October 2025 arXiv papers — page 33

Nikola Bogdanovic, Laura Cossu, Azeem Khadam

This paper investigates atomic factorizations in the monoid $\mathcal I(R)$ of nonzero ideals of a multivariate polynomial ring $R$, under ideal multiplication. Building on recent advances in factorization theory for unit-cancellative monoids, we extend techniques from the paper [Geroldinger and Khadam, Ark. Mat. 60 (2022), 67-106] to construct new families

Ekaterina Amerik, Andrey Soldatenkov, Misha Verbitsky

Wierzba and Wisniewski proved that in dimension 4, every bimeromorphic map of hyperkahler manifolds is represented as a composition of Mukai flops. Hu and Yau conjectured that this result can be generalized to arbitrary dimension. They defined ``Mukai's elementary transformation'' as the blow-up of a subvariety ruled by complex projective spaces, composed wi

Carolin Drenda, Dennis Dobler, Merle Munko, Andrew Titman

Various estimators for modelling the transition probabilities in multi-state models have been proposed, e.g., the Aalen-Johansen estimator, the landmark Aalen-Johansen estimator, and a hybrid Aalen-Johansen estimator. While the Aalen-Johansen estimator is generally only consistent under the rather restrictive Markov assumption, the landmark Aalen-Johansen es

Xi Cheng, Weijie Shen, Haoming Chen, Chaoyi Shen

Time series forecasting and anomaly detection are common tasks for practitioners in industries such as retail, manufacturing, advertising and energy. Two unique challenges stand out: (1) efficiently and accurately forecasting time series or detecting anomalies in large volumes automatically; and (2) ensuring interpretability of results to effectively incorpo

Il-Seung Jang, Jae-Hoon Kwon, Akito Uruno

We construct a crystal base of the negative half of a quantum orthosymplectic superalgebra. It can be viewed as a limit of the crystal bases of $q$-deformed irreducible oscillator representations. We also give a combinatorial description of the embedding from the crystal of a $q$-oscillator representation to that of the negative half subalgebra given in term

Eunju Kwon, Seungwon Oh, In-Chang Baek, Yucheon Park

Contact-rich manipulation has become increasingly important in robot learning. However, previous studies on robot learning datasets have focused on rigid objects and underrepresented the diversity of pressure conditions for real-world manipulation. To address this gap, we present a humanoid visual-tactile-action dataset designed for manipulating deformable s

Simon A. Lee, Cyrus Tanade, Hao Zhou, Juhyeon Lee

Wearable sensors provide abundant physiological time series, yet the principles governing their predictive utility remain unclear. We hypothesize that temporal resolution is a fundamental axis of representation learning, with different clinical and behavioral outcomes relying on structure at distinct scales. To test this resolution hypothesis, we introduce H

Špela Vintar, Taja Kuzman Pungeršek, Mojca Brglez, Nikola Ljubešić

While new benchmarks for large language models (LLMs) are being developed continuously to catch up with the growing capabilities of new models and AI in general, using and evaluating LLMs in non-English languages remains a little-charted landscape. We give a concise overview of recent developments in LLM benchmarking, and then propose a new taxonomy for the

A. Boasso, F. D. Mazzitelli

We analyze the renormalized stress-energy tensor (RSET) of a massless quantum scalar field in time-dependent gravitational backgrounds. Starting from its formal expression obtained within the covariant perturbative expansion to lowest order in the curvature, we evaluate the RSET in an arbitrary number of dimensions in terms of coordinate-space distributions.

Pablo Acuaviva, Aram Davtyan, Mariam Hassan, Sebastian Stapf

Large language models (LLMs) have demonstrated that large-scale pretraining enables systems to adapt rapidly to new problems with little supervision in the language domain. This success, however, has not translated as effectively to the visual domain, where models, including LLMs, continue to struggle with compositional understanding, sample efficiency, and

Angelika Abramiuk-Szurlej, Katarzyna Sznajd-Weron

Collective adaptation, whether in innovation adoption, pro-environmental or organizational change, emerges from the interplay between individual decisions and social influence. Agent-based modeling provides a useful tool for studying such processes. Here, we introduce the biased-independence $q$-voter model, a generalization of the $q$-voter model with indep

Viktoriia Zinkovich, Anton Antonov, Andrei Spiridonov, Denis Shepelev

Multimodal large language models (MLLMs) have shown impressive capabilities in vision-language tasks such as reasoning segmentation, where models generate segmentation masks based on textual queries. While prior work has primarily focused on perturbing image inputs, semantically equivalent textual paraphrases-crucial in real-world applications where users ex

Hitoshi Makino, Bernd Rellinghaus, Sebastian Schneider, Axel Lubk

Quantitative electron magnetic circular dichroism (EMCD) in transmission electron microscopy (TEM) enables the measurement of magnetic moments with elemental and atomic site sensitivity, but its practical application is fundamentally limited by noise. This study presents a comprehensive methodology for noise estimation and suppression in EMCD measurements, d

Taozhi Yang, Zhaoyu Zuo

Asteroseismic modelling of high-amplitude $\delta$ Scuti (HADS) variables critically depends on their accurate classification, which provides robust constraints on stellar physical parameters. As a foundational step in this direction, we present a detailed analysis of the pulsational behavior of 50 HADS star candidates using high-precision photometric data f

Tom Ó Nualláin

This project introduces the GNAR-HARX model, which combines Generalised Network Autoregressive (GNAR) structure with Heterogeneous Autoregressive (HAR) dynamics and exogenous predictors such as implied volatility. The model is designed for forecasting realised volatility by capturing both temporal persistence and cross-sectional spillovers in financial marke

Yiding Wang, Yuxuan Chen, Fanxu Meng, Xifan Chen

Since real-world legal experiments are often costly or infeasible, simulating legal societies with Artificial Intelligence (AI) systems provides an effective alternative for verifying and developing legal theory, as well as supporting legal administration. Large Language Models (LLMs), with their world knowledge and role-playing capabilities, are strong cand

Mária Lukáčová-Medvid'ová, Ferdinand Thein, Gerald Warnecke, Yuhuan Yuan

The paper is concerned with proving the equivalence of convexity or concavity properties of thermodynamic functions, such as energy and entropy, depending on different sets of variables. These variables are the basic thermodynamic state variables, specific state variables or the densities of state variables that are used in continuum mechanics. We prove resu

Abdullah Mushtaq, Rafay Naeem, Ezieddin Elmahjub, Ibrahim Ghaznavi

Large language models are increasingly used for Islamic guidance, but risk misquoting texts, misapplying jurisprudence, or producing culturally inconsistent responses. We pilot an evaluation of GPT-4o, Ansari AI, and Fanar on prompts from authentic Islamic blogs. Our dual-agent framework uses a quantitative agent for citation verification and six-dimensional

Zhineng Zhao, Zhihai He, Zikun Zhou, Siwei Ma

Learned image compression (LIC) has shown great promise for achieving high rate-distortion performance. However, current LIC methods are often limited in their capability to model the complex correlation structures inherent in natural images, particularly the entanglement of invariant global structures with transient local textures within a single monolithic

Pedro Freitas, Roméo Leylekian

We present examples of bounded planar domains with one single hole for which the nodal line of a second Dirichlet eigenfunction is closed and does not touch the boundary. This shows that Payne's nodal line conjecture can at most hold for simply-connected domains in the plane.

Masahiro Kato

This study proves that Nearest Neighbor (NN) matching can be interpreted as an instance of Riesz regression for automatic debiased machine learning. Lin et al. (2023) shows that NN matching is an instance of density-ratio estimation with their new density-ratio estimator. Chernozhukov et al. (2024) develops Riesz regression for automatic debiased machine lea

Seyed Mahdi Basiri Azad, Joschka Boedecker

Reinforcement learning (RL) in sparse-reward environments remains a significant challenge due to the lack of informative feedback. We propose a simple yet effective method that uses a small number of successful demonstrations to initialize the value function of an RL agent. By precomputing value estimates from offline demonstrations and using them as targets

Xiaoyu Kong, Leheng Sheng, Junfei Tan, Yuxin Chen

The recent success of large language models (LLMs) has renewed interest in whether recommender systems can achieve similar scaling benefits. Conventional recommenders, dominated by massive embedding tables, tend to plateau as embedding dimensions grow. In contrast, the emerging generative paradigm replaces embeddings with compact Semantic ID (SID) sequences

Yejin Kim, Shaghayegh Agah, Mayur Nankani, Neeraj Sharma

Most recommender systems treat timestamps as numeric or cyclical values, overlooking real-world context such as holidays, events, and seasonal patterns. We propose a scalable framework that uses large language models (LLMs) to generate geo-temporal embeddings from only a timestamp and coarse location, capturing holidays, seasonal trends, and local/global eve

Edward Rothberg

First-order methods based on the PDHG algorithm have recently emerged as a viable option for efficiently solving large-scale linear programming problems. One highly desirable property of these methods is that they can make effective use of GPUs. One undesirable property is that, as first-order methods, their convergence can be extremely slow. This property f

Anh Nguyen Hoang, Minh Le-Anh, Bach Le, Nghi D. Q. Bui

Given a large and evolving codebase, the ability to automatically generate holistic, architecture-aware documentation that captures not only individual functions but also cross-file, cross-module, and system-level interactions remains an open challenge. Comprehensive documentation is essential for long-term software maintenance and collaboration, yet current

Ken Gu, Advait Bhat, Mike A Merrill, Robert West

Evaluating the reasoning ability of language models (LMs) is complicated by their extensive parametric world knowledge, where benchmark performance often reflects factual recall rather than genuine reasoning. Existing datasets and approaches (e.g., temporal filtering, paraphrasing, adversarial substitution) cannot cleanly separate the two. We present SynthWo

Elena Caviglia, Luca Mesiti, Cerene Rathilal

In this paper, we generalize the concept of unicoherence to the context of frames. Unicoherence, originally introduced by Kuratowski, is a connectedness property that is well studied in classical topology and used to detect holes of a space. We extend the notion of unicoherence to locales and we then investigate its properties. In particular, we prove that m

Guangyu Xie, Yice Zhang, Jianzhu Bao, Qianlong Wang

Recent efforts leverage knowledge distillation techniques to develop lightweight and practical sentiment analysis models. These methods are grounded in human-written instructions and large-scale user texts. Despite the promising results, two key challenges remain: (1) manually written instructions are limited in diversity and quantity, making them insufficie

Louis-Pierre Arguin, Jad Hamdan

We continue the study of the Fourier coefficients of Gaussian multiplicative chaos (GMC) recently initiated by Garban and Vargas. We show that if $\{c_n\}_{n\geq 1}$ are the Fourier coefficients of critical GMC on the unit interval, then $(\log n)^{\alpha}c_n$ converges to zero in probability as $n$ tends to infinity for any $\alpha<1/4$.

E. Minguzzi

We provide a short introduction to ``Lorentzian metric spaces" i.e., spacetimes defined solely in terms of the two-point Lorentzian distance. As noted in previous work, this structure is essentially unique if minimal conditions are imposed, such as the continuity of the Lorentzian distance and the relative compactness of chronological diamonds. The latter co

Bijeet Basak, Nupur Patil, Kurian Polachan, Srinivas Vivek

Binarized Neural Networks (BNNs) deployed on memristive crossbar arrays provide energy-efficient solutions for edge computing but are susceptible to physical attacks due to memristor nonvolatility. Recently, Rajendran et al. (IEEE Embedded Systems Letter 2025) proposed a Physical Unclonable Function (PUF)-based scheme to secure BNNs against theft attacks. Sp

Miguel S. Oliveira, Cristiano Ciuti

We theoretically predict the interaction between polaritonic excitations arising from the coupling of a cavity photon mode with bound to continuum intersubband transitions in a doped quantum well. The resulting exciton bound by photon exchange, recently demonstrated experimentally, exhibits a binding energy that can be continuously tuned by varying the cavit

Federico Formica, Stefano Gregis, Aurora Francesca Zanenga, Andrea Rota

Understanding why neural networks make certain decisions is pivotal for their use in safety-critical applications. Feature-Guided Analysis (FGA) extracts slices of neural networks relevant to their tasks. Existing feature-guided approaches typically monitor the activation of the neural network neurons to extract the relevant rules. Preliminary results are en

Éric Fusy

Triangulations of the 5-gon with no separating triangle nor quadrangle, so called 5c-triangulations, are a planar map family closely related to 5-connected planar triangulations. We show that 5c-triangulations are in bijection with 5-regular plane trees satisfying a simple local constraint at inner edges. It yields explicit expressions for the generating fun

Edoardo Martinenghi, Valerio Toso, Fabrizio Bruno Armani, Andrea Castoldi

The DAMA experiment's long-standing claim of dark matter detection remains a key open issue in astroparticle physics. Independent verification requires NaI(Tl)-based detectors with enhanced low-energy sensitivity. Current detectors rely on photomultiplier tubes (PMTs) which features limited detection efficiency, intrinsic radioactivity, and high noise at keV

Margaret Beck, Jonathan Jaquette, Hannah Pieper

In this paper, we develop new theory connected with resonant vector bundles that will allow for the use of validated numerics to rigorously determine the stability of pulse solutions in the context of the Swift-Hohenberg equation. For many PDEs, the stability of stationary solutions is determined by the absence of point spectra in the open right half of the

Nikhat Khan, E. M. H. E. B. Ekanayake, Nicolas Casilli, Cristian Cassella

Networks of coupled nonlinear oscillators are emerging as powerful physical platforms for implementing Ising machines. Yet the relationship between parametric-oscillator implementations and traditional oscillator-based Ising machines remains underexplored. In this work, we develop a Kuramoto-style, canonical phase description of parametric oscillator Ising m

Reem Hammoud, Abdul Karim Gizzini, Ali J. Ghandour

Ensuring transparency and trust in artificial intelligence (AI) models is essential as they are increasingly deployed in safety-critical and high-stakes domains. Explainable AI (XAI) has emerged as a promising approach to address this challenge; however, the rigorous evaluation of XAI methods remains vital for balancing the trade-offs between model complexit

Aaron R. Allred, Erin E. Richardson, Sarah R. Bostrom, James Crum

Technological systems increasingly mediate human information exchange, spanning interactions among humans as well as between humans and artificial agents. The unprecedented scale and reliance on information disseminated through these systems substantially expand the scope of information-based influence that can both enable and undermine sound decision-making

Ali Ahmad Faour, Nabil Amacha, Ali J. Ghandour

The sustainable management of the Qaraaoun Reservoir, the largest surface water body in Lebanon located in the Bekaa Plain, depends on reliable monitoring of its storage volume despite frequent sensor malfunctions and limited maintenance capacity. This study introduces a sensor-free approach that integrates open-source satellite imagery, advanced water-exten

Joseph West

The motion of each particle in an N body system of identical masses interacting via an attractive or repulsive pairwise linear force law, the "Swarm," and with an external attractive or repulsive linear force law, the "Trap," is considered. In all Swarm and Trap combinations the motion of all N particles is completely separable and positions are found as a f

Yanzhen Cai, Mingtai Xie, Jing Kang, Weizhen Zhuo

Tuning magnetic anisotropy through chemical doping is a powerful strategy for designing functional materials with enhanced magnetic properties. Here, we report an enhanced Er^3+ magnetic moment resulting from nonmagnetic Lu^3+ substitution in the honeycomb-lattice antiferromagnet ErOCl. Unlike the Curie-Weiss type divergence typically observed in diluted mag

Yifan Wu, Xuewei Feng, Yuxiang Yang, Ke Xu

As the core of the Internet infrastructure, the TCP/IP protocol stack undertakes the task of network data transmission. However, due to the complexity of the protocol and the uncertainty of cross-layer interaction, there are often inconsistencies between the implementation of the protocol stack code and the RFC standard. This inconsistency may not only lead

Christian Klein, Svetlana Roudenko, Nikola Stoilov

We consider the nonlinear Schr\"odinger equation on a unit ball in one and two dimensions with Dirichlet boundary conditions, which have stabilizing effect on solutions behavior. In particular, we confirm that the ground state solutions are stable in subcritical and critical cases, and in the supercritical case the ground state solutions split into a stable

Marcelo R. Barbosa, Horatiu Nastase, Lucas S. Sousa

In this paper we analyze a generalized "single-trace $T\bar T$" deformation, defined by a TsT transformation, of the fibered $I$-brane solution from \cite{Nunez2023}. We use the Penrose limit to understand it, and we consider both the TsT followed by the Penrose limit, as well as the Penrose limit followed by TsT. We describe the spin chains obtained in fiel

Shantonu Mukherjee

An interaction between ferromagnetic and superconducting orders, to be realized in a two dimensional ferromagnetic superconductor, is proposed obeying necessary symmetry principles. This interaction allows us to formulate a duality, similar to the Boson-vortex duality in 2+1 dimensional superfluid. In the dual theory the Skyrmion and the vortex excitations i

Michail Dadopoulos, Anestis Ladas, Stratos Moschidis, Ioannis Negkakis

Retrieval-Augmented Generation (RAG) struggles on long, structured financial filings where relevant evidence is sparse and cross-referenced. This paper presents a systematic investigation of advanced metadata-driven Retrieval-Augmented Generation (RAG) techniques, proposing and evaluating a novel, multi-stage RAG architecture that leverages LLM-generated met

Vaibhav Katyal, A. Chakraborty, B. K. Sahoo

By critically evaluating higher-order nonlinear effects to the isotope shifts (ISs) in the low-lying transition frequencies of the singly charged calcium ion, stringent constraint on the electron-neutron coupling due to a hypothetical boson describing physics beyond the Standard Model is inferred. It shows an order magnitude difference compared to the previo

Francisco Díaz-Ruiz, Francisco J. Martín-Vega, José Antonio Cortés, Gerardo Gómez

Time division duplexing (TDD) has become the dominant duplexing mode in 5G and beyond due to its ability to exploit channel reciprocity for efficient downlink channel state information (CSI) acquisition. However, channel aging caused by user mobility and processing delays degrades the accuracy of CSI, leading to suboptimal link adaptation and loss of perform

Youwan Mahé, Elise Bannier, Stéphanie Leplaideur, Elisa Fromont

Post-stroke MRI not only delineates focal lesions but also reveals secondary structural changes, such as atrophy and ventricular enlargement. These abnormalities, increasingly recognised as imaging biomarkers of recovery and outcome, remain poorly captured by supervised segmentation methods. We evaluate REFLECT, a flow-based generative model, for unsupervise

Jiarui Qin, Yunjia Xi, Junjie Huang, Renting Rui

With the rapid development of LLM-based agents, there is a growing trend to incorporate agent-specific data into the pre-training stage of LLMs, aiming to better align LLMs with real-world autonomous task execution. However, current pre-training benchmarks primarily focus on isolated and static skills, e.g., common knowledge or mathematical/code reasoning, a

Dhananjaya Gowda, Seoha Song, Harshith Goka, Junhyun Lee

Large language models (LLMs) are increasingly deployed with task-specific adapters catering to multiple downstream applications. In such a scenario, the additional compute associated with these apparently insignificant number of adapter parameters (typically less than 1% of the base model) turns out to be disproportionately significant during inference time

Shu-Peng Wang, Zhen-Yan Lu, Zhi-Jun Ma, Rong-Yao Yang

A recent study shows that incorporating a new term into the thermodynamic potential density, as required by the thermodynamic consistency criterion, can effectively resolve the thermodynamic inconsistency problems of the conventional perturbative QCD model. This additional term plays a crucial role in resolving inconsistencies at relatively low densities and

Iwan Morton-Blake, Fuyudi Zhang, Qichao Chang, Shuhua Hao

High-energy neutrino astronomy has advanced rapidly in recent years, with IceCube, KM3NeT, and Baikal-GVD establishing a diffuse astrophysical flux and pointing to promising source candidates. These achievements mark the transition from first detections to detailed source studies, motivating next-generation detectors with larger volumes, improved angular res

Sandra Barragán, Adrián Pérez-Bote, Carlos Sáez, David Salgado

We provide a description of pilot and production experiences to streamline some business functions in the official statistical production process using statistical learning models. Our approach is quality-oriented searching for an improvement on accuracy, cost-efficiency, timeliness, granularity, response burden reduction, and frequency. Pilot experiences ha

Yan Meng, Jiachun Li, Matthew Pillari, Arjun Deopujari

Though playing an essential role in smart home systems, smart speakers are vulnerable to voice spoofing attacks. Passive liveness detection, which utilizes only the collected audio rather than the deployed sensors to distinguish between live-human and replayed voices, has drawn increasing attention. However, it faces the challenge of performance degradation

Matteo Giordano

I discuss the consequences of the constraints imposed on the Dirac spectrum by the restoration of chiral symmetry in the chiral limit of gauge theories with two light fermion flavors, with particular attention to the fate of the anomalous $\mathrm{U}(1)_A$ symmetry. Under general, physically motivated assumptions on the spectral density and on the two-point

Alvaro Detailleur, Dalim Wahby, Guillaume Ducard, Christopher Onder

Neural-network-based controllers (NNCs) can represent complex, highly nonlinear control laws, but verifying the closed-loop stability of dynamical systems using them remains challenging. This work presents contributions to a state-of-the-art stability verification procedure for NNC-controlled systems which relies on semialgebraic-set-based input-output model

Xianjun Gao, Jianchun Liu, Hongli Xu, Liusheng Huang

The integration of Large Language Models (LLMs) into real-time Web applications, such as AI-powered search and conversational agents, presents a fundamental Web infrastructure challenge: reconciling the demand for high-quality, complex reasoning with the stringent low-latency and high-throughput requirements of interactive services. Current LLM reasoning, hi

Lamine Chalal, Ahmed Rachid

This paper presents the development and validation of a digital twin for a scaled-down electric vehicle (EV) emulator, designed to replicate longitudinal vehicle dynamics under diverse operating conditions. The emulator integrates a separately excited DC motor (SEDCM), a four-quadrant DC-DC converter, a battery emulator, and a mechanical load emulator. The s

Andrew Beveridge, Ari Holcombe Pomerance

We consider random walks on a tree $G=(V,E)$ with stationary distribution $\pi_v = \mathrm{deg}(v)/2|E|$ for $v \in V$. Let the hitting time $H(v,w)$ denote the expected number of steps required for the random walk started at vertex $v$ to reach vertex $w$. We characterize the extremal tree structures for the best meeting time $T_{\mathrm{bestmeet}}(G) = \mi

Robert G. Radloff, Felix F. Zimmermann, Siqi Li, Stephan Kuschel

Lensless X-ray imaging provides element-specific nanoscale insights into thick samples beyond the reach of conventional light and electron microscopy. Coherent diffraction imaging (CDI) methods, such as ptychographic tomography, can recover three-dimensional (3D) nanoscale structures but require extensive sample rotation, adding complexity to experiments. X-

Herman Bergström, Zhongqi Yue, Fredrik D. Johansson

Medical images used to train machine learning models are often accompanied by radiology reports containing rich expert annotations. However, relying on these reports as inputs for clinical prediction requires the timely manual work of a trained radiologist. This raises a natural question: when can radiology reports be leveraged during training to improve ima

Chaopeng Tan, Jiarong Yao, Meng Wang

Reliable estimation of macro and micro traffic states is essential for urban traffic management. Unmanned Aerial Vehicles, with their airborne full-sample continuous trajectory observation, bring new opportunities for macro- and micro-traffic state estimation. In this study, we will explore the optimal UAV deployment problem in road networks in conjunction w

Juraj Mavračić

Policy Cards are introduced as a machine-readable, deployment-layer standard for expressing operational, regulatory, and ethical constraints for AI agents. The Policy Card sits with the agent and enables it to follow required constraints at runtime. It tells the agent what it must and must not do. As such, it becomes an integral part of the deployed agent. P

Haoyang Liu, Cuihua Du, Zhongcheng Li, Jian Zhang

We selected Bulge stars from APOGEE DR17 cross-matched with astrometric data from \textit{Gaia} DR3. Bulge stars were divided into sub-samples with line-of-sight velocity dispersion analyzed and the peaks of MDF were detected by both Gaussian Mixture Models (GMM) and \texttt{scipy.signal.find\_peaks}. GMM is also conducted to kinematically distinguish the me

Matteo Giordano

I study the consequences of chiral symmetry restoration for the Dirac spectrum in finite-temperature gauge theories in the two-flavor chiral limit, using Ginsparg--Wilson fermions on the lattice. I prove that chiral symmetry is restored at the level of the susceptibilities of scalar and pseudoscalar bilinears if and only if all these susceptibilities do not

Zhuangfan Huang, Xiaosong Li, Gao Wang, Tao Ye

Polarization image fusion combines S0 and DOLP images to reveal surface roughness and material properties through complementary texture features, which has important applications in camouflage recognition, tissue pathology analysis, surface defect detection and other fields. To intergrate coL-Splementary information from different polarized images in complex

Yann Kerverdo, Florent Leray, Youwan Mahé, Stéphanie Leplaideur

Deep learning frameworks such as nnU-Net achieve state-of-the-art performance in brain lesion segmentation but remain difficult to deploy clinically due to heavy dependencies and monolithic design. We introduce \textit{StrokeSeg}, a modular and lightweight framework that translates research-grade stroke lesion segmentation models into deployable applications

Hedieh Abdollahi

We present half-light radius measurements for the dwarf satellites of Andromeda, based on multi-epoch imaging from the Isaac Newton Telescope (INT) Monitoring Survey of Local Group dwarf galaxies. This analysis is conducted within a larger study to identify long-period variable (LPV) stars in these galaxies. The survey was performed with the Wide Field Camer

Da Chang, Peng Xue, Yu Li, Yongxiang Liu

Parameter-Efficient Fine-Tuning (PEFT) methods are crucial for adapting large pre-trained models. Among these, LoRA is considered a foundational approach. Building on this, the influential DoRA method enhances performance by decomposing weight updates into magnitude and direction. However, its underlying mechanism remains unclear, and it introduces significa

Yuanyuan Wu, Zhenlin Qin, Zhenliang Ma

Synthetic data offers a promising solution to the privacy and accessibility challenges of using smart card data in public transport research. Despite rapid progress in generative modeling, there is limited attention to comprehensive evaluation, leaving unclear how reliable, safe, and useful synthetic data truly are. Existing evaluations remain fragmented, ty

Yuda Zou, Zijian Zhang, Yongchao Xu

Referring Expression Counting (REC) extends class-level object counting to the fine-grained subclass-level, aiming to enumerate objects matching a textual expression that specifies both the class and distinguishing attribute. A fundamental challenge, however, has been overlooked: annotation points are typically placed on class-representative locations (e.g.,

Victor Galaz, Maria Schewenius, Jonathan F. Donges, Ingo Fetzer

Artificial intelligence (AI) is already driving scientific breakthroughs in a variety of research fields, ranging from the life sciences to mathematics. This raises a critical question: can AI be applied both responsibly and effectively to address complex and interconnected sustainability challenges? This report is the result of a collaboration between the S

Ziyang Zhang, Yifan Gao, Xuenan Xu, Baoxiang Li

Text-to-Speech (TTS) is inherently a "one-to-many" mapping characterized by intrinsic uncertainty, yet current paradigms often oversimplify it into a deterministic regression task. While continuous-valued autoregressive (AR) models have recently emerged as a promising alternative to discrete codec-based approaches, they typically rely on a fixed-variance pri

Francisco S. N. Lobo, Tiberiu Harko, Miguel A. S. Pinto

Modified gravity theories with a nonminimal coupling between curvature and matter offer a compelling alternative to dark energy and dark matter by introducing an explicit interaction between matter and curvature invariants. Two of the main consequences of such an interaction are the emergence of an additional force and the non-conservation of the energy--mom

Yue Wu

Pneumatic drying processes in industries such as agriculture, chemicals,and pharmaceuticals are notoriously difficult to model and control due to multi-source disturbances,coupled stage dynamics, and significant measurement delays. Traditional modeling paradigms often fail to simultaneously deliver accuracy, interpretability, and closed-loop applicability. T

Yutian Xiao, Meng Yuan, Fuzhen Zhuang, Wei Chen

The pre-ranking stage plays a pivotal role in large-scale recommender systems but faces an intrinsic trade-off between model expressiveness and computational efficiency. Owing to the massive candidate pool and strict latency constraints, industry systems often rely on lightweight two-tower architectures, which are computationally efficient yet limited in est

Dhananjaya Gowda, Seoha Song, Junhyun Lee, Harshith Goka

As the large language models (LLMs) grow in size each day, efficient training and fine-tuning has never been as important as nowadays. This resulted in the great interest in parameter efficient fine-tuning (PEFT), and effective methods including low-rank adapters (LoRA) has emerged. Although the various PEFT methods have been studied extensively in the recen

Maria Gabriela Valeriano, David Kohan Marzagão, Alfredo Montelongo, Carlos Roberto Veiga Kiffer

Machine Learning (ML) models are widely used in high-stakes domains such as healthcare, where the reliability of predictions is critical. However, these models often fail to account for uncertainty, providing predictions even with low confidence. This work proposes a novel two-step data-centric approach to enhance the performance of ML models by improving da

Junda He, Jieke Shi, Terry Yue Zhuo, Christoph Treude

The rapid integration of Large Language Models (LLMs) into software engineering (SE) has revolutionized tasks like code generation, producing a massive volume of software artifacts. This surge has exposed a critical bottleneck: the lack of scalable, reliable methods to evaluate these outputs. Human evaluation is costly and time-consuming, while traditional a

Hoang-Thien Nguyen, Thanh-Huy Nguyen, Ba-Thinh Lam, Vi Vu

Teacher-student frameworks have emerged as a leading approach in semi-supervised medical image segmentation, demonstrating strong performance across various tasks. However, the learning effects are still limited by the strong correlation and unreliable knowledge transfer process between teacher and student networks. To overcome this limitation, we introduce

Matthew Shardlow

Sentence embeddings can be decoded to give approximations of the original texts used to create them. We explore this effect in the context of text simplification, demonstrating that reconstructed text embeddings preserve complexity levels. We experiment with a small feed forward neural network to effectively learn a transformation between sentence embeddings

Louis Jourdan, Patrick Cassam-Chenaï

We show that the Zassenhaus decomposition for the exponential of the sum of two non-commuting operators, simplifies drastically when these operators satisfy a simple condition, called the no-mixed adjoint property. An important application to a Unitary Coupled Cluster method for strongly correlated electron systems is presented. This ansatz requires no Trott

Maxim N. Chernodub, Vladimir A. Goy, Alexander V. Molochkov, Alexey S. Tanashkin

We discuss a one-quark state in the confinement phase near a reflective chromometallic boundary both at finite and zero temperature. Using numerical simulations of lattice Yang-Mills theory, we show that the test quark is confined to the neutral mirror by an attractive potential of the Cornell type, suggesting the existence of a mirror-bound one-quark state,

Gabriel Montes-Rojas, Fernando Toledo, Nicolás Bertholet, Kevin Corfield

We study optimal monetary policy when a central bank maximizes a quantile utility objective rather than expected utility. In our framework, the central bank's risk attitude is indexed by the quantile index level, providing a transparent mapping between hawkish/dovish stances and attention to adverse macroeconomic realizations. We formulate the infinite-horiz

M. Rumo, G. Kremer, M. Heber, N. Wind

The transition metal dichalcogenide IrTe2 displays a remarkable series of first-order phase transitions below room temperature, involving lattice displacements as large as 20 percents of the initial bond length. This is nowadays understood as the result of strong electron-phonon coupling leading to the formation of local multicentre dimers that arrange thems

Pedram Fard, Alaleh Azhir, Neguine Rezaii, Jiazi Tian

Artificial intelligence in medicine is built to serve the average patient. By minimizing error across large datasets, most systems deliver strong aggregate accuracy yet falter at the margins: patients with rare variants, multimorbidity, or underrepresented demographics. This average patient fallacy erodes both equity and trust. We propose a different design:

Lingyue Fu, Bolun Zhang, Hao Guan, Yaoming Zhu

Recent advances in code agents have enabled automated software development at the project level, supported by large language models (LLMs). However, existing benchmarks for code agent evaluation face two major limitations. First, creating high-quality project-level evaluation datasets requires extensive domain expertise, leading to prohibitive annotation cos

Christopher W. Churchill

In this two-paper series, we present a straightforward mathematical model for synthesizing quasar absorption line profiles from sight lines through idealized, spatial-kinematic models of the circumgalactic medium (CGM) and their host galaxies. Here, in Paper I, we develop the spatial geometries of multiple galaxy/CGM structures and populate these structures

Jae Wan Shim

The particle number $N$ can be used as a quantitative gauge of non-Gaussianity. This idea extends to systems that are not literally finite by assigning them a notional $N$ that captures the same deviation. For an ideal gas with $N$ insufficiently large for the thermodynamic limit, the velocity distribution that maximises Havrda-Charv\'at entropy departs mark

Suman Sanyal

We introduce Perception Learning (PeL), a paradigm that optimizes an agent's sensory interface $f_\phi:\mathcal{X}\to\mathcal{Z}$ using task-agnostic signals, decoupled from downstream decision learning $g_\theta:\mathcal{Z}\to\mathcal{Y}$. PeL directly targets label-free perceptual properties, such as stability to nuisances, informativeness without collapse

Andrés E. Piatti

Multiple populations (MPs) is a intra-star cluster phenomenon consisting in star-to-star variation of the abundance of some light chemical elements. They have been observed in many star clusters, most of them old globular clusters, populating the Milky Way and other satellite galaxies. Since the study of MPs became more systematic, different astrophysical pa

Hannes Schulze, Lutz Schröder, Üsame Cengiz

Nominal automata models and transition systems serve as formalisms for languages and processes carrying data, and as such relate closely to classical register-based models. The paradigm of name allocation in nominal systems helps alleviate the pervasive computational hardness of register-based models in a tradeoff between expressiveness and computational tra

Hou Jian, Meng Tan, Tian Maozai

We propose a density-free method for frequentist inference on population quantiles, termed Self-Normalized Quantile Empirical Saddlepoint Approximation (SNQESA). The approach builds a self-normalized pivot from the indicator score for a fixed quantile threshold and then employs a constrained empirical saddlepoint approximation to obtain highly accurate tail

Lucio Mauro Carenza, Claudio Basilio Caporusso, Pasquale Digregorio, Antonio Suma

Polar patterns and topological defects are ubiquitous in active matter. In this paper, we study a paradigmatic polar active dumbbell system through numerical simulations, to clarify how polar patterns and defects emerge and shape evolution. We focus on the interplay between these patterns and morphology, domain growth, irreversibility, and compressibility, t

Yiming Zhu, Zhuhong Zhu, Xiaodong Xu, Hongwei Hou

OFDM is widely adopted in modern wireless communication systems, but its power efficiency is limited by high envelope fluctuations. Although various high power-efficiency waveforms have been proposed, most are incompatible with the CP-OFDMA framework and remain ineffective in multi-user downlink transmissions. To address this issue, we propose a constant-env

Luzia A. Trinca, Steven G. Gilmour

Fractional polynomial models are potentially useful for response surfaces investigations. With the availability of routines for fitting nonlinear models in statistical packages they are increasingly being used. However, as in all experiments the design should be chosen such that the model parameters are estimated as efficiently as possible. The design choice

Xin Wang, Rebing Wu

We derive a tight generalization bound for quantum machine learning that is applicable to a wide range of supervised tasks, data, and models. Our bound is both efficiently computable and free of big-O notation. Furthermore, we point out that previous bounds relying on big-O notation may provide misleading suggestions regarding the generalization error. Our g