October 2025 arXiv papers — page 17

Enpu Liang

This study aimed to systematically identify and quantify risks for drug-induced rhabdomyolysis (DIR) using real-world data and to propose an evidence-based risk mitigation framework. We conducted a retrospective pharmacovigilance study using the FDA Adverse Event Reporting System (FAERS) database from Q1 2005 to Q1 2025. A two-stage analysis involved initial

Dongjun Jang, Youngchae Ahn, Hyopil Shin

Current LLM evaluations often rely on a single instruction template, overlooking models' sensitivity to instruction style-a critical aspect for real-world deployments. We present RCScore, a multi-dimensional framework quantifying how instruction formulation affects model responses. By systematically transforming benchmark problems into multiple instruction s

Tim Browning, Will Sawin

Among the set of hypersurfaces of degree $d$ and dimension $\ell$ defined by the vanishing of a homogeneous polynomial with coefficients $\pm 1$, we investigate the probability that a hypersurface contains a rational point as $d$ and $\ell$ tend to infinity.

Hitomi Jin Ling Tee, Chaoren Wang, Zijie Zhang, Zhizheng Wu

The evaluation of intelligibility for TTS has reached a bottleneck, as existing assessments heavily rely on word-by-word accuracy metrics such as WER, which fail to capture the complexity of real-world speech or reflect human comprehension needs. To address this, we propose Spoken-Passage Multiple-Choice Question Answering, a novel subjective approach evalua

Yoshihiro Nambu

We propose a practical hybrid decoding scheme for the parity-encoding architecture. This architecture was first introduced by N. Sourlas as a computational technique for tackling hard optimization problems, especially those modeled by spin systems such as the Ising model and spin glasses, and reinvented by W. Lechner, P. Hauke, and P. Zoller to develop quant

Avinash Kadimisetty, Arun Rajagopalan, Vijendra SK

Reducing preventable hospital readmissions is a national priority for payers, providers, and policymakers seeking to improve health care and lower costs. The rate of readmission is being used as a benchmark to determine the quality of healthcare provided by the hospitals. In thisproject, we have used machine learning techniques like Logistic Regression, Rand

Mohammed Rafiq Namiq

In this paper, we introduce initially Cohen-Macaulay modules over a commutative Noetherian local ring $R$, a new class of $R$-modules that generalizes both Cohen-Macaulay and sequentially Cohen-Macaulay modules. A finitely generated $R$-module $N$ is initially Cohen-Macaulay if its depth is equal to its initial dimension, an invariant defined as the infimum

Jinho Choi, Hyesu Lim, Steffen Schneider, Jaegul Choo

Dataset bias, where data points are skewed to certain concepts, is ubiquitous in machine learning datasets. Yet, systematically identifying these biases is challenging without costly, fine-grained attribute annotations. We present ConceptScope, a scalable and automated framework for analyzing visual datasets by discovering and quantifying human-interpretable

Yunxiao Shi, Shuo Yang, Yixin Su, Rui Zhang

Modern data-centric AI needs precise per-sample influence. Standard SGD-IE approximates leave-one-out effects by summing per-epoch surrogates and ignores cross-epoch compounding, which misranks critical examples. We propose ACC-SGD-IE, a trajectory-aware estimator that propagates the leave-one-out perturbation across training and updates an accumulative infl

Shunya Minami, Tatsuya Ishigaki, Ikko Hamamura, Taku Mikuriya

Large language models are now integrated into many scientific workflows, accelerating data analysis, hypothesis generation, and design space exploration. In parallel with this growth, there is a growing need to carefully evaluate whether models accurately capture domain-specific knowledge and notation, since general-purpose benchmarks rarely reflect these re

Songxin Lei, Qiongyan Wang, Yanchen Zhu, Hanyu Yao

Public resource allocation involves the efficient distribution of resources, including urban infrastructure, energy, and transportation, to effectively meet societal demands. However, existing methods focus on optimizing the movement of individual resources independently, without considering their capacity constraints. To address this limitation, we propose

Allen Schmaltz

We introduce Similarity-Distance-Magnitude (SDM) language models (LMs), which are sequence prediction models fine-tuned to maximize the proportion of generations in the well-calibrated, high-probability region partitioned by a final-layer SDM activation layer used for binary classification of instruction-following. We demonstrate that existing pre-trained de

Shikhar Tuli, James Seale Smith, Haris Jeelani, Chi-Heng Lin

Large language models (LLMs) have significantly advanced generative applications in natural language processing (NLP). Recent trends in model architectures revolve around efficient variants of transformers or state-space/gated-recurrent models (SSMs, GRMs). However, prevailing SSM/GRM-based methods often emulate only a single attention head, potentially limi

Changjian Xie, Cheng Wang

To address the magnetization dynamics in ferromagnetic materials described by the Landau-Lifshitz-Gilbert equation under large damping parameters, a third-order accurate numerical scheme is developed by building upon a second-order method \cite{CaiChenWangXie2022} and leveraging its efficiency. This method boasts two key advantages: first, it only involves s

Dou Dai, Qiuqi Li, Huailing Song

This paper is dedicated to enhancing the computational efficiency of traditional parallel-in-time methods for solving stochastic initial-value problems. The standard parareal algorithm often suffers from slow convergence when applied to problems with stochastic inputs, primarily due to the poor quality of the initial guess. To address this issue, we propose

Haruki Hoshino, Jungjin Park, Osamu Kaneko, Kiminao Kogiso

Edge computing alleviates the computation burden of data-driven control in cyber-physical systems (CPSs) by offloading complex processing to edge servers. However, the increasing sophistication of cyberattacks underscores the need for security measures that go beyond conventional IT protections and address the unique vulnerabilities of CPSs. This study propo

Yanran Tang, Ruihong Qiu, Xue Li, Zi Huang

Legal case retrieval (LCR) is a cornerstone of real-world legal decision making, as it enables practitioners to identify precedents for a given query case. Existing approaches mainly rely on traditional lexical models and pretrained language models to encode the texts of legal cases. Yet there are rich information in the relations among different legal entit

Sagar Pandhare, Divya Kappara, Siuli Mukhopadhyay

Spatial regression models have a variety of applications in several fields ranging from economics to public health. Typically, it is of interest to select important exogenous predictors of the spatially autocorrelated response variable. In this paper, we propose variable selection in linear spatial lag models by means of the focussed information criterion (F

Shuma Komatsu

In the work of C. Donovan and N. A. Scoville, the homotopy type of the Morse complex of the extended star graph which is obtained as the one-point union of n paths of length 2 was determined by using star clusters and Cluster Lemma. In this paper, we determine the homotopy type of the Morse complex of extended star graph consisting of a path of length 1 and

Xiaohui Gao

We demonstrate substantial field enhancement in plasma nanoshells through high-order Mie resonances using combined Mie theory and particle-in-cell simulations. Optimal shell geometries yield approximately threefold electric field enhancement for 800 nm irradiation, with transient buildup times of tens of femtoseconds before plasma expansion disrupts resonanc

Liming Dong, Sung Une Lee, Zhenchang Xing, Muhammad Ejaz Ahmed

The increasing frequency and sophistication of software supply chain attacks pose severe risks to critical infrastructure sectors, threatening national security, economic stability, and public safety. Despite growing awareness, existing security practices remain fragmented and insufficient, with most frameworks narrowly focused on isolated life cycle stages

Wenli Xiao, Haotian Lin, Andy Peng, Haoru Xue

Supervised fine-tuning (SFT) has become the de facto post-training strategy for large vision-language-action (VLA) models, but its reliance on costly human demonstrations limits scalability and generalization. We propose Probe, Learn, Distill (PLD), a three-stage plug-and-play framework that improves VLAs through residual reinforcement learning (RL) and dist

Wontae Choi, Jaelin Lee, Hyung Sup Yun, Byeungwoo Jeon

Accurate estimation of motion information is crucial in diverse computational imaging and computer vision applications. Researchers have investigated various methods to extract motion information from a single blurred image, including blur kernels and optical flow. However, existing motion representations are often of low quality, i.e., coarse-grained and in

Hasnain Iqbal, Zerina Begum, Kazi Sakib

Flaky tests can make automated software testing unreliable due to their unpredictable behavior. These tests can pass or fail on the same code base on multiple runs. However, flaky tests often do not refer to any fault, even though they can cause the continuous integration (CI) pipeline to fail. A common type of flaky test is the order-dependent (OD) test. Th

Satoshi Kikuchi, Masaya Kato, Tsuyoshi Tasaki

Self-localization on a 3D map by using an inexpensive monocular camera is required to realize autonomous driving. Self-localization based on a camera often uses a convolutional neural network (CNN) that can extract local features that are calculated by nearby pixels. However, when dynamic obstacles, such as people, are present, CNN does not work well. This s

Dheer Noal Desai, Vishal Gupta

Let $\mathcal{D}_{n,\tau}$ be the set of all simple connected graphs of order $n$ and dissociation number $\tau.$ In this paper, we study the minimum size and the minimum spectral radius of graphs in $\mathcal{D}_{n,\tau}$ in connection with Tur\'an-type problems for complete multipartite graphs. We characterize the Tur\' an graphs for several complete multi

Sen-Peng Eu, Yi-Lin Lee

Recently, Brualdi and Cao studied $I_k$-avoiding $(0,1)$-matrices by decomposing them into zigzag paths and proved that the maximum number of $1$'s in such a matrix is given by an exact formula. We further study the structure of maximal $I_k$-avoiding $(0,1)$-matrices (IAMs) by interpreting them as families of non-intersecting lattice paths on the square lat

Renhao Li, Jianhong Tu, Yang Su, Yantao Liu

Reward models (RMs) play a critical role in aligning large language models (LLMs) with human preferences. Yet in the domain of tool learning, the lack of RMs specifically designed for function-calling tasks has limited progress toward more capable agentic AI. We introduce ToolRM, a family of lightweight reward models tailored for general tool-use scenarios.

Juncong Zhou, Chao Hu, Guanlin Wu, Zixiang Ren

This paper investigates the construction of channel knowledge map (CKM) from sparse channel measurements. Dif ferent from conventional two-/three-dimensional (2D/3D) CKM approaches assuming fixed base station configurations, we present a six-dimensional (6D) CKM framework named bidirectional wireless Gaussian splatting (BiWGS), which is capable of mod eling

Maarten P. Scholl, Mahmoud Mahfouz, Anisoara Calinescu, J. Doyne Farmer

While investment funds publicly disclose their objectives in broad terms, their managers optimize for complex combinations of competing goals that go beyond simple risk-return trade-offs. Traditional approaches attempt to model this through multi-objective utility functions, but face fundamental challenges in specification and parameterization. We propose a

Vincent Colin, Ko Honda, Yin Tian

The goal of this paper is twofold: (i) define a symplectic Khovanov type homology for a transverse link in a fibered closed $3$-manifold $M$ (with an auxiliary choice of a homotopy class of loops that intersect each fiber once) and (ii) give conjectural combinatorial dga descriptions of surface categories that appear in (i). These dgas are higher-dimensional

Li Zhenchao, Yang Aixiang

In the field of deep space radiation shielding design, traditional high-Z metals are being progressively replaced by novel low-Z materials such as hydrogenated graphene foam, polyethylene-carbon nanotube composite fibers, and boron-rich hydrogen-containing metal-organic frameworks. This transition stems from the constraints of the "gram-scale weight reductio

Qiumeng Li, Xinxi Yang, Suhong Zhou

As emerging mobility modes continue to expand, many cities face declining bus ridership, increasing fiscal pressure to sustain underutilized routes, and growing inefficiencies in resource allocation. This study employs an agent-based modelling (ABM) approach calibrated through a large language model (LLM) using few-shot learning to examine how progressive bu

Sebastian V. Colom, Magnus A. Haw, Jocelino Rodrigues

Langmuir probe diagnostics are a cornerstone of plasma characterization, providing critical measurements of electron temperature, electron density, and plasma potential. However, conventional swept Langmuir probes and other traditional electrostatic probes often lack the temporal resolution necessary to capture transient plasma behavior in dynamic environmen

Shubham Desai, Malapeta Hemasundara Rao, Sai Sidhardh

This study presents a meshfree two-dimensional fractional-order Element-Free Galerkin (2D f-EFG) method as a viable alternative to conventional mesh-based FEM for a numerical solution of (spatial) fractional-order differential equations (FDEs). The previously developed one-dimensional f-EFG solver offers a limited demonstration of the true efficacy of EFG fo

Jiaqi Wang, Xiao Yang, Kai Sun, Parth Suresh

Wearable devices such as smart glasses are transforming the way people interact with their surroundings, enabling users to seek information regarding entities in their view. Multi-Modal Retrieval-Augmented Generation (MM-RAG) plays a key role in supporting such questions, yet there is still no comprehensive benchmark for this task, especially regarding weara

Emilio Mastriani, Alessandro Costa, Federico Incardona, Kevin Munari

In this study, we investigate the effectiveness of advanced feature engineering and hybrid model architectures for anomaly detection in a multivariate industrial time series, focusing on a steam turbine system. We evaluate the impact of change point-derived statistical features, clustering-based substructure representations, and hybrid learning strategies on

A. V. Guglielmi

An attempt to construct a phenomenological theory of earthquakes is being undertaken by a small research team, which includes the author. We have only taken the first steps towards goal, but the results already obtained convincingly demonstrate the feasibility of the project. This paper is a kind of commentary on the path taken. The paper explains the essenc

Hyuntae Park, Yeachan Kim, SangKeun Lee

Molecule and text representation learning has gained increasing interest due to its potential for enhancing the understanding of chemical information. However, existing models often struggle to capture subtle differences between molecules and their descriptions, as they lack the ability to learn fine-grained alignments between molecular substructures and che

Mostafizar Khandakar, Bratati Pal, Palaniappan Vellaisamy

In this paper, we introduce and study two time-changed variants of the generalized fractional Skellam process. These are obtained by time-changing the generalized fractional Skellam process with an independent L\'evy subordinator with finite moments of any order and its inverse, respectively. We call the introduced processes the time-changed generalized frac

Haorong Qiu, Guodong Yu, Peiyou Jiang, Guoyong Fu

An optimized stellarator at finite plasma beta is realized by single-stage optimization of simply modifying the coil currents of the Compact Stellarator with Simple Coils (CSSC)[Yu et al., J. Plasma Physics 88,905880306 (2022)]. The CSSC is an optimized stellarator obtained by direct optimization via coil shapes, with its coil topology similar to that of the

Sudipto Das Sukanto, Diponker Roy, Fahim Shakil, Nirjhar Singha

Modes of transportation vary across countries depending on geographical location and cultural context. In South Asian countries rickshaws are among the most common means of local transport. Based on their mode of operation, rickshaws in cities across Bangladesh can be broadly classified into non-auto (pedal-powered) and auto-rickshaws (motorized). Monitoring

Yuan Yuan

This paper is concerned with the asymptotic stabilities of the inviscid and viscous shocks for the scalar conservation laws on the half-line $(-\infty,0)$ with shock speed $s<0$, subjected to the time-periodic boundary condition, which arises from the classical piston problems for fluid mechanics. Despite the importance, how time-periodic boundary conditions

Xiaoyu Zhang, Lile Wang, Yang Gao, Yao Zhou

We present direct numerical simulations demonstrating deflagration-to-detonation transition (DDT) driven by oxygen flames in Type Ia supernova progenitors. Using the Castro hydrodynamics code coupled with the ``aprox13'' 13-isotope nuclear network, we simulate combustion in isolated fuel regions where oxygen flames trail carbon flames. In a fiducial one-dime

Shunjie-Fabian Zheng, Hyeonjun Lee, Thijs Kooi, Ali Diba

Large annotated datasets are essential for training robust Computer-Aided Diagnosis (CAD) models for breast cancer detection or risk prediction. However, acquiring such datasets with fine-detailed annotation is both costly and time-consuming. Vision-Language Models (VLMs), such as CLIP, which are pre-trained on large image-text pairs, offer a promising solut

Jiaying Di, Kunlun Wang, Jing Xu, Wen Chen

This paper investigates a novel computation and communication co-design framework for large-scale split learning in intelligent reflecting surface (IRS)-assisted internet of things (IoT) networks integrated with digital twin (DT) technique. The considered system consists of a multi-antenna access point (AP), multiple heterogeneous user devices (UDs), and an

Wei Shang, Wanying Zhang, Shuhang Gu, Pengfei Zhu

Arbitrary-scale video super-resolution (AVSR) aims to enhance the resolution of video frames, potentially at various scaling factors, which presents several challenges regarding spatial detail reproduction, temporal consistency, and computational complexity. In this paper, we propose a strong baseline BasicAVSR for AVSR by integrating four key components: 1)

Kexing Liu

Human Activity Recognition (HAR) via Wi-Fi Channel State Information (CSI) presents a privacy-preserving, contactless sensing approach suitable for smart homes, healthcare monitoring, and mobile IoT systems. However, existing methods often encounter computational inefficiency, high latency, and limited feasibility within resource-constrained, embedded mobile

Kexing Liu

WiFi Channel State Information (CSI)-based human activity recognition (HAR) provides a privacy-preserving, device-free sensing solution for smart environments. However, its deployment on edge devices is severely constrained by domain shift, where recognition performance deteriorates under varying environmental and hardware conditions. This study presents max

Sena Aydin, Valerii Andreichev, Pantelis Maragkoudakis, Markus Meuwly

Molecular dynamics simulations for tripeptides in the gas phase and in solution using empirical and machine-learned energy functions are presented. For cationic AAA a machine-learned potential energy surface (ML-PES) trained on MP2 reference data yields quantitative agreement with measured splittings of the amide-I vibrations. Experimental spectroscopy in so

Annan Li, Chufan Wu, Zengle Ge, Yee Hin Chong

Large language models (LLMs) are catalyzing the development of autonomous AI research agents for scientific and engineering discovery. We present FM Agent, a novel and general-purpose multi-agent framework that leverages a synergistic combination of LLM-based reasoning and large-scale evolutionary search to address complex real-world challenges. The core of

Huanlin Gao, Ping Chen, Fuyuan Shi, Chao Tan

We present LeMiCa, a training-free and efficient acceleration framework for diffusion-based video generation. While existing caching strategies primarily focus on reducing local heuristic errors, they often overlook the accumulation of global errors, leading to noticeable content degradation between accelerated and original videos. To address this issue, we

Bo Pang, Deqian Kong, Silvio Savarese, Caiming Xiong

Reinforcement learning (RL) can elicit strong reasoning in large language models (LLMs), yet most open efforts focus on math and code. We propose Reasoning Curriculum, a simple two-stage curriculum that first elicits reasoning skills in pretraining-aligned domains such as math, then adapts and refines these skills across other domains via joint RL. Stage 1 p

Hahjin Lee, Young J. Kim

Trajectory planning for mobile robots in cluttered environments remains a major challenge due to narrow passages, where conventional methods often fail or generate suboptimal paths. To address this issue, we propose the adaptive trajectory refinement algorithm, which consists of two main stages. First, to ensure safety at the path-segment level, a segment-wi

Xin Hu, Pengfei Xu, Jin Zhou, Hongbo Fu

Structured layouts are preferable in many 2D visual contents (\eg, GUIs, webpages) since the structural information allows convenient layout editing. Computational frameworks can help create structured layouts but require heavy labor input. Existing data-driven approaches are effective in automatically generating fixed layouts but fail to produce layout stru

Lihe Ding, Shaocong Dong, Yaokun Li, Chenjian Gao

Part-based 3D generation holds great potential for various applications. Previous part generators that represent parts using implicit vector-set tokens often suffer from insufficient geometric details. Another line of work adopts an explicit voxel representation but shares a global voxel grid among all parts; this often causes small parts to occupy too few v

Minseo Kwon, Young J. Kim

Task and Motion Planning (TAMP) integrates high-level task planning with low-level motion feasibility, but existing methods are costly in long-horizon problems due to excessive motion sampling. While LLMs provide commonsense priors, they lack 3D spatial reasoning and cannot ensure geometric or dynamic feasibility. We propose a kinodynamic TAMP planner based

Naoki Takeishi, Junya Kobayashi, Shigeo Wada, Satoshi Ii

The relationship between the spatiotemporal distribution of oxygen transport and blood flow dynamics, accounting for the motion and deformation of individual red blood cells (RBCs), is of fundamental importance for understanding microcirculation systems. Three-dimensional (3D) modeling is indispensable for addressing complex oxygen transport and cellular beh

Boqin Zhuang, Jiacheng Qiao, Mingqian Liu, Mingxing Yu

The inference cost of Large Language Models (LLMs) has become a critical factor in determining their commercial viability and widespread adoption. This paper introduces a quantitative ``economics of inference'' framework, treating the LLM inference process as a compute-driven intelligent production activity. We analyze its marginal cost, economies of scale,

Max Aires, Jeff Kahn

An old conjecture of Kahn and Saks says, roughly, that any poset $P$ of large enough width contains elements $x,y$ which are "balanced" in the sense that the probability that $x$ precedes $y$ in a uniformly random linear extension of $P$ is close to $1/2$. We show this implies the seemingly stronger statement that the same conclusion holds if, instead of lar

Yuanhao Yan, Li He

Fej\'er's theorem guarantees norm convergence of Ces\`aro means of Taylor partial sums in the Hardy space, whereas such convergence generally fails in weighted Dirichlet-type spaces, especially in the higher-order setting. In this paper, we investigate summability problems in higher-order weighted Dirichlet spaces $\widehat{\mathcal{H}}_{\mu,m}$ and show tha

Nestor O. Perez-Arancibia

The term embodied intelligence (EI) conveys the notion that body morphology, material properties, interaction with the environment, and control strategies can be purposefully integrated into the process of robotic design to generate intelligent behavior; in particular, locomotion and navigation. In this paper, we discuss EI as a design principle for advanced

Ali Caglayan, Nevrez Imamoglu, Oguzhan Guclu, Ali Osman Serhatoglu

Attention models have recently emerged as a powerful approach, demonstrating significant progress in various fields. Visualization techniques, such as class activation mapping, provide visual insights into the reasoning of convolutional neural networks (CNNs). Using network gradients, it is possible to identify regions where the network pays attention during

Musfiqur Rahman, SayedHassan Khatoonabadi, Emad Shihab

Large language models (LLMs) have demonstrated strong performance on function-level code generation benchmarks, yet real-world software development increasingly demands class-level implementations that integrate multiple methods, attributes, and dependencies within authentic project contexts. This gap between benchmark performance and practical utility raise

Junya Ogiri, Hiroaki Minamide, Kunio Ishida

We present a theoretical framework for quantum-coherent nonlinear interferometry in which the nonlinear medium is modeled as active electron-phonon quantum systems rather than a passive $\chi^{(2)}$ converter. By explicitly retaining the quantum coherence of the coupled electron-phonon-photon dynamics, our model describes a two-stage buildup of entanglement

Klaus Ziegler

We investigate a two-dimensional superconducting system with a smoothly and periodically varying order parameter. The order parameter is modulated along one direction while remaining uniform in the perpendicular direction, leading to a spatially periodic superconducting phase. We show that the periodicity of the order parameter determines the winding number

Duncan McCoy, Connor Sell

Although every flat manifold occurs as a cusp cross-section in at least one commensurability class of arithmetic hyperbolic manifolds, it turns out that some flat manifolds have the property that they occur as cusp cross-sections in precisely one commensurability class of arithmetic hyperbolic manifolds -- a phenomena which we will refer to as the UCC proper

Runsheng Xu, Hubert Lin, Wonseok Jeon, Hao Feng

Vision-based end-to-end (E2E) driving has garnered significant interest in the research community due to its scalability and synergy with multimodal large language models (MLLMs). However, current E2E driving benchmarks primarily feature nominal scenarios, failing to adequately test the true potential of these systems. Furthermore, existing open-loop evaluat

Nawar Turk, Sevag Kaspar, Leila Kosseim

This paper investigates the relationship between Persuasion Techniques (PTs) and Discourse Relations (DRs) by leveraging Large Language Models (LLMs) and prompt engineering. Since no dataset annotated with both PTs and DRs exists, we took the SemEval 2023 Task 3 dataset labelled with 19 PTs as a starting point and developed LLM-based classifiers to label eac

Jacopo Borga, Ewain Gwynne

We study longest and shortest directed paths in the following natural model of directed random planar maps: the uniform infinite bipolar-oriented triangulation (UIBOT), which is the local limit of uniform bipolar-oriented triangulations around a typical edge. We construct the Busemann function which measures directed distance to $\infty$ along a natural inte

Feng Ju, Zeyu Qin, Rui Min, Zhitao He

While Test-Time Scaling (TTS) has proven effective in improving the reasoning ability of large language models (LLMs), low diversity in model outputs often becomes a bottleneck; this is partly caused by the common "one problem, one solution" (1P1S) training practice, which provides a single canonical answer and can push models toward a narrow set of reasonin

Mara Daniels, Liam Hodgkinson, Michael Mahoney

Physics-informed machine learning (PIML) integrates prior physical information, often in the form of differential equation constraints, into the process of fitting machine learning models to physical data. Popular PIML approaches, including neural operators, physics-informed neural networks, neural ordinary differential equations, and neural discrete equilib

Yashaswini, Sanjay Ghosh

Advances in data analysis and machine learning have revolutionized the study of brain signatures using fMRI, enabling non-invasive exploration of cognition and behavior through individual neural patterns. Functional connectivity (FC), which quantifies statistical relationships between brain regions, has emerged as a key metric for studying individual variabi

Yasuhiko Asao, Kiyonori Gomi

For an $n\times n$ positive definite symmetric matrix $Z$ with $Z_{ii} = 1$ for all $i$, we show that there exists a set of vectors $V_Z\subset \mathbb{R}^n$ such that the radius $R$ of the circumsphere of $V_Z$ satisfies ${\rm Mag}\ Z = (1-R^2)^{-1}$. This leads us to interpret geometrically several known and new facts on magnitude. In particular, we show t

Xianben Yang, Yuxuan Li, Tao Wang, Tao Wang

Traditional novel view synthesis methods heavily rely on external camera pose estimation tools such as COLMAP, which often introduce computational bottlenecks and propagate errors. To address these challenges, we propose a unified framework that jointly optimizes 3D Gaussian points and camera poses without requiring pre-calibrated inputs. Our approach iterat

Shanyan Chen, Ali Al-Bayaty, Xiaoyu Song, Marek Perkowski

An $(n+1)$-bit Toffoli gate is mainly utilized to construct other quantum gates and operators, such as Fredkin gates, arithmetical adders, and logical comparators, where $n \geq 2$. Several researchers introduced different methods to decompose $(n+1)$-bit Toffoli gates in a quantum circuit into a set of standard 3-bit Toffoli gates or a set of elementary qua

Louis Wai-Tong Fan, Maximillian Newman, John Wakeley

We introduce a general diploid population model with self-fertilization and possible overlapping generations, and study the genealogy of a sample of $n$ genes as the population size $N$ tends to infinity. Unlike traditional approach in coalescent theory which considers the unconditional (annealed) law of the gene genealogies averaged over the population pedi

Caoshuo Li, Zengmao Ding, Xiaobin Hu, Bang Li

As one of the earliest writing systems, Oracle Bone Script (OBS) preserves the cultural and intellectual heritage of ancient civilizations. However, current OBS research faces two major challenges: (1) the interpretation of OBS involves a complex workflow comprising multiple serial and parallel sub-tasks, and (2) the efficiency of OBS information organizatio

Zhen Cao, F. Aharonian, Y. X. Bai, Y. W. Bao

Supernova remnants (SNRs) have been considered as the primary contributors to cosmic rays (CRs) in our Galaxy. However, the maximum energy of particles that can be accelerated by shocks of SNRs is uncertain, and SNRs' contribution to CRs around PeV energies is unclear. In this study, we present observations of high-energy $\gamma$-ray emission from the SNR I

Cheng-Liang Jiao, Liying Zhu, Er-gang Zhao, Jia Zhang

Stellar-mass compact objects (COs) embedded in active galactic nucleus (AGN) discs are commonly assumed to accrete via Bondi or Bondi-Hoyle-Lyttleton (BHL) prescriptions, neglecting gas angular momentum. We show that differential rotation in AGN discs can impart non-negligible angular momentum, in which case accretion proceeds through a viscous disc rather t

Xiaoqian Qi, Haoye Chai, Sichang Liu, Lei Yue

The planning, management, and resource scheduling of cellular mobile networks require joint estimation of mobile traffic across different layers and nodes. Mobile traffic generation can proactively anticipate user demands and capture the dynamics of network load. However, existing methods mainly focus on generating traffic at a single spatiotemporal resoluti

Sándor Kisfaludi-Bak, Tze-Yang Poon, Geert van Wordragen

Hyperbolic tilings are natural infinite planar graphs where each vertex has degree $q$ and each face has $p$ edges for some $\frac1p+\frac1q<\frac12$. We study the structure of shortest paths in such graphs. We show that given a set of $n$ terminals, we can compute a so-called isometric closure (closely related to the geodesic convex hull) of the terminals i

Chenming Tang, Hsiu-Yuan Huang, Weijie Liu, Clive Bai

Reinforcement learning with verifiable rewards (RLVR) has significantly boosted the reasoning capability of language models (LMs). However, existing RLVR approaches train LMs based on their own on-policy responses and are constrained by the initial capability of LMs, thus prone to exploration stagnation, in which LMs fail to solve more training problems and

Aldo Garcia Guinto, Yuki Miyamoto

Given a locally compact second countable group $G$ with a 2-cocycle $\omega$, we show that the restriction of the twisted Plancherel weight $\varphi^\omega_G$ to the subalgebra generated by a closed subgroup $H$ in the twisted group von Neumann algebra $L_\omega(G)$ is semifinite if and only if $H$ is open. When $G$ is almost unimodular, i.e. $\ker\Delta_G$

Amal Mattoo

Johannes Krah showed that the blowup of $\mathbf{P}^{2}$ in $10$ general points admits a phantom subcategory. We construct three types of objects in such a phantom: a strong generator, projections of skyscraper sheaves, and a family of objects with two nonzero cohomology sheaves. We study the deformation theory of these objects to show that the phantom conta

Ratzanyel Rincón, Kyungchul Song

This paper examines methods of causal inference based on groupwise matching when we observe multiple large groups of individuals over several periods. We formulate causal inference validity through a generalized matching condition, generalizing the parallel trend assumption in difference-in-differences designs. We show that difference-in-differences, synthet

Xiaosen Wang, Zhijin Ge, Shaokang Wang

Despite the notable advancements and versatility of multi-modal diffusion models, such as text-to-image models, their susceptibility to adversarial inputs remains underexplored. Contrary to expectations, our investigations reveal that the alignment between textual and Image modalities in existing diffusion models is inadequate. This misalignment presents sig

Zhaoqi Zhang, Haolei Pei, Jun Guo, Tianyu Wang

In recommendation systems, scaling up feature-interaction modules (e.g., Wukong, RankMixer) or user-behavior sequence modules (e.g., LONGER) has achieved notable success. However, these efforts typically proceed on separate tracks, which not only hinders bidirectional information exchange but also prevents unified optimization and scaling. In this paper, we

Maximilian Schreiber, Pascal Tippe

This paper presents a comprehensive empirical analysis of security vulnerabilities in AI-generated code across public GitHub repositories. We collected and analyzed 7,703 files explicitly attributed to four major AI tools: ChatGPT (91.52\%), GitHub Copilot (7.50\%), Amazon CodeWhisperer (0.52\%), and Tabnine (0.46\%). Using CodeQL static analysis, we identif

Lisha Shuai, Jiuling Dong, Nan Zhang, Shaofeng Tan

Local Differential Privacy (LDP) is a widely adopted privacy-protection model in the Internet of Things (IoT) due to its lightweight, decentralized, and scalable nature. However, it is vulnerable to poisoning attacks, and existing defenses either incur prohibitive resource overheads or rely on domain-specific prior knowledge, limiting their practical deploym

Taku Mikuriya, Tatsuya Ishigaki, Masayuki Kawarada, Shunya Minami

Large language models (LLMs) have increasingly been applied to automatic programming code generation. This task can be viewed as a language generation task that bridges natural language, human knowledge, and programming logic. However, it remains underexplored in domains that require interaction with hardware devices, such as quantum programming, where human

Hongtao Guo Shuai Li Shu Li

This paper systematically reviews the research progress and application prospects of machine learning technologies in the field of polymer materials. Currently, machine learning methods are developing rapidly in polymer material research; although they have significantly accelerated material prediction and design, their complexity has also caused difficultie

Nick Masi, Randall Balestriero

The dominant paradigm in machine learning is to assess model performance based on average loss across all samples in some test set. This amounts to averaging performance geospatially across the Earth in weather and climate settings, failing to account for the non-uniform distribution of human development and geography. We introduce Stratified Assessments of

Chenrui Shi, Zedong Yu, Zhi Gao, Ruining Feng

Vision language models (VLMs) have advanced graphical user interface (GUI) task automation but still lag behind humans. We hypothesize this gap stems from missing core GUI knowledge, which existing training schemes (such as supervised fine tuning and reinforcement learning) alone cannot fully address. By analyzing common failure patterns in GUI task executio

Nan Lin, Linrui Zhang, Yuxuan Chen, Zhenrui Chen

For the task with complicated manipulation in unstructured environments, traditional hand-coded methods are ineffective, while reinforcement learning can provide more general and useful policy. Although the reinforcement learning is able to obtain impressive results, its stability and reliability is hard to guarantee, which would cause the potential safety t

Usman Akram, Fan Zhang, Yang Li, Haris Vikalo

Accurate channel state information (CSI) is essential for reliable multiuser MIMO operation. In 5G NR, reciprocity-based beamforming via uplink Sounding Reference Signals (SRS) face resource and coverage constraints, motivating sparse non-uniform SRS allocation. Prior masked-autoencoder (MAE) approaches improve coverage but overfit to training masks and degr

Weifei Jin, Yuxin Cao, Junjie Su, Minhui Xue

Recent advances in Audio-Language Models (ALMs) have significantly improved multimodal understanding capabilities. However, the introduction of the audio modality also brings new and unique vulnerability vectors. Previous studies have proposed jailbreak attacks that specifically target ALMs, revealing that defenses directly transferred from traditional audio

Jingyuan He, Jiongnan Liu, Vishan Vishesh Oberoi, Bolin Wu

Recommender systems are among the most impactful AI applications, interacting with billions of users every day, guiding them to relevant products, services, or information tailored to their preferences. However, the research and development of recommender systems are hindered by existing datasets that fail to capture realistic user behaviors and inconsistent

Yuxin Li, Minghao Liu, Ruida Wang, Wenzhao Ji

We present **Lean4PHYS**, a comprehensive reasoning framework for college-level physics problems in Lean4. **Lean4PHYS** includes *LeanPhysBench*, a college-level benchmark for formal physics reasoning in Lean4, which contains 200 hand-crafted and peer-reviewed statements derived from university textbooks and physics competition problems. To establish a soli

Qianchao Wang, Chuanzhen Jia, Yuxuan Ding, Zhe Li

Lightweight online detection of series arc faults is critically needed in residential and industrial power systems to prevent electrical fires. Existing diagnostic methods struggle to achieve both rapid response and robust accuracy under resource-constrained conditions. To overcome the challenge, this work suggests leveraging a multi-frequency neural network