verma modules over the exceptional linearly compact Lie superalgebras. As in the other cases all degenerate modules and morphisms between them give rise to infinite bilateral complexes which may be viewed as a generalizati

Adrian Göß

Solving mixed-integer nonlinear programs (MINLPs) typically relies on constructing relaxations that are easier to tackle than the original problem. Recently, global parabolic (PARA) relaxations were introduced, featuring separable quadratic functions -- paraboloids -- as global under- or overestimators of general nonlinear constraint functions. So far, the p

Pierre Krack, Tobias Jülg, Wolfram Burgard, Florian Walter

Well-designed dense reward functions in robot manipulation not only indicate whether a task is completed but also encode progress along the way. Generally, designing dense rewards is challenging and usually requires access to privileged state information available only in simulation, not in real-world experiments. This makes reward prediction models that inf

Jianquan Ge, Fagui Li, Yunheng Zhang

Let $M^n$ be an $n$-dimensional closed minimal submanifold immersed in the unit sphere $\mathbb{S}^{n+m}$. Denote by $S$ and $\rho^{\perp}$ the squared norm of the second fundamental form and the normal scalar curvature of $M^n$, respectively. Let $\{A^{\alpha}\}_{\alpha=n+1}^{n+m}$ be the shape operators of $M^n$ with respect to a local orthonormal normal f

Lauren Ervin, Harish Bezawada, Vishesh Vikas

Conventional mobile tensegrity robots constructed with straight links offer mobility at the cost of locomotion speed. While spherical robots provide highly effective rolling behavior, they often lack the stability required for navigating unstructured terrain common in many space exploration environments. This research presents a solution with a semi-circular

Zeyang Mou, Haijie Ren, Ding Xu, Igor S. Aranson

There is a recent interest in studying odd elasticity in soft solids. Current focus has been on simple solids. However, many soft solids are structured and can exhibit nematic elasticity or viscoelasticity. Here we generalize the concept of odd elasticity to nematic elasticity. By rewriting the governing equation for two-dimensional nematic liquid crystals (

Boo Carmans, Michael Petrov, Milos Nesladek

Quantum computing is a rapidly developing field. However, the most commonly used qubits require cryogenic conditions to operate, which increases the costs and puts constraints on the up-scaling. Ambient solid-state qubits provide an alternative with potential for large-scale application. The nitrogen-vacancy (NV) center in diamond is one of the main candidat

Yu Liu, Jiayang Xu, Tianzi Wang, Zichen Shi

Kangaroo mother care (KMC) is an intervention involving skin-to-skin contact that promotes physiological stability and supports long-term neurodevelopment in preterm infants. However, the underlying neurophysiological mechanisms remain unclear. We aimed to investigate the immediate effects of the first KMC on infants' brain function, mother-infant inter-brai

Xizhong Yang, Yinan Xia, Huiming Wang, Mofei Song

Leveraging the model's internal information as the self-reward signal in Reinforcement Learning (RL) has received extensive attention due to its label-free nature. While prior works have made significant progress in applying the Test-Time Scaling (TTS) strategies to RL, the discrepancy in internal information between test and training remains inadequately ad

Jia Liu, Li Ma, Wei Meng

Let $G$ be a finite $p$-group and $\delta(G)$ denote the number of all non-cyclic subgroups of $G$. In this paper, an upper bound for $\delta(G)$ is obtained. Furthermore, we prove that $\delta(G)\leq \delta(M_p(1, 1, 1) \times C_{p}^{n-3})$ (if $p=2$, then $\delta(G)\leq \delta(D_8\times C_{2}^{n-3})$), for any non-elementary abelian $p$-group $G$ of order

Subina Khanal, Seshu Tirupathi, Merim Dzaferagic, Marco Ruffini

Time series foundation models (TSFMs) require diverse, real-world datasets to adapt across varying domains and temporal frequencies. However, current large-scale datasets predominantly focus on low-frequency time series with sampling intervals, i.e., time resolution, in the range of seconds to years, hindering their ability to capture the nuances of high-fre

Shannan Yan, Jingchen Ni, Leqi Zheng, Jiajun Zhang

Large language model (LLM) agents increasingly rely on external memory to support long-horizon interaction, personalized assistance, and multi-step reasoning. However, existing memory systems still face three core challenges: they often rely too heavily on semantic similarity, which can miss evidence crucial for user-centric understanding; they frequently st

Zihe Wang, Yihuan Wang, Haiyang Yu. Zhiyong Cui, Xiaojian Liao

The current expressway operation relies on rule-based and isolated models, which limits the ability to jointly analyze knowledge across different systems. Meanwhile, Large Language Models (LLMs) are increasingly applied in intelligent transportation, advancing traffic models from algorithmic to cognitive intelligence. However, general LLMs are unable to effe

Poul H. Damgaard, Hojin Lee, Kanghoon Lee, Tabasum Rahnuma

Solving the classical equations of motion in general relativity recursively, we consider the metric of a spatially localized and stationary source of matter. Having in mind a star of general composition, we characterize it by means of its infinite set of mass and current multipoles. Specializing to de Donder gauge we set up the recursive equations that produ

Tony J. Puthenpurakal

Let $\mathbb{F}_q$ be a finite field with $q = p^s$ elements. Let $V$ be a $d$ dimensional vector space over $\mathbb{F}_q$ and let $G$ be a subgroup of $GL(V)$. Let $R = \mathbb{F}_q[V] = \text{Sym}_{\mathbb{F}_q}(V^*)$ and let $G$ act naturally on $R$. Set $S = R^G$. Let $\mathbf{d}_{d,0}, \mathbf{d}_{d, 1}, \ldots, \mathbf{d}_{d, d-1} \in S$ be the Dickso

Alexander Zuepke, Ashutosh Pradhan, Daniele Ottaviano, Andrea Bastoni

The Embedded Trace Macrocell (ETM) is a standard component of Arm's CoreSight architecture, present in a wide range of platforms and primarily designed for tracing and debugging. In this work, we demonstrate that it can be repurposed to implement a novel hardware-assisted memory bandwidth regulator, providing a portable and effective solution to mitigate mem

Hyundo Choi, Junhyeong An, Jinseong Park, Jaewoong Choi

Recent advances in one-step generative frameworks, such as flow map models, have significantly improved the efficiency of image generation by learning direct noise-to-data mappings in a single forward pass. However, machine unlearning for ensuring the safety of these powerful generators remains entirely unexplored. Existing diffusion unlearning methods are i

Allan McRobie

In Part 3 of this sequence of papers, the kinematic behaviour of 3D frame structures is described using the loop formalism that was developed in Part 2 to describe equilibrium. There, the notions of polygons, polyhedra and polytopes that form the geometric toolbox underlying graphic statics were replaced by the more general concept of CW-complexes from algeb

Gayathrini Premawardhana, Jonathan Beaumariage, M. V. Gurudev Dutt, David Pekker

We propose a system to achieve sub-zeptonewton force sensing and robust spin-mechanical entanglement in a levitated diamond system. By coupling a nitrogen-vacancy (NV) center spin to the motion of its host diamond within a magnetic trap, we develop a platform designed to surpass the standard quantum limit. We develop and compare three distinct pulse sequence

Marco Quantschnig, Yannick Werner, Thomas Klatzer, Sonja Wogrin

This paper presents a comprehensive, spatially disaggregated dataset of Austria's natural gas and hydrogen infrastructure towards 2040. The dataset covers the complete gas transmission and distribution networks down to the medium-pressure level and integrates hydrogen expansion plans from the Austrian Gas Grid Management. Transmission infrastructure is recon

Fumi Yoshida, Tsuyoshi Terai, Keiji Ohtsuki

We conducted a two-band imaging survey observation using the Subaru Telescope and its wide-field camera, Suprime-Cam, to study the visible colors and size distribution of Jupiter's Trojan asteroids. The survey covered an area around Jupiter's L4 Lagrange point totaling 9.2 square degrees. We detected 120 Trojan asteroids in this survey. From these Trojan ast

Sa Xiao, Hao Jing, Honglu Sun, Haoyu Li

This paper presents TorchNWP, a compilation library tool for the efficient coupling of artificial intelligence components and traditional numerical models. It aims to address the issues of poor cross-language compatibility, insufficient coupling flexibility, and low data transfer efficiency between operational numerical models developed in Fortran and Python

Yuan Ge, Haishu Zhao, Aokai Hao, Junxiang Zhang

Emotion is a core paralinguistic feature in voice interaction. It is widely believed that emotion understanding models learn fundamental representations that transfer to synthesized speech, making emotion understanding results a plausible reward or evaluation metric for assessing emotional expressiveness in speech synthesis. In this work, we critically exami

Michael Bidollahkhani, Freja Nordsiek, Julian M. Kunkel

GPU nodes are central to modern HPC and AI workloads, yet many failures do not manifest as immediate hard faults. While some instabilities emerge gradually as weak thermal or efficiency drift, a significant class occurs abruptly with little or no numeric precursor. In these detachment-class failures, GPUs become unavailable at the driver or interconnect leve

Yicui Shi, Yuhan Chen, Xiangfei Huang, Zhenguo Wang

Low-light image enhancement aims to restore the visibility of images captured by visual sensors in dim environments by addressing their inherent signal degradations, such as luminance attenuation and structural corruption. Although numerous algorithms attempt to improve image quality, existing methods often cause a severe loss of intrinsic signal priors. To

Kensuke Ishitani

The Ziggurat method is an efficient rejection sampling technique for generating one-dimensional normally distributed random numbers. This study proposes the pattern block method, a generalization of the Ziggurat method. The pattern block method enables the generation of random numbers from multimodal density functions and multidimensional distributions. The

Mohamed A. Mabrok

Large Language Models (LLMs) perform internal computations in continuous vector spaces yet produce discrete tokens -- a fundamental mismatch whose geometric consequences remain poorly understood. We develop a mathematical framework that interprets LLM hidden states as points on a latent semantic manifold: a Riemannian submanifold equipped with the Fisher inf

Zhihao Gong, Zeyu Sun, Dong Huang, Qingyuan Liang

While Large Language Models (LLMs) have substantially improved the functional correctness of code translation, the critical dimension of \textit{execution efficiency} remains overlooked. We present \textbf{\textsc{trace}}, the first benchmark to explicitly assess efficiency in LLM-translated code. \textsc{trace} includes 1,000 efficiency-critical tasks acros

Hui Zhang, Jincheng Yue, Jiongzhi Zheng, Ning Wang

Defective chalcopyrites have recently emerged as promising thermoelectric materials because their ordered intrinsic vacancies can profoundly reshape both lattice dynamics and electronic structure. Here, we present a comprehensive theoretical investigation of the lattice thermal and carrier transport properties of II-III$_2$-VI$_4$ defective chalcopyrites by

A. Arbieto, W. Britto, C. A. Morales, E. Rego

For any integer $n \geq 5$, we construct an $n$-dimensional $C^1$ vector field exhibiting a robustly transitive singular attractor which is not sectional-hyperbolic. Nevertheless, the attractor is singular-hyperbolic. This provides the first such examples improving some features of the constructions in [17, 32].

Gregor Hieronymus Eberwein, Gianluca Aglieri Rinella, Daniela Bortoletto, Szymon Bugiel

For the Inner Tracking System 3 (ITS3) upgrade, the ALICE experiment at CERN requires monolithic active pixel sensors of dimensions up to 97~mm$\,\times\,$266~mm, occupying a large fraction of a 300 mm wafer. To manufacture such a wafer-scale device, larger than the single design reticle size, stitching is employed. The MOnolithic Stitched Sensor (MOSS) is a

Vincent Gurgul, Robin Gubela, Stefan Lessmann

Generative Artificial Intelligence (GenAI) rapidly transforms software engineering, yet existing research remains fragmented across individual tasks in the Software Development Lifecycle. This study integrates a systematic literature review with a survey of 65 software developers. The results show that GenAI exerts its highest impact in design, implementatio

Wei Xu, Lipeng Zhu, Wenyan Ma, An Liu

In conventional antenna arrays, mutual coupling between antenna elements is often regarded as detrimental. However, under specific conditions, it can be harnessed to enhance the far-field directivity (i.e., beamforming gain). Theoretically, the directivity of an N-antenna superdirective array over the endfire direction can reach N^{2}, significantly exceedin

Joshua Raymond Bettles, Jiaxu Wu, Bruno Vilhena Adorno, Joaquin Carrasco

Robotic inspection of radioactive areas enables operators to be removed from hazardous environments; however, planning and operating in confined, cluttered environments remain challenging. These systems must autonomously reconstruct the unknown environment and cover its surfaces, whilst estimating and avoiding collisions with objects in the environment. In t

Marios Aristodemou, Yasaman Omid, Sangarapillai Lambotharan, Mahsa Derakhshan

The integration of satellite communication networks with next-generation (NG) technologies is a promising approach towards global connectivity. However, the quality of services is highly dependant on the availability of accurate channel state information (CSI). Channel estimation in satellite communications is challenging due to the high propagation delay be

Yipeng Xie, Mingwei Lei, Wenbo Dong, Meng Shi

This study demonstrates a significant enhancement in ultra-low-frequency (ULF) electric field sensitivity using Rydberg atoms via an optical chopping amplification (OCA) technique. Conventional Rydberg-based ULF measurements are fundamentally limited by 1/f noise, which severely degrades sensitivity. Our approach modulates the coupling laser with an optical

Klaudiusz Czudek, Jacopo De Simoi, Andrew Gad, Marco Poon

In this short paper, we show a characterization of Birkhoff Billiards inside discs which is related to the expansion of the formal Lazutkin conjugacy at the boundary.

Yongtao Wang

This paper investigates tangent measures in the sense of Preiss for self-similar sets on ${{\mathbb{R}}^d}$ that satisfy the strong separation condition. Through the dynamics of ``zooming in'' on any typical point, we derive an explicit and uniform formula for the tangent measures associated with this category of self-similar sets on ${{\mathbb{R}}^d}$. Furt

R. B. Koshi, M. Doi, S. Sako, K. Maeda

Attempts to reveal the spectroscopic diversity of Type Ia supernovae (SNe Ia) have led to subclassification schemes such as the Branch system, which classifies SNe Ia into four categories: core normal (CN), broad line (BL), cool (CL), and shallow silicon (SS). The physical origin of these spectroscopic differences, including progenitor channels, explosion me

Zhong-Xuan Mao, Jing-Feng Tian

In this paper, we investigate the recurrence relations for the Maclaurin coefficients of the products of elementary functions and hypergeometric functions. Specifically, we focus on the confluent hypergeometric function $\mathcal{M}(z) = h(z) M(a,c;z)$ and the Gaussian hypergeometric function $\mathcal{F}(z) = h(z) F(a,b;c;z)$, considering several specific c

Shubham Kanodia, Caleb I. Cañas, Jacob Lustig-Yaeger, Giannina Guzmán Caloca

The growing sample of Giant Exoplanets around M-dwarf Stars (GEMS) helps probe the extremes of giant planet formation. Comparing the properties of this sample with their FGK counterparts can help us understand how planet formation and migration depend on stellar mass. We initiated a large Cycle 2 JWST transmission spectroscopy survey of seven GEMS. Here we p

Yu Liu, Lei Zhang, Haoxun Li, Hanlei Shi

Open-Vocabulary Multimodal Emotion Recognition (OV-MER) is inherently challenging due to the ambiguity of equivocal multimodal cues, which often stem from distinct unobserved situational dynamics. While Multimodal Large Language Models (MLLMs) offer extensive semantic coverage, their performance is often bottlenecked by premature commitment to dominant data

Daniel Windhager, Bernhard A. Moser, Michael Lunglmayr

Spiking Neural Networks (SNNs) offer an energy efficient alternative to conventional Artificial Neural Networks (ANNs) but typically still require a large number of parameters. This work introduces Linearized Bregman Iterations (LBI) as an optimizer for training SNNs, enforcing sparsity through iterative minimization of the Bregman distance and proximal soft

Svetoslav Nizhnichenkov, Rahul Nair, Elizabeth Daly, Brian Mac Namee

We investigate how successful bias mitigation reshapes the embedding space of encoder-only and decoder-only foundation models, offering an internal audit of model behaviour through representational analysis. Using BERT and Llama2 as representative architectures, we assess the shifts in associations between gender and occupation terms by comparing baseline an

Wellars Banzi, Froduald Minani, Solange Mukeshimana, David Rule

We prove pointwise bounds for rough Fourier integral operators by the $L^p$ Hardy-Littlewood maximal function. We assume the Fourier integral operators have amplitudes in $L^\infty S^m_ρ$ and phases $φ$ such that $φ(x,ξ) - x\cdotξ\in L^\infty Φ^1$, and assume a non-degeneracy condition on the matrix $\partial^2_ξφ(x,ξ)$. The pointwise bound holds when \begin

Pengyu Zhang, Klim Zaporojets, Jie Liu, Jia-Hong Huang

Multi-Modal Knowledge Graphs (MMKGs) benefit from visual information, yet large-scale image collection is hard to curate and often excludes ambiguous but relevant visuals (e.g., logos, symbols, abstract scenes). We present Beyond Images, an automatic data-centric enrichment pipeline with optional human auditing. This pipeline operates in three stages: (1) la

Yanyu Qian, Yue Tan, Yixin Liu, Wang Yu

Diffusion large language models (D-LLMs) have emerged as a promising alternative to auto-regressive models due to their iterative refinement capabilities. However, hallucinations remain a critical issue that hinders their reliability. To detect hallucination responses from model outputs, token-level uncertainty, such as entropy, has been widely used to indic

Linghao Zhang, Haitao Zhao, Bo Xu, Hongbo Zhu

Space-air-ground integrated networks (SAGIN) promise ubiquitous 6G connectivity but face significant resource management challenges due to heterogeneous infrastructure, dynamic topologies, and stringent quality-of-service (QoS) requirements. Conventional model-driven approaches struggle with scalability and adaptability in such complex environments. This pap

Maria Vrellou, Alexander Welle, Stefan Wagner, Marco Weber

Accurate nanoscale detection of hydrogen is essential for understanding hydrogen-related phenomena in materials, yet conventional deuterium tracing is often complicated by residual background hydrogen. This study evaluates tritium as an unambiguous isotopic marker for nanoscale hydrogen analysis in metals using atom probe tomography (APT). Titanium was selec

Francis J. Headley

We derive the quantum Fisher information for entropy estimation in a Gibbs state and show that it equals the inverse of the heat capacity, which is dual to the temperature Fisher information given by the heat capacity divided by the square of the temperature. Their product is independent of the Hamiltonian and depends only on the temperature, leading to a me

Weiqing Li, Jinyue Guo, Yaqi Wang, Haiyang Xiao

Visual-language models (VLMs) excel at data mappings, but real-world document heterogeneity and unstructuredness disrupt the consistency of cross-modal embeddings. Recent late-interaction methods enhance image-text alignment through multi-vector representations, yet traditional training with limited samples and static strategies cannot adapt to the model's d

Tom Bohman, Marcus Michelen, Dhruv Mubayi

For $r \ge 2$ and a graph $G$, let $\alpha_{{r}}(G)$ be the maximum number of vertices in a $K_r$-free subgraph of $G$. We investigate the value $\alpha_{r}(G)$ when $G$ is the random graph $G \sim G_{n, 1/2}$ and discover the following phenomenon: with high probability, $\alpha_r(G)$ lies in an interval of constant length that varies in a non-monotonic fash

Zijian Wang, Jeremy G. Hoskins, Manas Rachh, Alex H. Barnett

In this paper, we study the scale-invariant quantity \[\mathcal{G}(\Omega)=\frac{\|\partial_n u_1\|_{L^\infty(\partial\Omega)}}{\lambda_1},\]where $u_1$ is the first $L^2$-normalized Dirichlet Laplace eigenfunction of a Euclidean domain $\Omega$ and $\lambda_1$ is its eigenvalue. This is related to the peak boundary heat flux in the long time limit. For conv

Yikun Wang, Yang Li, Zeyi Ren, Jingreng Lei

The recently emerged movable antenna (MA) and fluid antenna technologies offer promising solutions to enhance the spatial degrees of freedom in wireless systems by dynamically adjusting the positions of transmit or receive antennas within given regions. In this paper, we aim to address the joint optimization problem of antenna positioning and beamforming in

Ai Jian, Xiaoyun Zhang, Eryu Guo, Wanrou Du

Text-to-SQL parsing has achieved remarkable progress under the Full Schema Assumption. However, this premise fails in real-world enterprise environments where databases contain hundreds of tables with massive noisy metadata. Rather than injecting the full schema upfront, an agent must actively identify and verify only the relevant subset, giving rise to the

Kaiwen Song, Jinkai Cui, Juyong Zhang

In practical real-time XR and telepresence applications, network and computing resources fluctuate frequently. Therefore, a progressive 3D representation is needed. To this end, we propose ProgressiveAvatars, a progressive avatar representation built on a hierarchy of 3D Gaussians grown by adaptive implicit subdivision on a template mesh. 3D Gaussians are de

Xinyi Yang, Chenheng Xu, Weijun Hong, Ce Mo

Moral reasoning is fundamental to safe Artificial Intelligence (AI), yet ensuring its consistency across modalities becomes critical as AI systems evolve from text-based assistants to embodied agents. Current safety techniques demonstrate success in textual contexts, but concerns remain about generalization to visual inputs. Existing moral evaluation benchma

Hunain Ahmed Jillani, Ahmed Tawfik Aboukhadra, Ahmed Elhayek, Jameel Malik

Fast and accurate 3D hand reconstruction is essential for real-time applications in VR/AR, human-computer interaction, robotics, and healthcare. Most state-of-the-art methods rely on heavy models, limiting their use on resource-constrained devices like headsets, smartphones, and embedded systems. In this paper, we investigate how the use of lightweight neura

Daniil I. Sarypov, Dmitriy A. Pokhabov, Arthur G. Pogosov, Evgeny Yu. Zhdanov

In solids, frequent electron-electron collisions can induce collective, fluid-like electron transport. While this regime offers a powerful framework for exploring many-body phenomena, there is still a lack in functional electronic device actively exploiting hydrodynamic behaviour of electrons. Here, we introduce a solid-state analogue of a Tesla valve $\unic

Jingying Bao, J. Andrew Zhang, Kai Wu, Christos Masouros

In this correspondence, we investigate networked sensing in perceptive mobile networks under a bistatic multi-transmitter single-receiver uplink topology, where multiple user equipments (UEs) transmit signals over orthogonal frequency-division multiple access (OFDMA) resources and a single base station performs joint sensing. Uplink clock asynchronism introd

Leonid Erlygin, Alexey Zaytsev

Accurate uncertainty estimation is essential for building robust and trustworthy recognition systems. In this paper, we consider the open-set text classification (OSTC) task - and uncertainty estimation for it. For OSTC a text sample should be classified as one of the existing classes or rejected as unknown. To account for the different uncertainty types enc

Mykael Cardoso, Gleison do N. Santos, Roger P. de Moura

In this paper we establish exponential decay results for solutions of the damped $n$-dimensional Zakharov--Kuznetsov equation for $2 \le n \le 3$. More precisely, we prove the exponential decay of the $L^2(\mathbb{R}^n)$ norm when the damping is localized. In addition, when the dissipative mechanism acts on the whole space $\mathbb{R}^n$, we prove the expone

Rishaank Gupta

Large language model compression has made substantial progress through pruning, quantization, and low-rank decomposition, yet a fundamental limitation persists across all existing methods: compression budgets are allocated without any representation of what individual model components functionally encode. We term this the capability-blind compression problem

Junseok Lee, Sungho Shin, Seongju Lee, Kyoobin Lee

Single-domain generalization is essential for object detection, particularly when training models on a single source domain and evaluating them on unseen target domains. Domain shifts, such as changes in weather, lighting, or scene conditions, pose significant challenges to the generalization ability of existing models. To address this, we propose Cross-Doma

Weiping Shang, Shanshan Zhang

Let $G$ be a graph with vertex set $V=V(G)$. A double Roman dominating function on a graph $G$ is a function $f : V \to \{0,1,2,3\}$ satisfying the conditions that if $f(v) = 0$, then vertex $v$ must have at least two neighbors in $V_2$ or one neighbor in $V_3$, if $f(v) = 1$, then vertex $v$ must have at least one neighbor in $V_2 \cup V_3$. The weight of a

Manjari Dutta

The central theme of my thesis is to explore various simple prototype models that are exactly solvable in the framework of time dependent noncommutative spaces. By adopting the methodology provided by the Lewis Riesenfeld theory, we developed a procedure for obtaining a class of exact solutions for such model systems. We analyzed these solutions by deriving

Yixuan Wang, Qingyu Shi, Jiayu Zhou, Dianbo Liu

The growing context length of Large Language Models (LLMs) enlarges the Key-Value (KV) cache, limiting deployment in resource-limited environments. Prior training-free approaches for KV cache compression typically rely on low-rank approximation or scalar quantization, which fail to simultaneously achieve high compression ratios and high reconstruction fideli

Haochen Luo, Zhengzhao Lai, Junjie Xu, Yifan Li

Large Language Models (LLMs) excel at general code generation, yet translating natural-language trading intents into correct option strategies remains challenging. Real-world option design requires reasoning over massive, multi-dimensional option chain data with strict constraints, which often overwhelms direct generation methods. We introduce the Option Que

Yaroslav V. Kartashov, Vladimir V. Konotop, Dmitry A. Zezyulin

We demonstrate that linear and nonlinear Thouless pumping can be realized in two-component elongated Bose-Einstein condensates using helicoidal spin-orbit coupling that slides with respect to a static optical lattice, identical for both spinor components. Stable quantized transport is found for solitons in semi-infinite and finite gaps, within certain interv

G. Ciarfaglia, A. Rosanova, S. Cipolla, J. Bartoli

EngGPT2-16B-A3B is the latest iteration of Engineering Group's Italian LLM and it's built to be a Sovereign, Efficient and Open model. EngGPT2 is trained on 2.5 trillion tokens - less than Qwen3's 36T or Llama3's 15T - and delivers performance on key benchmarks, including MMLU-Pro, GSM8K, IFEval and HumanEval, comparable to dense models in the 8B-16B range,

Yicheng Rui, Xiao-Wei Duan, Licai Deng, Fan Yang

Ground-based time-domain observatories require minute-by-minute, site-scale awareness of cloud cover, yet existing all-sky datasets are short, daylight-biased, or lack astrometric calibration. We present LenghuSky-8, an eight-year (2018-2025) all-sky imaging dataset from a premier astronomical site, comprising 429,620 $512 \times 512$ frames with 81.2% night

Joona Kareinen, Veikka Immonen, Tuomas Eerola, Lumi Haraguchi

This paper considers self-supervised cross-modal coordination as a strategy enabling utilization of multiple modalities and large volumes of unlabeled plankton data to build models for plankton recognition. Automated imaging instruments facilitate the continuous collection of plankton image data on a large scale. Current methods for automatic plankton image

Muhammad Ahmad

Hyperspectral image classification (HSIC) has been significantly advanced by deep learning methods that exploit rich spatial-spectral correlations. However, existing approaches still face fundamental limitations: transformer-based models suffer from poor scalability due to the quadratic complexity of self-attention, while recent Fourier transform-based metho

O. R. Jadhav, L. K. Dewangan, I. I. Zinchenko, Thushara G. S. Pillai

We present the SOFIA/HAWC+ 214 $\mu$m polarimetric observations toward the infrared dark cloud G351.77-0.53 (hereafter G351), complemented by existing multi-wavelength data sets. Infrared excess from the embedded sources indicate ongoing star formation activity in the cloud. The G351 cloud hosts two prominent star-forming clumps, i.e., c1 and c2. The plane-o

Alessandro Gianola, Zeeshan Hameed, Marco Montali, Anjo Seidel

Object-centric process mining examines how processes interact with multiple co-evolving objects, and has gained great interest in recent years. However, object-centric event logs (OCELs) leave object relationships underspecified in several respects, especially if relationships are dynamic, i.e., they change over time. In this paper, we identify and formally

Masakazu Yoshimura, Teruaki Hayashi, Yuki Hoshino, Wei-Yao Wang

The realm of Mamba for vision has been advanced in recent years to strike for the alternatives of Vision Transformers (ViTs) that suffer from the quadratic complexity. While the recurrent scanning mechanism of Mamba offers computational efficiency, it inherently limits non-causal interactions between image patches. Prior works have attempted to address this

Xiaolin Wang, Jiayao Liu, Zimeng Zeng, Hongyu Yuan

Reconfigurable and sustainable microcavity lasers are highly desirable for next-generation integrated photonics. Here, we report a recyclable, low-threshold quasi-bound state in the continuum (q-BIC) laser fabricated via low-cost, high-throughput interference lithography. By introducing a polyvinyl alcohol (PVA) coating on a dye-doped photonic crystal, we su

Jing Dai, Chen Wu, Ming Wu, Qibin Zhang

Recent advances in multimodal learning have significantly improved cancer survival risk prediction. However, the joint prognostic potential of protein markers and histopathology images remains underexplored, largely due to the high cost and limited availability of protein expression profiling. To address this challenge, we propose HGP-Mamba, a Mamba-based mu

Zhengdao Li, Penggao Yan, Baoshan Song, Li-Ta Hsu

A Gaussian error assumption is commonly adopted in the pseudorange measurement model for global navigation satellite system (GNSS) positioning, which leads to the conventional least squares (LS) estimator. In urban environments, however, multipath and non-line-of-sight (NLOS) receptions produce heavy-tailed pseudorange errors that are not well represented by

Juan Carlos Verano-Espitia, Tero Mäkinen, Mikko J. Alava, Jérôme Weiss

Creep under a sustained load can persist for long times yet culminate in abrupt yielding or rupture, implying a finite lifetime even when the material appears solid. Here, we formulate lifetime prediction as Bayesian inference over an evolving activation-energy landscape. A time-dependent distribution of activation barriers controls deformation: stress lower

Saber Zerhoudi, Kanishka Ghosh Dastidar, Felix Klement, Artur Romazanov

Moltbook is a social network where every participant is an AI agent. We analyze 1,312,238 posts, 6.7~million comments, and over 120,000 agent profiles across 5,400 communities, collected over 40 days (January 27 to March 9, 2026). We evaluate the platform through three layers. At the interaction layer, 91.4% of post authors never return to their own threads,

Quentin Muller, Tommaso Tufarelli, Madalin Guta, Katherine Inzani

Surface roughness is an important quantity to many engineering and precision manufacturing disciplines. In this paper we investigate the problem of estimating the root-mean-square roughness of a sample by passive linear optics. By adopting quantum parameter estimation methods, we determine the ultimate precision limits for estimating spatial moments of a gen

Quan Cheng

Recent empirical results have demonstrated that training large language models (LLMs) with negative-only feedback can match or exceed standard reinforcement learning from human feedback (RLHF). Negative Sample Reinforcement achieves parity with PPO on mathematical reasoning; Distributional Dispreference Optimization trains effectively using only dispreferred

Jakub Leśkiewicz, Bartosz Furmanek, Michał Lipiński, Dmitriy Morozov

Topological simplification is the process of reducing complexity of a function while maintaining its essential features. Its goal is to find a new filter function, which reorders cells of the input complex in a way which eliminates some persistent homological features, without affecting the rest. We present a new approach to simplification based on the conce

Zhenghua Bao, Yi Shi

Multi-hop question answering (QA) requires reasoning across multiple documents, yet existing retrieval-augmented generation (RAG) approaches address this either through graph-based methods requiring additional online processing or iterative multi-step reasoning. We present IndexRAG, a novel approach that shifts cross-document reasoning from online inference