October 2025 arXiv papers — page 99

Yongsheng Zhang

We discover some very general configuration results for constructing area-minimizing cones. In particular, given any closed minimal submanifold in some Euclidean sphere, every cone over the minimal product of sufficiently many copies of the submanifold turns out to be area-minimizing; meanwhile every cone over the minimal product of the submanifold and a rou

Fern Gossow, Andrew Huchala

The Borel-Weil-Bott theorem can be used to decompose the cohomology of twisted sheaves of holomorphic forms on the complex Grassmannian into irreducible representations of the general linear group. By analyzing this decomposition, we provide two effective formulae for computing the associated Hodge numbers, and give examples in special cases. One of these in

Junlong Tong, Yingqi Fan, Anhao Zhao, Yunpu Ma

Large language models (LLMs) have demonstrated remarkable capabilities in chain of thought (CoT) reasoning. However, the current LLM reasoning paradigm initiates thinking only after the entire input is available, which introduces unnecessary latency and weakens attention to earlier information in dynamic scenarios. Inspired by human cognition of thinking whi

Wuhao Xie, Kanji Tanaka

Long-term autonomy for mobile robots requires both robust self-localization and reliable map maintenance. Conventional landmark-based methods face a fundamental trade-off between landmarks with high detectability but low distinctiveness (e.g., poles) and those with high distinctiveness but difficult stable detection (e.g., local point cloud structures). This

Emeline Noël, Erwan Jézéquel, Pierre-Antoine Joulin

A purely analytical wake model for wind turbines is derived, anchored exclusively in physical interactions between atmospheric turbulence and turbine dynamics, and thus inherently accounting for atmospheric stratification. Unlike empirical models relying on assumed wake deficit shapes or tunable coefficients, this model predicts the wake deficit solely from

Chong Chen, Ze Liu, Lingfeng Bao, Yanlin Wang

The cryptocurrency market offers significant investment opportunities but faces challenges including high volatility and fragmented information. Data integration and analysis are essential for informed investment decisions. Currently, investors use three main approaches: (1) Manual analysis across various sources, which depends heavily on individual experien

Yuyang Hong, Qi Yang, Tao Zhang, Zili Wang

Recently, significant progress has been made in multi-modal continual learning, aiming to learn new tasks sequentially in multi-modal settings while preserving performance on previously learned ones. However, existing methods mainly focus on coarse-grained tasks, with limitations in addressing modality entanglement in fine-grained continual learning settings

Chunhao Cai, Cong Zhang

This paper deals with the Local Asymptotical normality for the joint drift parameter and Hurst parameter $H>3/4$ in the mixed fractional Ornstein-Uhlenbeck process. Different from the only estimation of the drift parameter when $H$ is known, we will use the fact that the mixed fractional Brownian motion is a semimartingale with its own filtering when $H>3/4$

Jinyu Hu, Chen Wu

In this paper, we carefully look at the $\alpha$ -decay half-lives of 196 even-even nuclei using a two-potential approach that is made better by taking into account an alpha particle's effective mass that changes with coordinates. The result shows that the accuracy of this model has been improved after considering effective mass for the alpha particle. Furth

Simeon Ball, Raven Zhang

A major difficulty in quantum computation is the ability to implement fault tolerant computations, protecting information against undesired interactions with the environment. Stabiliser codes were introduced as a means to protect information when storing or applying computations in Hilbert spaces where the local dimension is fixed, i.e. in Hilbert spaces of

Nicholas Lindsay

Let $X$ be a smooth complex prime Fano fourfold having a semi-free action of $\mathbb{C}^*$, then $X$ is contained in one of the families $\mathbb{P}^4,Q^4,W_5, X^{m}_{8}$. All of the families contain members that have a semi-free $\mathbb{C}^*$-action.

Hideyuki Miura, Jin Takahashi, Erbol Zhanpeisov

We consider the semilinear heat equation $u_t=\Delta u+|u|^{p-1} u$ in possibly non-convex and unbounded domains. Our main result shows the nonexistence of type II blow-up for possibly sign-changing solutions in the energy subcritical range $(n-2)p<n+2$. This resolves a long-standing open question dating back to the 1980s and also deduces the blow-up of the

Qing Shi, Jing He, Qiaosheng Chen, Gong Cheng

Dataset search is a well-established task in the Semantic Web and information retrieval research. Current approaches retrieve datasets either based on keyword queries or by identifying datasets similar to a given target dataset. These paradigms fail when the information need involves both keywords and target datasets. To address this gap, we investigate a ge

Igor V. Smetanin, Alexander V. Uskov

It has been demonstrated that a quantum well with optically excited, homogeneous oscillations of a two-dimensional (2D) electron plasma behaves as a photonic time crystal when it scatters an obliquely incident, weak probe electromagnetic wave. The hydrodynamic approximation is used to describe self-consistently the interaction between the probe wave and the

Gengyu Wang, Runze Zhang, Zhongzhi Zhang

Public opinion governance in social networks is critical for public health campaigns, political elections, and commercial marketing. In this paper, we addresse the problem of maximizing overall opinion in social networks by strategically modifying the internal opinions of key nodes. Traditional matrix inversion methods suffer from prohibitively high computat

Darashan Saikia, Liza Devi, Biplob Sarkar, Asish Jyoti Boruah

In astrophysical systems like X-ray binaries (XRBs), active galactic nuclei (AGN), and young stellar objects (YSOs), we often observe a very fundamental structure called accretion discs(ADs). Conventional AD theory usually supposes that the gravitational field is controlled by a central compact object. This assumption breaks down when the mass of the disc be

Eduard Naichuk, Jeroen van den Brink, Flavio S. Nogueira

In quantum optics and condensed matter physics non-Hermitian phenomena are often studied under the assumption of an open physical system. However, there are examples of intrinsically non-Hermitian, though often $\mathcal{PT}$ (parity-time) symmetric, not necessarily open systems, in which case the concept of gain and loss relative to an underlying environmen

Botong Gai, Shuanhong Wang

Let $H$ be a Hopf algebra. In this paper, we study a class of $H$-operators on $H$-pseudoalgebras, which resemble the Rota-Baxter $H$-operator, and they are called Rota-Baxter type $H$-operators. We firstly present some basic properties and examples. Then by using Rota-Baxter type $H$-operators, we construct a number of associative (resp. Lie, NS-) $H$-pseud

Jacek Wszoła, Krzysztof Burnecki, Marek Teuerle, Martyna Zdeb

This paper introduces a novel multidimensional insurance-linked instrument: a contingent convertible bond (CoCoCat bond) whose conversion trigger is activated by predefined natural catastrophes across multiple geographical regions. We develop such a model explicitly accounting for the complex dependencies between regional catastrophe losses. Specifically, we

Giulia Giusti

The concept of linearity plays a central role in both mathematics and computer science, with distinct yet complementary meanings. In mathematics, linearity underpins functions and vector spaces, forming the foundation of linear algebra and functional analysis. In computer science, it relates to resource-sensitive computation. Linear Logic (LL), for instance,

M. Molero, S. Mangano, C. Delgado

Iron cosmic rays represent the most abundant heavy nuclei at energies above 1 TeV, with their production thought to be primarily originated by astrophysical sources. Therefore, measuring the iron spectrum provides crucial insights into the origin, acceleration, and propagation mechanisms of cosmic rays. While recent results from space-based detectors have re

Zhuo Cao, Heming Du, Bingqing Zhang, Xin Yu

Existing Moment retrieval (MR) methods focus on Single-Moment Retrieval (SMR). However, one query can correspond to multiple relevant moments in real-world applications. This makes the existing datasets and methods insufficient for video temporal grounding. By revisiting the gap between current MR tasks and real-world applications, we introduce a high-qualit

A. Chernyavskiy, I. S. Cojocaru, S. M. Drofa, P. G. Vilyuzhanina

Nitrogen-vacancy (NV) centers in diamond are widely used in the development of a number of sensors. The sensitivity of these devices is limited by both the number of centers used and their coherent properties. While the effects on the coherent properties of paramagnetic impurities such as carbon 13-isotopes and p1 centers are rather well understood, the mutu

Botong Gai, Shuanhong Wang

In this paper, we mainly provide a new approache to construct Hom-Hopf algebras. For this, we introduce and study the notion of a left $(m,k)$-Hom-crossed product structure as a generalization of $k$-Hom-smash product structure. Then one combines this $(m,k)$-Hom-crossed product structure and a left $m$-Hom-smash coproduct structure to build Radford $[(m,k),

Keuntae Kim, Eunhye Jeong, Sehyeon Lee, Seohee Yoon

Recent advances in enhancing the reasoning ability of large language models (LLMs) have been remarkably successful. LLMs trained with reinforcement learning (RL) for reasoning demonstrate strong performance in challenging tasks such as mathematics and coding, even with relatively small model sizes. However, despite these improvements in task accuracy, the as

Fumiya Iwashige, Tomoya Wakayama, Shonosuke Sugasawa, Shintaro Hashimoto

Nonparametric Bayesian approaches provide a flexible framework for clustering without pre-specifying the number of groups, yet they are well known to overestimate the number of clusters, especially for functional data. We show that a fundamental cause of this phenomenon lies in misspecification of the error structure: errors are conventionally assumed to be

Chenyan Fei, Dalin Zhang, Chen Melinda Dang

Effective and accurate diagnosis of fuel cell health status is crucial for ensuring the stable operation of fuel cell stacks. Among various parameters, high-frequency impedance serves as a critical indicator for assessing fuel cell state and health conditions. However, its online testing is prohibitively complex and costly. This paper employs a deep sparse a

Botong Gai

Let $H=U(\delta)$ be the universal enveloping algebra of finite dimension Lie algebra $\delta$. The central result of the paper is the classification of pre-Lie $H$-pseudoalgebras of low ranks over the Hopf algebra $H$. We firstly study pre-Lie pseudoalgebras that are free of rank $1$ over $H$. Then we introduce and classify a class of pre-Lie $H$-pseudoalge

Jundong Zhang, Yuhui Situ, Fanji Zhang, Rongji Deng

Tasks involving high-risk-high-return (HRHR) actions, such as obstacle crossing, often exhibit multimodal action distributions and stochastic returns. Most reinforcement learning (RL) methods assume unimodal Gaussian policies and rely on scalar-valued critics, which limits their effectiveness in HRHR settings. We formally define HRHR tasks and theoretically

Tingsong Xiao, Yao An Lee, Zelin Xu, Yupu Zhang

Disease progression modeling aims to characterize and predict how a patient's disease complications worsen over time based on longitudinal electronic health records (EHRs). For diseases such as type 2 diabetes, accurate progression modeling can enhance patient sub-phenotyping and inform effective and timely interventions. However, the problem is challenging

Santhosh Kumar Ravindran

We introduce CosmoCore, a neuroscience-inspired reinforcement learning (RL) architecture that integrates affective signals to enhance code generation in large language models (LLMs). Motivated by human and animal learning where embarrassment from mistakes drives rapid correction, as observed in training a puppy to avoid repeating errors after a single scoldi

Chenchen Tan, Youyang Qu, Xinghao Li, Hui Zhang

The increase in computing power and the necessity of AI-assisted decision-making boost the growing application of large language models (LLMs). Along with this, the potential retention of sensitive data of LLMs has spurred increasing research into machine unlearning. However, existing unlearning approaches face a critical dilemma: Aggressive unlearning compr

Dandan Chen, Tianjian Xu

We establish some new bilateral double-sum Rogers-Ramanujan identities involving parameters. As applications, these identities yield several new multi-sum Rogers-Ramanujan type identities. Our proofs utilize the theory of basic hypergeometric series in conjunction with the integral method.

Bin Sui, Yihao Wang, Jiaju Zhang

We investigate the short-interval expansion of the subsystem fidelity in two-dimensional conformal field theories (2D CFTs) using the operator product expansion (OPE) of twist operators. We obtain universal contributions from general quasiprimary operators valid for arbitrary 2D CFTs, along with specific results in free massless boson and fermion theories. T

Matteo Barbon, Nadir Ijaz, Marco Peloso

We investigate the Stochastic Gravitational Wave Background (SGWB) produced in models of axion inflation coupled to gauge fields. Achieving a detectable signal at Pulsar Timing Array, astrometry, or interferometer frequencies requires a sufficiently strong amplification of the gauge fields, at a level that induces significant backreaction on the inflaton bac

Michael Hersche, Samuel Moor-Smith, Thomas Hofmann, Abbas Rahimi

Diffusion models have demonstrated strong potential in language modeling, offering various advantages over traditional autoregressive approaches. Their ability to generate and revise entire responses in parallel enables faster generation and built-in self-correction mechanisms. Most modern diffusion-based language models employ masked diffusion, where decodi

Yingqi Fan, Anhao Zhao, Jinlan Fu, Junlong Tong

Multimodal Large Language Models (MLLMs) have achieved strong performance across vision-language tasks, but suffer from significant computational overhead due to the quadratic growth of attention computations with the number of multimodal tokens. Though efforts have been made to prune tokens in MLLMs, \textit{they lack a fundamental understanding of how MLLM

Rui Zhang, Shuaiting Yan, Wenru Jie, Jiayi Chen

Velocity-map imaging of electrons is a pivotal technique in chemical physics. A recent study reported a quantum offset as large as 0.2 cm-1 in velocity imaging-based electron spectrometry [Phys. Rev. Lett. 134, 043001 (2025)]. In this work, we assess the existence this offset through a combination of simulations and experiments. Our simulations reveal that t

Ryota Soga, Masataka Kobayashi, Tsukasa Shimizu, Shintaro Shiba

Event cameras, featuring high temporal resolution and high dynamic range, offer visual sensing capabilities comparable to conventional image sensors while capturing fast-moving objects and handling scenes with extreme lighting contrasts such as tunnel exits. Leveraging these properties, this study proposes a novel self-localization system that integrates vis

Connor Mooney, Ravi Shankar

We prove smoothness and interior derivative estimates for viscosity solutions to the special Lagrangian equation with almost negative phases and small enough semi-convexity. We show by example that the range of phases we consider and the semi-convexity condition are sharp. As an application, we find a new Liouville theorem for entire such solutions of the sp

Mika Feng, Pierre Gallin-Martel, Koichi Ito, Takafumi Aoki

Face recognition systems are designed to be robust against variations in head pose, illumination, and image blur during capture. However, malicious actors can exploit these systems by presenting a face photo of a registered user, potentially bypassing the authentication process. Such spoofing attacks must be detected prior to face recognition. In this paper,

Bingrong Liu, Jun Shi, Yushan Zheng

Class-incremental learning (CIL) for endoscopic image analysis is crucial for real-world clinical applications, where diagnostic models should continuously adapt to evolving clinical data while retaining performance on previously learned ones. However, existing replay-based CIL methods fail to effectively mitigate catastrophic forgetting due to severe domain

Nirai Hayakawa, Kazumasa Shimari, Kazuma Yamasaki, Hirotatsu Hoshikawa

Recently, research on predicting match outcomes in esports has been actively conducted, but much of it is based on match log data and statistical information. This research targets the FPS game VALORANT, which requires complex strategies, and aims to build a round outcome prediction model by analyzing minimap information in match footage. Specifically, based

M. Saifuzzaman Rafat, Akif Islam, Mohd Ruhul Ameen, Momen Khandoker Ope

Riverbank erosion is a serious environmental problem in Bangladesh, causing land loss, damage to infrastructure, and displacement of local communities. Manual analysis of satellite images is often slow and difficult to apply consistently across large river networks. This study uses a parameter-efficient adaptation of the Segment Anything Model (SAM) to detec

Pu Zhang, Yuwei Li, Xingyuan Xian, Guoming Tang

As the capabilities of Vision-Language Models (VLMs) advance, they can process increasingly large inputs, which, unlike in LLMs, generates significant visual token redundancy and leads to prohibitive inference costs. While many methods aim to reduce these costs by pruning visual tokens, existing approaches, whether based on attention or diversity, typically

Jiaqi Leng, Xiang Hu, Junxiong Wang, Jianguo Li

Effectively processing long contexts is a critical challenge for language models. While standard Transformers are limited by quadratic complexity and poor length extrapolation, alternative architectures like sliding window attention and state space models sacrifice the ability to effectively utilize the full context due to their fixed-size memory. Chunk-base

Jing-Bin Zhou, Shihai Yang

In this paper, we first characterize the polar decomposition of unbounded weighted composition operator pairs $\textbf{C}_{\phi,\omega}$ in an $L^2$-space. Based on this characterization, we introduce the $\lambda$-spherical mean transform $\mathcal{M}_\lambda(\textbf{C}_{\phi,\omega})$ for $\lambda\in[0,1]$. We then investigate the dense definiteness of $\m

Bo Zhang

The theory of rational motives admits several models, including those of Morel, Beilinson, Ayoub, and Voevodsky. An open question has been the equivalence of Voevodsky's Nisnevich-based $\mathrm{DM}(S, \mathbb{Q})$ with the others, which was only known over excellent and geometrically unibranch base schemes. In this paper, we prove that $\mathrm{DM}(S, \math

D. Abriola, M. Lombardi, C. Grillo, P. Bergamini

We present a new high precision parametric strong lensing total mass reconstruction of the Euclid Early Release Observations (ERO) galaxy cluster Abell 2390, at redshift z = 0.231. We include in this analysis 35 multiple images from 13 background sources, of which 25 are spectroscopically confirmed thanks to observations from the MUSE, spanning a redshift ra

Peiru Zheng, Yun Zhao, Zhan Gong, Hong Zhu

End-to-end autonomous driving has emerged as a promising paradigm for achieving robust and intelligent driving policies. However, existing end-to-end methods still face significant challenges, such as suboptimal decision-making in complex scenarios. In this paper,we propose SimpleVSF (Simple VLM-Scoring Fusion), a novel framework that enhances end-to-end pla

A. Aligayev, U. Jabbarli, U. Samadova, F. J. Dominguez-Gutierrez

In this study, we employ a multi-scale computational modeling approach, combining density functional theory (DFT) and self-consistent charge density functional tight binding (SCC-DFTB), to investigate hydrogen (H2) production and dissociation mechanisms from ammonia (NH3) and methane (CH4) on pristine and nickel-doped graphene. These two-dimensional material

Wenxun Wang, Shuchang Zhou, Wenyu Sun, Peiqin Sun

Transformers have shown remarkable performance in both natural language processing (NLP) and computer vision (CV) tasks. However, their real-time inference speed and efficiency are limited due to the inefficiency in Softmax and Layer Normalization (LayerNorm). Previous works based on function approximation suffer from inefficient implementation as they place

Vaibhav Rathore, Divyam Gupta, Biplab Banerjee

Generalized Category Discovery (GCD) aims to classify test-time samples into either seen categories** -- available during training -- or novel ones, without relying on label supervision. Most existing GCD methods assume simultaneous access to labeled and unlabeled data during training and arising from the same domain, limiting applicability in open-world sce

Alexander Aghili, Andy Bruce, Daniel Sabo, Sanya Murdeshwar

The rapid evolution of molecular dynamics (MD) methods, including machine-learned dynamics, has outpaced the development of standardized tools for method validation. Objective comparison between simulation approaches is often hindered by inconsistent evaluation metrics, insufficient sampling of rare conformational states, and the absence of reproducible benc

Dikran Dikranjan, Wei He, Dekui Peng

Generalizing results from \cite{DTk,DU} we study the fine structure of locally minimal (locally) precompact Abelian groups (these are the locally essential subgroups $G$ of LCA groups $L$, i.e., such that $G$ non-trivially meets all ``small" closed subgroup of $L$). More precisely we prove that if $G$ is a dense locally minimal and sequentially closed subgro

Runlin Lei, Lu Yi, Mingguo He, Pengyu Qiu

While Graph Neural Networks (GNNs) and Large Language Models (LLMs) are powerful approaches for learning on Text-Attributed Graphs (TAGs), a comprehensive understanding of their robustness remains elusive. Current evaluations are fragmented, failing to systematically investigate the distinct effects of textual and structural perturbations across diverse mode

Nicolas Robert, Fabien Gandon, Maxime Lefrançois

Agile and collaborative approaches to ontologies design are crucial because they contribute to making them userdriven, up-to-date, and able to evolve alongside the systems they support, hence proper continuous validation tooling is required to ensure ontologies match developers' requirements all along their development. We propose OLIVAW (Ontology Long-lived

Mu Qiao, Romain Martin, Lukas Homeier, Ivan Morera

Understanding how particles bind into composite objects is a ubiquitous theme in physics, from the formation of molecules to hadrons in quantum chromodynamics and the pairing of charge carriers in superconductors. The formation of bound states usually originates from attractive interactions between particles. However, the binding can also arise purely from t

Aaron Bernstein, Joakim Blikstad, Jason Li, Thatchaphol Saranurak

We give a combinatorial algorithm for computing exact maximum flows in directed graphs with $n$ vertices and edge capacities from $\{1,\dots,U\}$ in $\tilde{O}(n^{2}\log U)$ time, which is near-optimal on dense graphs. This shaves an $n^{o(1)}$ factor from the recent result of [Bernstein-Blikstad-Saranurak-Tu FOCS'24] and, more importantly, greatly simplifie

Haonan He, Yufeng Zheng, Jie Song

Photorealistic 3D head avatars are vital for telepresence, gaming, and VR. However, most methods focus solely on facial regions, ignoring natural hand-face interactions, such as a hand resting on the chin or fingers gently touching the cheek, which convey cognitive states like pondering. In this work, we present a novel framework that jointly learns detailed

Ningan Lai, Silu Yin, Yi Zhou

In this paper, we introduce a modified scaling Morawetz multiplier, which produces a weighted Morawetz type energy (non-negative) estimate for the inhomogeneous wave equation in $\mathbb{R}^2$. With this estimate in hand, an alternative proof of global existence for the Cauchy problem of quasilinear wave equation with small and compactly supported data is gi

Yingzi Han, Jiakai He, Chuanlong Xie, Jianping Li

Automated plankton recognition models face significant challenges during real-world deployment due to distribution shifts (Out-of-Distribution, OoD) between training and test data. This stems from plankton's complex morphologies, vast species diversity, and the continuous discovery of novel species, which leads to unpredictable errors during inference. Despi

Xueying Yu, Haitian Yue, Zehua Zhao

In this work, we introduce and study nonlinear Schr\"odinger equations (NLS) with anisotropic dispersion, where the standard Laplacian acts on the Euclidean variable \(x \in \mathbb{R}^d\), and an Ornstein-Uhlenbeck ($\mathcal{OU}$) operator governs the confined direction \(\alpha \in \mathbb{R}\). We consider models with two natural variants of $\mathcal{OU

Yann Bugeaud, Hajime Kaneko, Dong Han Kim

Let $\xi$ be a real number and $b \ge 2$ an integer. We study the relationship between the irrationality exponent of $\xi$ and the subword complexity $p(n, \mathbf{x})$ of the $b$-ary expansion $\mathbf{x}$ of $\xi$, where $p(n, \mathbf{x})$ counts the number of distinct blocks of length $n$ in $\mathbf{x}$, for $n \ge 1$. If the irrationality exponent of $\

Lakshmikanta Sau, Priyadarshi Mukherjee, Sasthi C. Ghosh

Reconfigurable intelligent surface (RIS) is a cutting-edge communication technology that has been proposed as aviable option for beyond fifth-generation wireless communication networks. This paper investigates various group selection strategies in the context of grouping-based self-sustainable RIS-aided device-to-device (D2D) communication with spatially cor

Muhammad Wahid Akram, Keshav Sood, Muneeb Ul Hassan

Globally, individuals and organizations employ Quick Response (QR) codes for swift and convenient communication. Leveraging this, cybercriminals embed falsify and misleading information in QR codes to launch various phishing attacks which termed as Quishing. Many former studies have introduced defensive approaches to preclude Quishing such as by classifying

Hezhishi Jiang, Liyan Xu, Tianshu Li, Jintong Tang

Urban science has largely relied on universal models, rendering the heterogeneous and locally specific nature of cities effectively invisible. Here we introduce a topological framework that defines and detects localities in human mobility networks. We empirically demonstrate that these human mobility network localities are rigorous geometric entities that ma

Melik Ozolcer, Sang Won Bae

We study a web-deployed, tool-augmented LLM health coach with real users. In a pilot with seven users (280 rated turns), offline policy evaluation (OPE) over factorized decision heads (Tool/Style) shows that a uniform heavy-tool policy raises average value on logs but harms specific subgroups, most notably low-health-literacy/high-self-efficacy users. A ligh

Shun Huang, Wenlu Xing, Shijia Geng, Hailong Wang

Malignant ventricular arrhythmias (VT/VF) following acute myocardial infarction (AMI) are a major cause of in-hospital death, yet early identification remains a clinical challenge. While traditional risk scores have limited performance, end-to-end deep learning models often lack the interpretability needed for clinical trust. This study aimed to develop a hy

Feihong Yan, Peiru Wang, Yao Zhu, Kaiyu Pang

Masked Autoregressive (MAR) models promise better efficiency in visual generation than autoregressive (AR) models for the ability of parallel generation, yet their acceleration potential remains constrained by the modeling complexity of spatially correlated visual tokens in a single step. To address this limitation, we introduce Generation then Reconstructio

Luan Yuxi, Rana Mondal

We propose two interrelated advances in the theory of adjointable operators on Hilbert C*-modules. First, we give a set of equivalent, verifiable conditions characterizing positivity of general $n\times n$ block operator matrices acting on finite direct sums of Hilbert C*-modules. Our conditions generalize and remove several classical range-closedness and Mo

Roland Croft, Brian Du, Darcy Joseph, Sharath Kumar

Face Recognition (FR) models have been shown to be vulnerable to adversarial examples that subtly alter benign facial images, exposing blind spots in these systems, as well as protecting user privacy. End-to-end FR systems first obtain preprocessed faces from diverse facial imagery prior to computing the similarity of the deep feature embeddings. Whilst face

Zihao Ding, Mufeng Zhu, Yao Liu

Model Context Protocol (MCP) has recently gained increased attention within the AI community for providing a standardized way for large language models (LLMs) to interact with external tools and services, significantly enhancing their capabilities. However, the inclusion of extensive contextual information, including system prompts, MCP tool definitions, and

Yong Wu, Yanwei Fu, Shouyan Wang, Yizhou Wang

Inferring causal relationships between variable pairs in the observational study is crucial but challenging, due to the presence of unmeasured confounding. While previous methods employed the negative controls to adjust for the confounding bias, they were either restricted to the discrete setting (i.e., all variables are discrete) or relied on strong assumpt

Anbang Yang, Xiaoyu Nie, Hao Lin Yu, Yiming Liu

The narrow transition from the lowest rovibrational level of the $\mathrm{X}^{1}\Sigma^{+}$ electronic ground state to the lowest vibrational level of the $\mathrm{b}^{3}\Pi_{0}$ potential provides opportunities for achieving magic-wavelength trapping of ultracold bialkali molecules for enhancing their rotational coherence times. Guided by existing spectrosc

Jinrui Zhang

Foundation models - already transformative in domains such as natural language processing - are now starting to emerge for time-series tasks in finance. While these pretrained architectures promise versatile predictive signals, little is known about how they shape the risk profiles of the trading strategies built atop them, leaving practitioners reluctant to

Maria Deolinda Santana, Cleyton Magalhaes, Ronnie de Souza Santos

\textit{Background:} The use of large language models in software testing is growing fast as they support numerous tasks, from test case generation to automation, and documentation. However, their adoption often relies on informal experimentation rather than structured guidance. \textit{Aims:} This study investigates how software testing professionals use LL

Shuzheng Gao, Eric John Li, Man Ho Lam, Jingyu Xiao

Large foundation models are fundamentally transforming the software engineering landscape, demonstrating exceptional capabilities across diverse tasks such as code generation, debugging, and testing. Despite this rapid progress, a significant gap remains in how to comprehensively evaluate these models' trustworthiness in real-world software engineering scena

Guanjie Cheng, Siyang Liu, Xinkui Zhao, Yishan Chen

Mobile edge crowdsensing (MECS) enables large-scale real-time sensing services, but its continuous data collection and transmission pipeline exposes terminal devices to dynamic privacy risks. Existing privacy protection schemes in MECS typically rely on static configurations or coarse-grained adaptation, making them difficult to balance privacy, data utility

Tongji Gao, Zhan Li, Lei Zhang

Let $X$ be a smooth projective variety with a nef anticanonical divisor over an algebraically closed field of characteristic $p>0$. In this paper, we establish a precise structure of $X$ under the condition that $a_X: X \to {\rm Alb}(X)$ is of relative dimension one.

Derda Kaymak, Gyuhak Kim, Tomoya Kaichi, Tatsuya Konishi

In classic supervised learning, once a model is deployed in an application, it is fixed. No updates will be made to it during the application. This is inappropriate for many dynamic and open environments, where unexpected samples from unseen classes may appear. In such an environment, the model should be able to detect these novel samples from unseen classes

Christopher Simmonds, Matt Visser

Herein we shall argue for the utility of "spacetime geodesy", a point of view where one delays as long as possible worrying about dynamical equations, in favour of the maximal utilization of both symmetries and geometrical features. This closely parallels Weinberg's distinction between "cosmography" and "cosmology", wherein maximal utilization of both the sy

Daniel Nichols, Konstantinos Parasyris, Charles Jekel, Abhinav Bhatele

Language models are now prevalent in software engineering with many developers using them to automate tasks and accelerate their development. While language models have been tremendous at accomplishing complex software engineering tasks, there are still many areas where they fail to deliver desirable results, for instance code performance related tasks. Task

Yinghui Wang, Xinyu Zhang, Peng Du

Generating editable, parametric CAD models from a single image holds great potential to lower the barriers of industrial concept design. However, current multi-modal large language models (MLLMs) still struggle with accurately inferring 3D geometry from 2D images due to limited spatial reasoning capabilities. We address this limitation by introducing GACO-CA

Mohammadamin Lari

This paper introduces a novel two-stage framework for online mitigation of False Data Injection (FDI) signals to improve the resiliency of Networked Control Systems (NCSs) and ensure their safe operation in the presence of malicious activities. The first stage involves meta learning to select a base time series forecasting model within a stacked ensemble lea

Christopher J. Whyte, Hugh R. Wilson, Shay Tobin, Brandon R. Munn

When conflicting images are presented to either eye, binocular fusion is disrupted. Rather than experiencing a blend of both percepts, often only one eye's image is experienced, whilst the other is suppressed from awareness. Importantly, suppression is transient - the two rival images compete for dominance, with stochastic switches between mutually exclusive

Hyunjin Choo, Fanchen Bu, Hyunjin Hwang, Young-Gyu Yoon

Higher-order interactions (HOIs) in complex systems, such as scientific collaborations, multi-protein complexes, and multi-user communications, are commonly modeled as hypergraphs, where each hyperedge (i.e., a subset of nodes) represents an HOI among the nodes. Given a hypergraph, hyperedge prediction aims to identify hyperedges that are either missing or l

Kwanghee Park, Ju-Yeon Jo, Yoshitaka Tanimura

Molecular vibrations in solutions, especially OH stretching and bending in water, drive ultrafast energy relaxation and dephasing in chemical and biological systems. We present a machine learning approach for constructing system-bath models of intramolecular vibrations in solution, compatible with quantum simulations via the hierarchical equations of motion

Fan-Shun Meng, Shuhei Shinzato, Zhiqiang Zhao, Jun-Ping Du

A neuroevolution potential (NEP) for the ternary $\alpha$-Fe--C--H system was developed based on a database generated from spin-polarized density functional theory (DFT) calculations, achieving empirical potential efficiency with DFT accuracy. At the same power consumption, simulation speeds using NEP are comparable to, or even faster than, those with bond o

Heng Zhang, Wei-Hsing Huang, Gokhan Solak, Arash Ajoudani

We present OmniVIC, a universal variable impedance controller (VIC) enhanced by a vision language model (VLM), which improves safety and adaptation in any contact-rich robotic manipulation task to enhance safe physical interaction. Traditional VIC have shown advantages when the robot physically interacts with the environment, but lack generalization in unsee

Yu Gao, Anqing Jiang, Yiru Wang, Wang Jijun

Conventional end-to-end (E2E) driving models are effective at generating physically plausible trajectories, but often fail to generalize to long-tail scenarios due to the lack of essential world knowledge to understand and reason about surrounding environments. In contrast, Vision-Language-Action (VLA) models leverage world knowledge to handle challenging ca

Linhan Xia, Mingzhan Yang, Jingjing Wang, Ziwei Yan

Transformer-based large language models (LLMs) are increasingly being adopted in networking research to address domain-specific challenges. However, their quadratic time complexity and substantial model sizes often result in significant computational overhead and memory constraints, particularly in resource-constrained environments. Drawing inspiration from

Subin Lin, Chuanbo Hua

Heating, Ventilation, and Air-Conditioning (HVAC) systems account for a substantial share of global building energy use, making reliable anomaly detection essential for improving efficiency and reducing emissions. Classical rule-based approaches offer explainability but lack adaptability, while deep learning methods provide predictive power at the cost of tr

Phi-Hung Hoang, Nam-Thuan Trinh, Van-Manh Tran, Thi-Thu-Hong Phan

Accurate assessment of fish freshness remains a major challenge in the food industry, with direct consequences for product quality, market value, and consumer health. Conventional sensory evaluation is inherently subjective, inconsistent, and difficult to standardize across contexts, often limited by subtle, species-dependent spoilage cues. To address these

Chungil Lee, Yoshiaki Abe, Yu Kawano, Tomoki Yamazaki

An interface control principle is proposed for unsteady fluid-structure in- teraction (FSI) analyses. This principle introduces a method of explicitly controlling the interface motion in the temporal direction to minimize the residual force on the interface, which is defined as the discrepancy between the fluid and structural forces. The interface model is c

Shantnav Agarwal, Javier Alonso-Mora, Sihao Sun

Existing approaches for transporting and manipulating cable-suspended loads using multiple UAVs along reference trajectories typically rely on either centralized control architectures or reliable inter-agent communication. In this work, we propose a novel machine learning based method for decentralized kinodynamic planning that operates effectively under par

Bishal Chhetri, B. V. Rathish Kumar

The dynamics of tumor-immune interactions within a complex tumor microenvironment are typically modeled using a system of ordinary differential equations or partial differential equations. These models introduce some unknown parameters that need to be estimated accurately and efficiently from the limited and noisy experimental data. Moreover, due to the intr

Xiaoxue Ren, Jun Wan, Yun Peng, Zhongxin Liu

Large Language Models (LLMs) have demonstrated significant capability in code generation, but their potential in code efficiency optimization remains underexplored. Previous LLM-based code efficiency optimization approaches exclusively focus on function-level optimization and overlook interaction between functions, failing to generalize to real-world develop

Joshua Tomlin

In ordinary Seiberg-Witten theory, there are well known connected sum formulae such as the vanishing formula and the blow up formula. For families Seiberg-Witten theory, there are results such as Liu's families blow-up formula and Baraglia-Konno's gluing formula, but these have limited uses. In this paper, we prove a general connected sum formula which incor

Jhen-Dong Lin, Pao-Wen Tu, Kuan-Yi Lee, Neill Lambert

Certifying nonclassical correlations typically requires access to all subsystems, presenting a major challenge in open quantum systems coupled to inaccessible environments. Recent works have shown that, in autonomous pure dephasing scenarios, quantum discord with the environment can be certified from system-only dynamics via the Hamiltonian ensemble formulat