December 2025 arXiv papers — page 46

Shuzheng Si, Qingyi Wang, Haozhe Zhao, Yuzhuo Bai

Recognizing whether outputs from large language models (LLMs) contain faithfulness hallucination is crucial for real-world applications, e.g., retrieval-augmented generation and summarization. In this paper, we introduce FaithLens, a cost-efficient and effective faithfulness hallucination detection model that can jointly provide binary predictions and corres

Sadia Nasrin Tisha, Md Nazmus Sakib, Sanorita Dey

Hackathons have emerged as dynamic platforms for fostering innovation, collaboration, and skill development in the technology sector. Structural differences across hackathon formats raise important questions about how event design can shape student learning experiences and engagement. This study examines two distinct hackathon formats: a gender-specific hack

Zeev Nutov, Anael Vaknin

A set-family ${\cal F}$ is disjointness-compliable if $A' \subseteq A \in {\cal F}$ implies $A' \in {\cal F}$ or $A \setminus A' \in {\cal F}$; if ${\cal F}$ is also symmetric then ${\cal F}$ is proper. A classic result of Goemans and Williamson [SODA 92:307-316] states that the problem of covering a proper set-family by a min-cost edge set admits approximat

Dan Chen, Heye Huang, Tiantian Chen, Zheng Li

Current LLM-based driving agents that rely on unstructured plain-text memory suffer from low-precision scene retrieval and inefficient reflection. To address this limitation, we present RESPOND, a structured decision-making framework for LLM-driven agents grounded in explicit risk patterns. RESPOND represents each ego-centric scene using a unified 5 by 3 mat

Yunan Lin, Sebastian Bathiany, Maha Badri, Maximilian Gelbrecht

Global gridded crop models (GGCMs) are crucial to project the impacts of climate change on agricultural productivity and assess associated risks for food security. Despite decades of development, state-of-the-art GGCMs retain substantial uncertainties stemming from process representations. Recently, machine learning approaches trained on observational data p

Aaron Chan, Alex Ding, Frank Chen, Alan Wu

The rapid integration of Large Language Models (LLMs) into decentralized physical infrastructure networks (DePIN) is currently bottlenecked by the Verifiability Trilemma, which posits that a decentralized inference system cannot simultaneously achieve high computational integrity, low latency, and low cost. Existing cryptographic solutions, such as Zero-Know

Izumi Hachisu, Maiko Kato

V1674 Her is the fastest ($t_2\sim 1$ day) classical nova in our Galaxy and its absolute $V$ peak of $M_{V,\rm max}\sim -10.2$ is one magnitude brighter than typical very fast novae. Such a nova is sometimes called a superbright nova. Using our fully self-consistent nova outburst model combined with the optically thick winds on a $1.35 ~M_\odot$ white dwarf

Hao Guo, Xugong Qin, Jun Jie Ou Yang, Peng Zhang

Document image retrieval (DIR) aims to retrieve document images from a gallery according to a given query. Existing DIR methods are primarily based on image queries that retrieve documents within the same coarse semantic category, e.g., newspapers or receipts. However, these methods struggle to effectively retrieve document images in real-world scenarios whe

Ze Gong, Pradeep Varakantham, Akshat Kumar

Offline Preference-based Reinforcement Learning (PbRL) learns rewards and policies aligned with human preferences without the need for extensive reward engineering and direct interaction with human annotators. However, ensuring safety remains a critical challenge across many domains and tasks. Previous works on safe RL from human feedback (RLHF) first learn

Yan Zhang, Li Deng, Lixin Duan, Ivor W. Tsang

The fast online recommendation is critical for applications with large-scale databases; meanwhile, it is challenging to provide accurate recommendations in sparse scenarios. Hash technique has shown its superiority for speeding up the online recommendation by bit operations on Hamming distance computations. However, existing hashing-based recommendations suf

Toshiki Takadera, Shin'ichi Hirano, Tsutomu Kobayashi

Cosmic voids in the large-scale structure are among the useful probes for testing gravity on cosmological scales. In this paper, we investigate the evolution of voids in the Horndeski theory using the effective field theory (EFT) of dark energy. Modeling the void formation with the dynamics of spherical mass shells, we study how modifications of gravity enco

Vili Heinonen, Jani Lukkarinen

We study the long-time behavior of a non-interacting two-dimensional quantum gas in a weak random potential with long-range correlations. Any peaked initial momentum distribution will eventually become isotropic and broaden due to scattering events with the random potential. We derive an expression for the long-time average of the momentum distribution and t

Z. -P. Xie, C. -P. Hao, D. Sheng

Atomic cells made by anodically bonding silicon and borosilicate glasses are widely used in atomic devices. One inherent problem in these cells is that the silicon material blocks beams with wavelengths shorter than 1000 nm, which limits available optical accesses when alkali metal atoms are involved. In this work, we investigate the possibility of the silic

Junghyun Lee, Branislav Kveton, Anup Rao, Subhojyoti Mukherjee

Large language models (LLMs) solve reasoning problems by first generating a rationale and then answering. We formalize reasoning as a latent variable model and derive a reward-based filtered expectation-maximization (FEM) objective for learning to reason. This view connects EM and modern reward-based optimization, and shows that the main challenge lies in de

Songze Li, Jiameng Cheng, Yiming Li, Xiaojun Jia

By integrating language understanding with perceptual modalities such as images, multimodal large language models (MLLMs) constitute a critical substrate for modern AI systems, particularly intelligent agents operating in open and interactive environments. However, their increasing accessibility also raises heightened risks of misuse, such as generating harm

Florian De Leger

From a coloured operad $\mathcal{P}$ and a $\mathcal{P}$-algebra $A$, we construct a new operad $\mathrm{SC}(\mathcal{P})$ and a Hochschild object $\mathrm{Hoch}(A)$ together with an $\mathrm{SC}(\mathcal{P})$-action on the pair $(\mathrm{Hoch}(A),A)$. We prove that if $\mathcal{P}$ is the little $n$-disks operad, then $\mathrm{SC}(\mathcal{P})$ is equivalen

Shalender Singh, Santosh Kumar

Wavefunction collapse is commonly associated with unavoidable physical disturbance of the measured system. Here we show that in driven-dissipative quantum systems, continuous measurement can induce strong trajectory-level collapse while leaving the ensemble-averaged steady state strictly invariant. We identify measurement-invariant steady states whose uncond

Xiaofan Wang, Xingyu Gao, Jianlong Fu, Zuolei Li

The capability of performing long-horizon, language-guided robotic manipulation tasks critically relies on leveraging historical information and generating coherent action sequences. However, such capabilities are often overlooked by existing Vision-Language-Action (VLA) models. To solve this challenge, we propose LoLA (Long Horizon Latent Action Learning),

Aviad Eisenberg, Sharon Gannot, Shlomo E. Chazan

This paper presents a robust multi-channel speaker extraction algorithm designed to handle inaccuracies in reference information. While existing approaches often rely solely on either spatial or spectral cues to identify the target speaker, our method integrates both sources of information to enhance robustness. A key aspect of our approach is its emphasis o

Honglin Mu, Jinghao Liu, Kaiyang Wan, Rui Xing

Large Language Models (LLMs) excel at text comprehension and generation, making them ideal for automated tasks like code review and content moderation. However, our research identifies a vulnerability: LLMs can be manipulated by "adversarial instructions" hidden in input data, such as resumes or code, causing them to deviate from their intended task. Notably

Arghavan Bazigaran, Hansem Sohn

We compare human and large language model (LLM) generalization in the number game, a concept inference task. Using a Bayesian model as an analytical framework, we examined the inductive biases and inference strategies of humans and LLMs. The Bayesian model captured human behavior better than LLMs in that humans flexibly infer rule-based and similarity-based

Dhivya Dharshini Kannan, Anupam Trivedi, Dipti Srinivasan

Data centers account for significant global energy consumption and a carbon footprint. The recent increasing demand for edge computing and AI advancements drives the growth of data center storage capacity. Energy efficiency is a cost-effective way to combat climate change, cut energy costs, improve business competitiveness, and promote IT and environmental s

Dhruv Ringe

In this Ph.D. thesis, we study the methods to constrain particle physics models using the stochastic GW imprints from cosmological phase transitions (PTs). Beginning with the theory and background, we describe how the GW background from first-order PTs (FOPTs) and topological defects such as domain walls (DWs) can constrain the model parameter space at upcom

Ruiqi Wang, Xinchen Wang, Cuiyun Gao, Chun Yong Chong

Large language models (LLMs) have been increasingly deployed in real-world software engineering, fostering the development of code evaluation metrics to study the quality of LLM-generated code. Conventional rule-based metrics merely score programs based on their surface-level similarities with reference programs instead of analyzing functionality and code qu

Arne Hendrickx

We extend the classical Titchmarsh theorems to the Fourier transform of two types of H\"older-Lipschitz functions - additive and multiplicative - defined on fundamental domains of lattices in $\mathbb{R}^d$. Our approach is based on generalizations of Duren's lemma, which we first illustrate in the classical Euclidean setting. As an application of the second

Sofian Chaybouti, Sanath Narayan, Yasser Dahou, Phúc H. Lê Khac

Vision foundation models trained via multi-teacher distillation offer a promising path toward unified visual representations, yet the learning dynamics and data efficiency of such approaches remain underexplored. In this paper, we systematically study multi-teacher distillation for vision foundation models and identify key factors that enable training at low

Tongyi Fun Team, Qian Chen, Luyao Cheng, Chong Deng

Recent advancements in joint speech-text models show great potential for seamless voice interactions. However, existing models face critical challenges: temporal resolution mismatch between speech tokens (25Hz) and text tokens (~3Hz) dilutes semantic information, incurs high computational costs, and causes catastrophic forgetting of text LLM knowledge. We in

Ying Zhu, Jiaxin Wan, Xiaoran Liu, Siyang He

Diffusion Language Models (dLLMs) have emerged as promising alternatives to Auto-Regressive (AR) models. While recent efforts have validated their pre-training potential and accelerated inference speeds, the post-training landscape for dLLMs remains underdeveloped. Existing methods suffer from computational inefficiency and objective mismatches between train

Tomona Kinugawa, Tetsuo Hyodo

Understanding the internal structure of near-threshold states is essential for revealing the nature of exotic hadrons. Motivated by this challenge, we discuss the clustering structures of near-threshold $s$-wave eigenstates using the compositeness, which characterizes the clustering nature of the states. We show that shallow bound states usually possess clus

Luca Barbieri, Marcus Henninger, Paolo Tosi, Artjom Grudnitsky

Integrated Sensing and Communication (ISAC) systems enable cellular networks to jointly operate as communication technology and sense the environment. While opportunities and potential performance have been largely investigated in simulations, few experimental works have showcased Automatic Target Recognition (ATR) effectiveness in a real-world deployment ba

M. Dotti, F. Mannucci, R. Buscicchio, M. Colpi

The Laser Interferometer Space Antenna (LISA), adopted by ESA and scheduled for the second half of the next decade, will drive a new revolution in the rapidly growing field of gravitational-wave astronomy, by extending GW observations into the hiterto unexplored millihertz regime. One of the key source classes of LISA is merging massive black hole binaries i

Pooyan Amir-Ahmadi, Marko Mlikota, Dalibor Stevanović

For a general class of dynamic and stochastic structural models, we show that (i) non-linearity in economic dynamics is a necessary and sufficient condition for time-varying parameters (TVPs) in the reduced-form VARMA process followed by observables, and (ii) all parameters' time-variation is driven by the same, typically few sources of stochasticity: the st

Chengwei Liu, Haoyin Yan, Shaofei Xue, Xiaotao Liang

Many existing audio processing and generation models rely on task-specific architectures, resulting in fragmented development efforts and limited extensibility. It is therefore promising to design a unified framework capable of handling multiple tasks, while providing robust instruction and audio understanding and high-quality audio generation. This requires

Daniele Simeoni, Andrea Renato Rossi, Gianmarco Parise, Fabio Guglietta

We focus on the process of plasma acceleration in the presence of non-negligible thermal effects, wherein a driver of relativistic electrons perturbs a warm neutral plasma and generates a wakefield structure. We study the acceleration process via numerical simulations based on fluid models with different thermal closure assumptions, and also provide systemat

Jakob Hedicke

Often it is possible to equip the space of all cone geodesics of a strongly convex cone structure with the structure of a smooth contact manifold. This generalizes the analogous notions for the space of light rays of a Lorentzian spacetime. After reviewing these constructions on the space of cone geodesics, with a focus on the natural contact structure, we e

Natasha Fernandes, Annabelle McIver, Parastoo Sadeghi

"f differential privacy" (fDP) is a recent definition for privacy privacy which can offer improved predictions of "privacy loss". It has been used to analyse specific privacy mechanisms, such as the popular Gaussian mechanism. In this paper we show how fDP's foundation in statistical hypothesis testing implies equivalence to the channel model of Quantitative

Robert van de Ven, Trim Bresilla, Bram Nelissen, Ard Nieuwenhuizen

Automating tasks in orchards is challenging because of the large amount of variation in the environment and occlusions. One of the challenges is apple pose estimation, where key points, such as the calyx, are often occluded. Recently developed pose estimation methods no longer rely on these key points, but still require them for annotations, making annotatin

Sourav Bhattacharya, Ashish Yadav

We study the problem of relating cycles on a \emph{triod} $Y$ to \emph{circle rotations}. We prove that the simplest cycles on a \emph{triod}~$Y$ with a given \emph{rotation number}~$\rho$, called \emph{triod--twist cycles} are conjugate, via a piece-wise monotone map of \emph{modality} at most~$m + 3$, where~$m$ is the \emph{modality} of~$P$ to the rotation

Yiming Chen, Zijun Chen, Yizhe Zhu

We study the limiting spectral distribution of the normalized Laplacian $\mathcal L$ of an Erdős-Rényi graph $G(n,p)$. To account for the presence of isolated vertices in the sparse regime, we define $\mathcal L$ using the Moore-Penrose pseudoinverse of the degree matrix. Under this convention, we show that the empirical spectral distribution of a suitably n

Xiang Chen, Yixin Ou, Quan Feng, Lei Li

The pre-trained foundation models (PFMs) have become essential for facilitating large-scale multimodal learning. Researchers have effectively employed the ``pre-train, prompt, and predict'' paradigm through prompt learning to induce improved few-shot performance. However, prompt learning approaches for PFMs still follow a parametric learning paradigm. As suc

Yuxin Wang, Shicheng Fang, Bo Wang, Qi Luo

Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for Large Language Models (LLMs) to address knowledge-intensive queries requiring domain-specific or up-to-date information. To handle complex multi-hop questions that are challenging for single-step retrieval, iterative RAG approaches incorporating reinforcement learning have been propo

Marco D'Addezio

In the proof of Crew's parabolicity conjecture, we established a key property concerning the slopes of $\dagger$-hulls of $F$-isocrystals, extending a result of Tsuzuki. This article presents an alternative proof of this theorem for a specific class of $F$-isocrystals. The central ingredient is a local extension property for \'etale $p$-divisible subgroups.

Zhihan Cao, Xiao Yang, Gaolei Li, Jun Wu

Video semantic communication, praised for its transmission efficiency, still faces critical challenges related to privacy leakage. Traditional security techniques like steganography and encryption are challenging to apply since they are not inherently robust against semantic-level transformations and abstractions. Moreover, the temporal continuity of video e

Deriyan Senjaya

This study explores the influence of a Stark-like perturbative potential on a quantum particle confined to a cylindrical surface (QPCS) and its implications for extra-dimensional theories. The QPCS framework is particularly relevant to Kaluza-Klein (KK) theory, which postulates extra spatial dimensions to unify electromagnetism and gravity. In KK theory, the

Xingyou Yin, Ceyao Zhang, Min Hu, Kai Chen

Large Language Models (LLMs) have demonstrated effectiveness as zero-shot time series (TS) forecasters. The key challenge lies in tokenizing TS data into textual representations that align with LLMs' pre-trained knowledge. While existing work often relies on fine-tuning specialized modules to bridge this gap, a distinct, yet challenging, paradigm aims to lev

I. A. Pshenichnov, S. D. Savenkov, A. O. Svetlichnyi

We investigate the production of secondary nuclei in the hadronic fragmentation and electromagnetic dissociation (EMD) of $^{20}$Ne beams at the LHC and $^{124}$Xe beams at NICA. For light nuclei at LHC energies, our calculations show that hadronic interactions are the dominant channel for nuclear transmutation. This contrasts with the previously established

Masanori Nakamura, Teruo Jyo, Munehiko Nagatani, Hitoshi Wakita

We successfully transmitted a net 582-Gb/s probabilistically shaped PAM12 C-band signal over 11-km dispersion-shifted fibre and net 4$\times$526-Gb/s uniform PAM8 O-band signals over 2-km four-core fibre using a single-carrier 216-GBd IMDD system based on a 150-GHz bandwidth InP-DHBT electrical mixer and a thin-film lithium-niobate modulator.

V. N. Zverev, N. A. Abdullayev, Z. S. Aliev, I. R. Amiraslanov

The results of a comprehensive study of MnSb$_2$Te$_4$ single crystals are presented. The structure, Raman spectra, low-temperature transport, Hall effect, magnetization, and magnetic susceptibility are studied. It was established that the crystals are ferromagnetic, with a Curie temperature ranging from 22 to 45\,K for different samples. Hall and magnetizat

Hyeongcheol Park, Jiyoung Seo, Jaewon Mun, Hogun Park

Retrieval-Augmented Generation (RAG) has recently been extended to multimodal settings, connecting multimodal large language models (MLLMs) with vast corpora of external knowledge such as multimodal knowledge graphs (MMKGs). Despite their recent success, multimodal RAG in the audio-visual domain remains challenging due to 1) limited modality coverage and mul

Zhuo Yang, Yeyun Chen, Jiaqing Xie, Ben Gao

Molecular editing and optimization are multi-step problems that require iteratively improving properties while keeping molecules chemically valid and structurally similar. We frame both tasks as sequential, tool-guided decisions and introduce MolAct, an agentic reinforcement learning framework that employs a two-stage training paradigm: first building editin

Guan-Qiang Li, Yu-Qi Zhang, Hao Guo, You-Jiao Dong

The optimal control of population transfer for multi-level systems is investigated from the perspective of quantum geometry. Firstly, the general theoretical framework of optimizing the stimulated Raman adiabatic passage (STIRAP) scheme based on the dynamical quantum geometric tensor is given, and then the dynamical quantum geometric tensor and the nonadiaba

Yuval Sallem, Nahala Yadid, Xi Wang, Irina volotsenko

Recent advances have shown that introducing dependency interactions between two superconducting networks can trigger abrupt, hysteretic normal-superconductor phase transitions. In this study, we demonstrate that such behavior can also arise in a single-network superconducting system that features two distinct types of interactions: short-range electrical con

Cyrus Vachha, Yixiao Kang, Zach Dive, Ashwat Chidambaram

Authoring 3D scenes is a central task for spatial computing applications. Competing visions for lowering existing barriers are (1) focus on immersive, direct manipulation of 3D content or (2) leverage AI techniques that capture real scenes (3D Radiance Fields such as, NeRFs, 3D Gaussian Splatting) and modify them at a higher level of abstraction, at the cost

Niraj Prakash Kini, Shiau-Rung Tsai, Guan-Hsun Lin, Wen-Hsiao Peng

Millimeter-wave radar offers a privacy-preserving and lighting-invariant alternative to RGB sensors for Human Pose Estimation (HPE) task. However, the radar signals are often sparse due to specular reflection, making the extraction of robust features from radar signals highly challenging. To address this, we present milliMamba, a radar-based 2D human pose es

Ruobing Chen, Sirui Yu

Let $\mathbb R^{m|n}$ be the usual superspace. The algebra of functions on it is Koszul, but its Koszul dual is not graded commutative, and in particular not the algebra of functions on $\mathbb R^{n|m}$. This contrasts with the two extreme cases, in which the polynomial and exterior algebras are Koszul dual. We remedy this discrepancy by realizing the algeb

Eric E. Mamajek, Jason T. Wright, Noah W. Tuchow, Patrick A. Young

Measurements of physical parameters for stars and (exo)planets are often quoted in units normalized to the Sun and/or Earth. The nominal total solar irradiance, ${S}^{\rm N}_{\odot}$, while based on a current best estimate with uncertainties, was adopted to be an exact reference value of 1361 W m$^{-2}$ by IAU 2015 Resolution B3, corresponding to ``the mean

Habib Alizadeh, Marcelo S. Atallah, Dylan Cant, Jianqiao Shang

The diameter of the spectral pseudometric on the universal cover of the Hamiltonian diffeomorphism group of $\mathrm{Gr}(2,p)$ is shown to be finite whenever $p$ is a prime number. On the other hand, it is shown that the diameter is infinite in the case of $\mathrm{Gr}(2k,2n)$ for all natural numbers $k<n$.

Philipp Weder, Annalisa Buffa

Defeaturing, the process of simplifying computational geometries, is a critical step in industrial simulation pipelines for reducing computational cost. Rigorous a posteriori estimators exist for the global energy-norm error introduced by geometry simplifications. However, practitioners are usually more concerned with the accuracy of specific quantities of i

Mustafa Bakr, Smain Amari

The transverse magnetic (TM) modes of a spherical cavity satisfy a dispersion relation connecting the angular eigenvalue $\nu$ to the resonant frequency through zeros of the spherical Bessel function derivative. Analytic continuation of this dispersion relation to $\nu = -1$ yields a formal zero-frequency endpoint where $j_{-1}(x) = \cos x / x$ admits the ro

Dor Shamay, Boaz Rafaely

Binaural reproduction is gaining increasing attention with the rise of devices such as virtual reality headsets, smart glasses, and head-tracked headphones. Achieving accurate binaural signals with these systems is challenging, as they often employ arbitrary microphone arrays with limited spatial resolution. The Binaural Signals Matching with Magnitude Least

Lechuan Sun, Shan Zhang, Bin Xu, Rui Su

Shear band propagation and interaction are critical to the mechanical performance of metallic glasses and are strongly governed by thermal history, yet their microscopic mechanisms remain unclear. Here, using molecular dynamics simulations combined with a state-of-the-art annealing protocol, we systematically investigate these behaviors in a model metallic g

Mohammad Helal Uddin, Liam Seymour, Sabur Baidya

Vision Transformers (ViTs) deliver state-of-the-art accuracy but their quadratic attention cost and redundant computations severely hinder deployment on latency and resource-constrained platforms. Existing pruning approaches treat either tokens or heads in isolation, relying on heuristics or first-order signals, which often sacrifice accuracy or fail to gene

Mustafa Bakr, Smain Amari

The spherical harmonics $Y_\ell^m$ fall into three families -- sectoral ($\ell = |m|$), tesseral ($\ell > |m| > 0$), and zonal ($m = 0$) -- which exhibit fundamentally different behaviour under analytic continuation to non-integer parameters. We demonstrate that this trichotomy has a natural explanation in the representation theory of SO(3). Sectoral harmoni

R. M. I. D Gamage, F. Ambrosino, L. Cimmino, G. Nyitrai

This work presents a comprehensive characterization of the FERS 5202 front-end readout unit when processing signals from Silicon Photo-multipliers (SiPMs). The readout system's performance is characterized in terms of its charge resolution, dynamic range, and noise performance at the single photoelectron level, which is critical for applications requiring de

Yongchan Son, Jahun Jang, Been An, Jimoon Kang

Team communication plays a vital role in supporting collaboration in multiplayer online games. Therefore, numerous studies were conducted to examine communication patterns in esports teams. While non-verbal communication has been extensively investigated, research on assessing voice-based verbal communication patterns remains relatively understudied. In this

Yuanhao Chen, Qi Liu, Pengbin Chen, Zhongjian Qiao

Offline reinforcement learning (RL) aims to learn a policy that maximizes the expected return using a given static dataset of transitions. However, offline RL faces the distribution shift problem. The policy constraint offline RL method is proposed to solve the distribution shift problem. During the policy constraint offline RL training, it is important to e

Kashinath T. Chavan, Ihsan Boustani, Alok Shukla

A systematic study of the structural, electronic, and optical properties of cage-like boron clusters, with the number of constituent atoms ranging from 20 to 122, has been carried out within the framework of density-functional theory (DFT), employing 6-31G(d, p) extended basis set. The dynamic stability of the clusters is analyzed through the vibrational fre

Xian-Rong Zhang, Yue-Jiao Gong, Wei-Neng Chen, Jun Zhang

Evolutionary Neural Architecture Search (ENAS) has gained attention for automatically designing neural network architectures. Recent studies use a neural predictor to guide the process, but the high computational costs of gathering training data -- since each label requires fully training an architecture -- make achieving a high-precision predictor with { li

Akilan Sankaran, Diego Israel Chavez

We investigate the ability of millimetric walking droplets to tunnel between spatially-structured cavities. By synthesizing experimental and theoretical analysis, we provide a comprehensive framework for droplet tunneling mechanics in three spatial dimensions. We define a generalized Dirichlet-to-Neumann operator that enables explicit characterization of dro

Siyuan Tong, Qilong Zhai, Qian Zhang, Ran Zhang

We develop an enriched Galerkin (EG) method for the incompressible Navier-Stokes equations that conserves both kinetic energy and helicity in the inviscid limit without introducing any additional projection variables. The method employs an EG velocity space, which is the first-order continuous Galerkin space enriched with piecewise constants defined on mesh

Thanh-Tung Le, Tuan Pham, Tung Nguyen, Deying Kong

Novel view synthesis (NVS) seeks to render photorealistic, 3D-consistent images of a scene from unseen camera poses given only a sparse set of posed views. Existing deterministic networks render observed regions quickly but blur unobserved areas, whereas stochastic diffusion-based methods hallucinate plausible content yet incur heavy training- and inference-

Subrata Kumer Paula, Dewan Nafiul Islam Noora, Rakhi Rani Paula, Md. Ekramul Hamidb

Human Activity Recognition (HAR) plays a vital role in healthcare, surveillance, and innovative environments, where reliable action recognition supports timely decision-making and automation. Although deep learning-based HAR systems are widely adopted, the impact of Activation Functions (AFs) and Model Optimizers (MOs) on performance has not been sufficientl

Md Mahfuzur Rahman, Nishith Tripathi, Jeffrey H. Reed, Lingjia Liu

Non-Terrestrial Networks (NTN) are emerging as critical enablers of global connectivity, particularly in remote, unserved, underserved, or maritime regions lacking traditional infrastructure. While much of the existing work on NTN focuses on theoretical or simulated evaluations, practical implementations remain limited. In this paper, we present SpaceNET, a

Xuan Yang. Junhui Li, Bin Luo, Ziyang Chen, Hong guo

In precision frequency transfer systems, stringent requirements are imposed on the phase stability of transmitted signals. Throughout the transmission process, the inherent challenges of long-haul signal propagation inevitably introduce multiple noise components, including but not limited to thermal noise, phase fluctuations, and environmental interference.

Neha Hotwani, T. S. S. R. K. Rao

We provide a characterization of the $C^*$-extreme points of the closed unit ball of a von Neumann algebra and demonstrate that $C^*$-extremality is equivalent to both linear extremality and strong extremality. As an application, we characterize certain classes of von Neumann algebras in terms of their $C^*$-extreme points.

W. Horiuchi, Y. Suzuki, R. B. Wiringa

The application of Glauber theory has been playing an increasingly important role with the study of unstable or exotic nuclei. Its adaptation to medium and high-energy nucleus-nucleus collisions is severely limited because one has to evaluate the matrix elements of multiple-scattering operators. The extraction of physical observables has been done using 'app

R. D'Amicis, J. M. Raines, S. Benella, M. Velli

Fast and slow solar wind have distinct properties linked to their solar sources.Alfv\'enic slow wind complicates the usual speed-based classification, especially at intermediate speeds. Solar Orbiter's Solar Wind Analyzer (SWA) offers unique capabilities to investigate how Alfv\'enic slow wind differs from fast wind and relate these differences to their sola

Zuo Wang, Ye Yuan

Text classification plays an important role in various downstream text-related tasks, such as sentiment analysis, fake news detection, and public opinion analysis. Recently, text classification based on Graph Neural Networks (GNNs) has made significant progress due to their strong capabilities of structural relationship learning. However, these approaches st

Annika Hirling, Giorgio Nicoletti, Antonio Celani

The Multi-Armed Bandit problem provides a fundamental framework for analyzing the tension between exploration and exploitation in sequential learning. This paper explores Information Directed Sampling (IDS) policies, a class of heuristics that balance immediate regret against information gain. We focus on the tractable environment of two-state Bernoulli band

Nachiappan Chockalingam, Akshay Deshpande, Lokesh Butra, Ram Sekhar Bodala

Phasor Measurement Units (PMUs) generate high-frequency, time-synchronized data essential for real-time power grid monitoring, yet the growing scale of PMU deployments creates significant challenges in latency, scalability, and reliability. Conventional centralized processing architectures are increasingly unable to handle the volume and velocity of PMU data

W. Horiuchi, Y. Suzuki, R. B. Wiringa

Experiments using intermediate- to high-energy radioactive nuclear beams present numerous findings. Extracting important properties of physical observables relies on a firm theoretical analysis. Though Glauber theory is believed to work well, no convincing calculation has so far been done. We perform ab initio Glauber theory calculations of both elastic diff

Zuo Wang, Ye Yuan

In this paper, we investigate how the widely existing contextual and structural divergence may influence the representation learning in rich-text graphs. To this end, we propose Jensen-Shannon Divergence Message-Passing (JSDMP), a new learning paradigm for rich-text graph representation learning. Besides considering similarity regarding structure and text, J

Daichi Arai, Yuichi Kondo, Kyohei Unno, Yasuko Sugito

This study proposes a practical approach for compressing 360-degree equirectangular videos using pretrained neural video compression (NVC) models. Without requiring additional training or changes in the model architectures, the proposed method extends quantization parameter adaptation techniques from traditional video codecs to NVC, utilizing the spatially v

Yiming Du, Baojun Wang, Yifan Xiang, Zhaowei Wang

Temporal reasoning over long, multi-session dialogues is a critical capability for conversational agents. However, existing works and our pilot study have shown that as dialogue histories grow in length and accumulate noise, current long-context models struggle to accurately identify temporally pertinent information, significantly impairing reasoning perform

Aritra Das, Simon K. Yung, Lorcan O. Conlon, Ozlem Erkilic

Quantum measurements, alongside quantum states and processes, form a cornerstone of quantum information processing. However, unlike states and processes, their efficient characterisation remains relatively unexplored. We resolve this asymmetry by introducing a comprehensive framework for efficient detector estimation that reveals the fundamental limits to ex

Feiyu Wang, Liheng Wang, Mingrui Yuan, Zhen Han

High modulation efficiency and a compact footprint are critical for next-generation electro-optic (EO) modulators. We introduce a new class of non-resonant recirculating phase modulators (PMs) that boosts modulation efficiency by repeatedly modulating the optical field within a single, non-resonant waveguide, while fundamentally removing the loop-length matc

Zengzhao Xu, Ligong Wang, Weige Xi

A fractional $(a,b,m)$-covered graph is a generalization of the concept of a fractional $[a,b]$-covered graph. For any $H \subseteq G$ with edge set $|E(H)| = m$, if there exists a fractional $[a,b]$-factor (the corresponding fractional indicator function is $h$) such that $h(e) = 1$ for any $e \in H$, then the graph $G$ is called a fractional $(a,b,m)$-cove

Jinyoung Choi, Youngchae Kwon, Injung Kim

Fashion style classification is a challenging task because of the large visual variation within the same style and the existence of visually similar styles. Styles are expressed not only by the global appearance, but also by the attributes of individual items and their combinations. In this study, we propose an item region-based fashion style classification

Guang-Yao Xiao, Fabo Feng, Song Wang, Kai Li

The third body is expected to shape the formation and evolution of close binary systems. In this work, we develop a method to detect and characterize the tertiary companion around eclipsing binaries through the combined analysis of eclipse timing variation, Hipparcos and/or Gaia astrometry. This method allows us to determine both the true mass and the inclin

Jeehong Kim, Youngseok Hwang, Minchan Kim, Sungho Bae

Spatio-temporal graph neural networks (ST-GNNs) have achieved notable success in structured domains such as road traffic and public transportation, where spatial entities can be naturally represented as fixed nodes. In contrast, many real-world systems including maritime traffic lack such fixed anchors, making the construction of spatio-temporal graphs a fun

Tarak Chand, Saurabh Sharma, Koshvendra Singh, Joe P. Ninan

We present a detailed analysis of the long-term photometric and spectroscopic evolution of V1180 Cas over a decade, aiming to identify the dominant mechanisms behind its variability. We combine multi-band light curves from 1999 to 2025 with over 30 epochs of optical to near-infrared spectroscopy (0.5-2.5 $\mu$m), analyzing variability patterns, color behavio

Yanjie Li, Jian Xu, Xueqing Chen, Lina Yu

Adsorption energy is a key descriptor of catalytic reactivity. It is fundamentally defined as the difference between the relaxed total energy of the adsorbate-surface system and that of an appropriate reference state; therefore, the accuracy of relaxed-energy prediction directly determines the reliability of machine-learning-driven catalyst screening. E(3)-e

Wenzhao Wu, Yahui Tang, Mingfei Cheng, Wenbing Tang

As embodied agents advance toward real-world deployment, ensuring optimal decisions becomes critical for resource-constrained applications. Current evaluation methods focus primarily on functional correctness, overlooking the non-functional optimality of generated plans. This gap can lead to significant performance degradation and resource waste. We identify

Chaithra, Kamesh Kadimisetty, Biju R Mohan

Financial sentiment analysis plays a crucial role in informing investment decisions, assessing market risk, and predicting stock price trends. Existing works in financial sentiment analysis have not considered the impact of stock prices or market feedback on sentiment analysis. In this paper, we propose an adaptive framework that integrates large language mo

Alolika Roy, Amarendra K. Sarma

We theoretically investigate quantum measurement noise in a hybrid optomechanical system, focusing on radiation pressure back action and its impact on force sensing. The setup consists of an optomechanical cavity with a movable mirror, a fixed semi transparent mirror, an ensemble of quantum dots (QD) coupled to the cavity mode, and an intracavity optical par

Dianxuan Fu, Xiaomin Liu, Yihao Zhang, Shikui Shen

We propose a communication-bound-aware cross-domain resource assignment framework for pipeline-parallel distributed training over multi-datacenter optical networks, which lowers iteration time by 31.25% and reduces 13.20% blocking requests compared to baselines.

Tarakanta Nayak, Pooja Phogat

The Julia set of the Chebyshev's method applied to polynomials with exactly two distinct roots is shown to be connected, and its Fatou set is proved to be the union of attracting basins corresponding to the two roots. Further, if the two roots have the same multiplicity then the common boundary of the two immediate basins is proved to be a connected subset o

Taekyun Kim, Dae San Kim, Hyunseok Lee, Kyo-Shin Hwang

This paper introduces a degenerate version of the Euler-Seidel method by incorporating a parameter lambda into the classical recurrence relation. We define a degenerate Euler-Seidel matrix associated with an initial sequence and establish corresponding lambda-generalized binomial identities and generating function relations. By applying this method to the de

Abolfazl Lavaei, David Angeli

In this work, we propose a data-driven divide and conquer strategy for the stability analysis of interconnected homogeneous nonlinear networks of degree one with unknown models and a fully unknown topology. The proposed scheme leverages joint dissipativity-type properties of subsystems described by storage functions, while providing a stability certificate o

Christian Tantardini, Quentin Pitteloud, Boris Yakobson, Martin Andersson

Electromagnetic response is commonly computed in two languages: length-gauge molecular polarizabilities and velocity-gauge (Kubo) conductivities for periodic solids. We introduce a compact, gauge-invariant bridge that carries the same microscopic inputs-transition dipoles and interaction kernels-from molecules to crystals and heterogeneous media, with explic