May 2025 arXiv papers — page 110

Christopher Parmeter, Artem Prokhorov, Valentin Zelenyuk

Big data and machine learning methods have become commonplace across economic milieus. One area that has not seen as much attention to these important topics yet is efficiency analysis. We show how the availability of big (wide) data can actually make detection of inefficiency more challenging. We then show how machine learning methods can be leveraged to ad

Victor Falgas-Ravry, Adva Mond, Rik Sarkar, Victor Souza

A multigraph G is said to be an (s,q)-graph if every s-set of vertices in G supports at most q edges (counting multiplicities). In this paper we consider the maximal sum and product of edge multiplicities in an (s,q)-graph on n vertices. These are multigraph analogues of a problem of Erd\H{o}s raised by F\"uredi and K\"undgen and Mubayi and Terry respectivel

Armin Pournaki, Felix Gaisbauer, Eckehard Olbrich

We investigate the polarization of the German Twittersphere by extracting the main issues discussed and the signaled opinions of users towards those issues based on (re)tweets concerning trending topics. The dataset covers daily trending topics from March 2021 to July 2023. At the opinion level, we show that the online public sphere is largely divided into t

Jennifer D'Souza, Hamed Babaei Giglou, Quentin Münch

Large Language Models (LLMs) drive scientific question-answering on modern search engines, yet their evaluation robustness remains underexplored. We introduce YESciEval, an open-source framework that combines fine-grained rubric-based assessment with reinforcement learning to mitigate optimism bias in LLM evaluators. We release multidisciplinary scienceQ&A d

Xuan Mao, Meng Liu, Yuxiang Li

We study the Neumann initial-boundary value problem for the parabolic-elliptic chemotaxis system, proposed by J\"ager and Luckhaus (1992). We confirm that their comparison methods can be simplified and refined, applicable to seek the critical mass $8\pi$ concerning finite-time blowup in the unit disk. As an application, we deal with a parabolic-elliptic-para

Paolo Fachin, Francesco Macheda, Paolo Barone, Francesco Mauri

Using first principles techniques, we show that infrared optical response can be used to discriminate between the topological and the trivial phases of two-dimensional quantum spin Hall insulators (QSHI). We showcase germanene and jacutingaite, of recent experimental realization, as prototypical systems where the infrared spectrum is discontinuous across the

Tharathep Plienbumrung, Jean-Baptiste Morée, Andrzej M. Oleś, Maria Daghofer

We present an effective two-band model for infinite-layer nickelates NdNiO$_2$ that consisting of a $d$ band centered at Ni site and an interstitial $s$-like band centered at Nd site. To the large extent of the wave functions, we find intersite Coulomb interactions to be substantial. We then use the variational cluster approach together with mean-field theor

Ryan James Campbell, Mihalis Mathioudakis, Carlos Quintero Noda, Peter Keys

The morphology of circular polarisation profiles from solar spectropolarimetric observations encode information about the magnetic field strength, inclination, and line-of-sight velocity gradients. Previous studies used manual methods or unsupervised machine learning (ML) to classify the shapes of circular polarisation profiles. We trained a multi-layer perc

Elizaveta I. Malevannaya, Viktor I. Polozov, Anton I. Ivanov, Aleksei R. Matanin

In this review, we provide a practical guide on protection of superconducting quantum circuits from broadband electromagnetic and infrared-radiation noise by using cryogenic shielding and filtering of microwave lines. Recently, superconducting multi-qubit processors demonstrated quantum supremacy and quantum error correction below the surface code threshold.

Hiroki Shiraishi, Hisao Ishibuchi, Masaya Nakata

Function approximation is a critical task in various fields. However, existing neural network approaches struggle with locally complex or discontinuous functions due to their reliance on a single global model covering the entire problem space. We propose X-KAN, a novel method that optimizes multiple local Kolmogorov-Arnold Networks (KANs) through an evolutio

Faeze Ghorbanpour, Daryna Dementieva, Alexander Fraser

Considering the importance of detecting hateful language, labeled hate speech data is expensive and time-consuming to collect, particularly for low-resource languages. Prior work has demonstrated the effectiveness of cross-lingual transfer learning and data augmentation in improving performance on tasks with limited labeled data. To develop an efficient and

Minh Ngoc Ta, Dong Cao Van, Duc-Anh Hoang, Minh Le-Anh

The growing collaboration between humans and AI models in generative tasks has introduced new challenges in distinguishing between human-written, LLM-generated, and human-LLM collaborative texts. In this work, we collect a multilingual, multi-domain, multi-generator dataset FAIDSet. We further introduce a fine-grained detection framework FAID to classify tex

Yoorhim Cho, Hongyeob Kim, Semin Kim, Youjia Zhang

Visuo-tactile perception aims to understand an object's tactile properties, such as texture, softness, and rigidity. However, the field remains underexplored because collecting tactile data is costly and labor-intensive. We observe that visually distinct objects can exhibit similar surface textures or material properties. For example, a leather sofa and a le

Suhan Guo, Zhenghao Xu, Furao Shen, Jian Zhao

Accurate prediction of contagious disease outbreaks is vital for informed decision-making. Our study addresses the gap between machine learning algorithms and their epidemiological applications, noting that methods optimal for benchmark datasets often underperform with real-world data due to difficulties in incorporating mobility information. We adopt a two-

Patrick Hendra, Josué R. León-Torres, Valerio Flavio Gili, Markus Gräfe

We investigate the dual generation of type-0 and type-II spontaneous parametric down conversions (SPDCs) within a single periodically poled rubidium-doped KTiOPO$_4$ (PPRKTP) waveguide. By coupling a 45 degree linearly polarized pump laser into the waveguide, both SPDC processes are concurrently excited: the type-0 SPDC process is facilitated via third-order

Hui Huang, Yancheng He, Hongli Zhou, Rui Zhang

LLM-as-a-Judge refers to the automatic modeling of preferences for responses generated by Large Language Models (LLMs), which is of significant importance for both LLM evaluation and reward modeling. Although generative LLMs have made substantial progress in various tasks, their performance as LLM-Judge still falls short of expectations. In this work, we pro

Youhong Chen, Debraj Bhattacharjee, Balarko Chaudhuri

We present a purely data-driven method to pinpoint generation plants that significantly contribute to poorly damped oscillations as part of post-event analysis. First, Extended Dynamic Mode Decomposition (EDMD) is applied on PMU data from the point of interconnection (POI) of the plants to obtain the finite-dimensional Koopman operator. Then, modal analysis

Nikhil Sobanbabu, Guanqi He, Tairan He, Yuxiang Yang

Sim-to-real discrepancies hinder learning-based policies from achieving high-precision tasks in the real world. While Domain Randomization (DR) is commonly used to bridge this gap, it often relies on heuristics and can lead to overly conservative policies with degrading performance when not properly tuned. System Identification (Sys-ID) offers a targeted app

Juri Kopp, Gerald Richwien, Markus Heidelmann, Soma Salamon

It is established that the Morin transition, a spin reorientation in hematite, is shifted to lower temperatures with decreasing nanoparticle volume. However, our findings indicate an opposite effect in a series of hematite nanospindles: The particles, synthesized by hydrothermal decomposition of iron(III) chloride solution, with aspect ratios $p$ between $1.

Jian Xiong, Jingbo Zhou, Jingyong Ye, Qiang Huang

Reinforcement learning (RL) has emerged as an effective approach for enhancing the reasoning capabilities of large language models (LLMs), especially in scenarios where supervised fine-tuning (SFT) falls short due to limited chain-of-thought (CoT) data. Among RL-based post-training methods, group relative advantage estimation, as exemplified by Group Relativ

Zaid Abdullah, Mario R. Camana, Abuzar B. M. Adam, Chandan K. Sheemar

We investigate the performance of a multiple reconfigurable intelligence surface (RIS)-aided millimeter wave (mmWave) beamspace multiple-input multiple-output (MIMO) system with multiple users (UEs). We focus on a challenging scenario in which the direct links between the base station (BS) and all UEs are blocked, and communication is facilitated only via RI

Yudong Wang, Hongjiong Tian

In this work, we present a general technique for establishing the strong convergence of numerical methods for stochastic delay differential equations (SDDEs) in the infinite horizon. This technique can also be extended to analyze certain continuous function-valued segment processes associated with the numerical methods, facilitating the numerical approximati

Henry Dumant

The purpose of this paper is to introduce the resonances of Dirac operators by continuing meromorphically the truncated resolvent and to establish a result about their localization : a kind of Rellich Theorem. Firstly, we consider the case of the Dirac operator in an external field which is essentially bounded and compactly supported. Secondly, we consider t

Luxi Lin, Zhihang Lin, Zhanpeng Zeng, Rongrong Ji

This paper introduces Multimodal Speculative Decoding (MSD) to accelerate Multimodal Large Language Models (MLLMs) inference. Speculative decoding has been shown to accelerate Large Language Models (LLMs) without sacrificing accuracy. However, current speculative decoding methods for MLLMs fail to achieve the same speedup as they do for LLMs. To address this

Petr Sedlák, Miroslav Frost, Martin Ševčík, Lukáš Kadeřávek

A laser-ultrasonic approach was used to measure elastic properties of a superelastic nickel-titanium wire with the aim to evaluate their evolution with stress and temperature in stress-induced martensite. It was observed that this evolution can be well described by a single smooth surface in the stress-temperature space, with the values of Young's modulus ra

Ivan Ridkokasha, Andres Aramburo-Garcia, Kyrylo Bondarenko, Anastasia Sokolenko

The 3D distribution of matter at small scales encodes valuable information about the nature of dark matter and other fundamental physics. A prominent probe of such scales outside galaxies is the Lyman-alpha forest, which studies absorption features in the spectra of high-redshift quasars caused by neutral hydrogen. The measured quantity is the power spectrum

Jianfei Zhao, Feng Zhang, Xin Sun, Chong Feng

Due to the unidirectional masking mechanism, Decoder-Only models propagate information from left to right. LVLMs (Large Vision-Language Models) follow the same architecture, with visual information gradually integrated into semantic representations during forward propagation. Through systematic analysis, we observe that the majority of the visual information

Shaolin Zhu, Tianyu Dong, Bo Li, Deyi Xiong

In this paper, we present FuxiMT, a novel Chinese-centric multilingual machine translation model powered by a sparsified large language model (LLM). We adopt a two-stage strategy to train FuxiMT. We first pre-train the model on a massive Chinese corpus and then conduct multilingual fine-tuning on a large parallel dataset encompassing 65 languages. FuxiMT inc

Vladimir Panov, Anton Ryabchenko

This study focuses on statistical inference for the class of quasi-infinitely divisible (QID) distributions, which was recently introduced by Lindner, Pan and Sato (2018). The paper presents a Fourier approach, based on the analogue of the L{\'e}vy-Khintchine theorem with a signed spectral measure. We prove that for some subclasses of QID distributions, the

Yuanyuan Chang, Yinghua Yao, Tao Qin, Mengmeng Wang

Text-to-image diffusion models have emerged as powerful tools for high-quality image generation and editing. Many existing approaches rely on text prompts as editing guidance. However, these methods are constrained by the need for manual prompt crafting, which can be time-consuming, introduce irrelevant details, and significantly limit editing performance. I

Haibo Wu, Marina I. Knight, Keiland W. Cooper, Norbert J. Fortin

Understanding the evolving dependence between two clusters of multivariate signals is fundamental in neuroscience and other domains where sub-networks in a system interact dynamically over time. Despite the growing interest in multivariate time series analysis, existing methods for between-clusters dependence typically rely on the assumption of stationarity

Mouad Elaarabi, Domenico Borzacchiello, Philippe Le Bot, Nathan Lauzeral

In this work, we explore the integration of Sequence Encoding for Online Parameter Identification with Physics-Informed Neural Networks to create a model that, once trained, can be utilized for real time applications with variable parameters, boundary conditions, and initial conditions. Recently, the combination of PINNs with Sparse Regression has emerged as

Zengfeng Huang, Zhongzheng Xiong, Xiaoyi Zhu, Zhewei Wei

We consider the problems of distributed heavy hitters and frequency moments in both the coordinator model and the distributed tracking model (also known as the distributed functional monitoring model). We present simple and optimal (up to logarithmic factors) algorithms for $\ell_p$ heavy hitters and $F_p$ estimation ($p \geq 2$) in these distributed models.

Felicia Servina Djuang, Indah Emilia Wijayanti, Yeni Susanti

Let $R$ be a finite ring with identity. The idempotent graph $I(R)$ is the graph whose vertex set consists of the non-trivial idempotent elements of $R$, where two distinct vertices $x$ and $y$ are adjacent if and only if $xy = yx = 0$. The clean graph $Cl(R)$ is a graph whose vertices are of the form $(e, u)$, where $e$ is an idempotent element and $u$ is a

Noy Vaisleib, Michal Arbel-Haddad, Amir Goldbourt

Geopolymers are aluminosilicate materials that exhibit effective immobilization properties for low-level radioactive nuclear waste, and more specifically for the immobilization of radioactive cesium. The identification of the cesium-binding sites and their distribution between the different phases making up the geopolymeric matrix can be obtained using solid

Nicanor Carrasco-Vargas

The study of subshifts on groups different from $\mathbb{Z}$, such as $\mathbb{Z}^d$, $d\geq 2$, has been a subject of intense research in recent years. These investigations have unveiled aremarkable connection between dynamics and recursion theory. Different questions about the dynamics of these systems have been answered in recursion-theoretical terms. In

Ziyu Liu, Yuhang Zang, Yushan Zou, Zijian Liang

A key trend in Large Reasoning Models (e.g., OpenAI's o3) is the native agentic ability to use external tools such as web browsers for searching and writing/executing code for image manipulation to think with images. In the open-source research community, while significant progress has been made in language-only agentic abilities such as function calling and

A. A. Solovykh, N. E. Rybin, I. S. Novikov, A. V. Shapeev

Accounting for nuclear quantum effects (NQEs) can significantly alter material properties at finite temperatures. Atomic modeling using the path-integral molecular dynamics (PIMD) method can fully account for such effects, but requires computationally efficient and accurate models of interatomic interactions. Empirical potentials are fast but may lack suffic

Haijun Li, Tianqi Shi, Zifu Shang, Yuxuan Han

Machine translation (MT) has become indispensable for cross-border communication in globalized industries like e-commerce, finance, and legal services, with recent advancements in large language models (LLMs) significantly enhancing translation quality. However, applying general-purpose MT models to industrial scenarios reveals critical limitations due to do

Syed M. Arslan, Saif Al-Kuwari, M. T. Rahim, Hashir Kuniyal

Quantum Key Distribution (QKD) is based on the laws of quantum mechanics to enable provably secure communication. Despite its theoretical security promise, practical QKD systems are vulnerable to serious attacks, including side-channel attacks and detector loopholes, and assumes a trusted device characterization. Device-Independent Quantum Key Distribution (

Ziang Wang, Amir Aryani

This technical report presents a natural language processing (NLP)-based approach for systematically classifying scientific literature on childhood speech disorders. We retrieved and filtered 4,804 relevant articles published after 2015 from the PubMed database using domain-specific keywords. After cleaning and pre-processing the abstracts, we applied two to

Theis E. Jendal, Matteo Lissandrini, Peter Dolog, Katja Hose

Inductive Recommender Systems are capable of recommending for new users and with new items thus avoiding the need to retrain after new data reaches the system. However, these methods are still trained on all the data available, requiring multiple days to train a single model, without counting hyperparameter tuning. In this work we focus on graph-based recomm

Germain Vivier-Ardisson, Mathieu Blondel, Axel Parmentier

Integrating combinatorial optimization layers into neural networks has recently attracted significant research interest. However, many existing approaches lack theoretical guarantees or fail to perform adequately when relying on inexact solvers. This is a critical limitation, as many operations research problems are NP-hard, often necessitating the use of ne

Bin-Bin Gao, Xiaochen Chen, Zhongyi Huang, Congchong Nie

This paper focus on few-shot object detection~(FSOD) and instance segmentation~(FSIS), which requires a model to quickly adapt to novel classes with a few labeled instances. The existing methods severely suffer from bias classification because of the missing label issue which naturally exists in an instance-level few-shot scenario and is first formally propo

Raghav Singhal, Kaustubh Ponkshe, Rohit Vartak, Praneeth Vepakomma

Large Language Models have demonstrated strong performance across a wide range of tasks, but adapting them efficiently to new domains remains a key challenge. Parameter-Efficient Fine-Tuning (PEFT) methods address this by introducing lightweight, trainable modules while keeping most pre-trained weights fixed. The prevailing approach, LoRA, models updates usi

Marian Nita, Mugurel Tolea, Catalina Marinescu

We propose an algorithm for determining the zeros of the electric conductivity in large molecular nanonstructures such as graphene sheets. To this end, we employ the inverse graph method, whereby non-zeros of the Green's functions are represented graphically by a segment connecting two atomic sites, to visually signal the existence of a conductance zero as a

Kapila W. S. Palitharathna, Constantinos Psomas, Ioannis Krikidis

This paper presents a tunable liquid lens (TLL)-assisted indoor mobile visible light communication system. To mitigate performance degradation caused by user mobility and random receiver orientation, an electrowetting cuboid TLL is used at the receiver. By dynamically controlling the orientation angle of the liquid surface through voltage adjustments, signal

Yequan Wang, Aixin Sun

Artificial General Intelligence (AGI) is often envisioned as inherently embodied. With recent advances in robotics and foundational AI models, we stand at the threshold of a new era-one marked by increasingly generalized embodied AI systems. This paper contributes to the discourse by introducing a systematic taxonomy of Embodied AGI spanning five levels (L1-

Illia Horenko, Davide Bassetti, Lukáš Pospíšil

Entropic measures like Shannon entropy (SE), its quantum mechanical analogue von Neumann entropy, and Kullback-Leibler divergence (KL) are key components in many tools used in physics, information theory, machine learning (ML) and quantum computing. Besides of the significant amounts of SE and KL computations required in these fields, the singularity of thei

Hakaze Cho, Peng Luo, Mariko Kato, Rin Kaenbyou

In-context Learning (ICL) utilizes structured demonstration-query inputs to induce few-shot learning on Language Models (LMs), which are not originally pre-trained on ICL-style data. To bridge the gap between ICL and pre-training, some approaches fine-tune LMs on large ICL-style datasets by an end-to-end paradigm with massive computational costs. To reduce s

Adrijan Rogan, Andrej Kolar-Požun, Gregor Kosec

This paper focuses on RBF-based meshless methods for approximating differential operators, one of the most popular being RBF-FD. Recently, a hybrid approach was introduced that combines RBF interpolation and traditional finite difference stencils. We compare the accuracy of this method and RBF-FD on a two-dimensional Poisson problem for standard five-point a

Sule Bai, Mingxing Li, Yong Liu, Jing Tang

Traditional visual grounding methods primarily focus on single-image scenarios with simple textual references. However, extending these methods to real-world scenarios that involve implicit and complex instructions, particularly in conjunction with multiple images, poses significant challenges, which is mainly due to the lack of advanced reasoning ability ac

Daniel G. Boettger, Shane R. Keating, Michael L. Banner, Russel P. Morison Xavier Barthélémy

The process of breaking in surface gravity waves can be characterized by two distinct stages. Breaking onset, defined as the first visible surface manifestation of breaking, is preceded by breaking inception, which is characterized by the initiation of an irreversible process within the crest that leads inevitably to breaking. Breaking inception diagnostics

Sašo Grozdanov, Mile Vrbica

The physics of gravitational waves and other classical fields on specifically four-dimensional backgrounds of black holes exhibits electric-magnetic-like dualities. In this paper, we discuss the structure of such dualities in terms of geometrical quantities with a physically-intuitive interpretation. In turn, we explain the interplay between the algebraic st

Ramūnas Garunkštis, Athanasios Sourmelidis, Jörn Steuding

We provide conditional and unconditional asymptotic formulae for the exponential sums $\sum_\gamma\,\gamma^{-i\tau}$, where the summation is over the ordinates of the nontrivial zeros $\rho=\beta+i\gamma$ of the Riemann zeta-function. In particular, the obtained results are related to the Lindel\"of Hypothesis for these ordinates (in the sense of Gonek et al

Jianing An, Luyang Jiang, Jie Luo, Wenjun Wu

Visual understanding requires interpreting both natural scenes and the textual information that appears within them, motivating tasks such as Visual Question Answering (VQA). However, current VQA benchmarks overlook scenarios with visually embedded questions, whereas advanced agents should be able to see the question without separate text input as humans. We

Darpan Aswal, Siddharth D Jaiswal

Safety-aligned LLMs remain vulnerable to digital phenomena like textese that introduce non-canonical perturbations to words but preserve the phonetics. We introduce CMP-RT (code-mixed phonetic perturbations for red-teaming), a novel diagnostic probe that pinpoints tokenization as the root cause of this vulnerability. A mechanistic analysis reveals that phone

Ignatios Antoniadis, Auttakit Chatrabhuti, Jules Cunat, Hiroshi Isono

It was proposed that five-dimensional (5D) inflation can blow up the size of a compact dimension from the 5D Planck length to the micron size, as required by the dark dimension proposal, relating the weakness of the actual gravitational force to the size of the observable universe. Moreover, it was shown that 5D inflation can generate the (approximate) flat

Atanu Manna, Bhakti Bhusan Manna

In this article, we study the following Hardy-Sobolev-Maz'ya type equation: \begin{equation} -\Delta u - \mu \frac{u}{|z|^2} = \frac{|u|^{q-2}u}{|z|^t}, \quad u \in D^{1,2} (\mathbb{R}^n), \end{equation} where $x = (y,z) \in \mathbb{R}^h \times \mathbb{R}^k = \mathbb{R}^n$, with $n \geq 5$, $2 < k <n$, and $t = n - \frac{(n-2)q}{2}$. We establish the existen

Jiefeng Ye, Zhigao Huang, Xianxin Wu, Jian-Min Zhang

Fermionic and bosonic localized states induced by geometric frustration in the kagome lattice provide a distinctive research platform for investigating emergent exotic quantum phenomena in strongly correlated systems. Here, we report the discovery of coexisting electronic and phononic flat bands induced by geometric frustration in a novel kagome superconduct

Kaixing Yang, Xulong Tang, Ziqiao Peng, Yuxuan Hu

Music-to-dance generation represents a challenging yet pivotal task at the intersection of choreography, virtual reality, and creative content generation. Despite its significance, existing methods face substantial limitation in achieving choreographic consistency. To address the challenge, we propose MatchDance, a novel framework for music-to-dance generati

H. Mannel, J. Zöllner, E. Kleinherbers, M. Zöllner

Stochastic resonance is a phenomenon in which fluctuations enhance an otherwise weak signal. It has been found in many different systems in paleoclimatology, biology, medicine, and physics. The classical stochastic resonance due to thermal noise has recently been experimentally extended to the quantum regime, where the fundamental randomness of individual qu

Tomi Suomi, Petri Ihantola, Tommi Mikkonen, Niko Mäkitalo

Agile software development relies on self-organized teams, underlining the importance of individual responsibility. How developers take responsibility and build ownership are influenced by external factors such as architecture and development methods. This paper examines the existing literature on ownership in software engineering and in psychology, and argu

Dotan Gazith, Barak Zackay

Fast Radio Bursts (FRBs) are bright and short radio flashes of cosmological origin. Although a great number of FRBs were detected in the last two decades, their progenitors and the physical processes that create them are unknown. In recent years, magnetars have been proposed as one of the leading progenitor candidates. A striking feature that can hint at suc

Jie Li, Shengwei Tian, Long Yu, Xin Ning

The Chamfer Distance (CD) is a cornerstone objective function for point cloud completion, yet its inherent symmetric weighting mechanism limits the quality of the generated results. By penalizing local detail deviations and global coverage deficiencies equally, standard CD often causes structural defects such as point aggregation and incomplete spatial struc

Jorge Fabila, Lidia Garrucho, Víctor M. Campello, Carlos Martín-Isla

This study explores the use of Federated Learning (FL) for tuberculosis (TB) diagnosis using chest X-rays in low-resource settings across Africa. FL allows hospitals to collaboratively train AI models without sharing raw patient data, addressing privacy concerns and data scarcity that hinder traditional centralized models. The research involved hospitals and

Minwu Kim, Anubhav Shrestha, Safal Shrestha, Aadim Nepal

Recent studies have shown that reinforcement learning with verifiable rewards (RLVR) enhances overall accuracy (pass@1) but often fails to improve capability (pass@k) of LLMs in reasoning tasks, while distillation can improve both. In this paper, we investigate the mechanisms behind these phenomena. First, we demonstrate that RLVR struggles to improve capabi

Mattes Mollenhauer, Nicole Mücke, Dimitri Meunier, Arthur Gretton

This paper examines the performance of ridge regression in reproducing kernel Hilbert spaces in the presence of noise that exhibits a finite number of higher moments. We establish excess risk bounds consisting of subgaussian and polynomial terms based on the well known integral operator framework. The dominant subgaussian component allows to achieve converge

Zhoulai Fu, Freek Verbeek, Binoy Ravindran

This work advances floating-point program verification by introducing Augmented Weak-Distance (AWD), a principled extension of the Weak-Distance (WD) framework. WD is a recent approach that reformulates program analysis as a numerical minimization problem, providing correctness guarantees through non-negativity and zero-target correspondence. It consistently

Sizhe Yuen, Ting Su, Ziyang Wang, Yali Du

A question-answering (QA) system is to search suitable answers within a knowledge base. Current QA systems struggle with queries requiring complex reasoning or real-time knowledge integration. They are often supplemented with retrieval techniques on a data source such as Retrieval-Augmented Generation (RAG). However, RAG continues to face challenges in handl

Qu Wang, Yan Xia

Link prediction in dynamic networks remains a fundamental challenge in network science, requiring the inference of potential interactions and their evolving strengths through spatiotemporal pattern analysis. Traditional static network methods have inherent limitations in capturing temporal dependencies and weight dynamics, while tensor-based methods offer a

Gabriele Gennari, Christian Gorges, Fabian Denner, Berend van Wachem

The handling of topology changes in two-phase flows, such as breakup or coalescence of interfaces, with front tracking is a well-known problem that requires an additional effort to perform explicit manipulations of the Lagrangian front. In this work, we present an approach that allows to perform topology changes with interfaces made of connected triangular e

Li Wang, Xin Yu, Xuxin Lv, Gangzheng Ai

With the rapid advancement of unmanned aerial vehicles (UAVs) and missile technologies, perimeter-defense game between attackers and defenders for the protection of critical regions have become increasingly complex and strategically significant across a wide range of domains. However, existing studies predominantly focus on small-scale, simplified two-dimens

Rasmus S. Nielsen, Ángel Labordet Álvarez, Axel G. Medaille, Ivan Caño

Chalcohalides are an emerging family of semiconductors with irresistible material properties, shaped by the intricate interplay between their unique structural chemistry and vibrational dynamics. Despite their promise for next-generation solar energy conversion devices, their intrinsic optoelectronic properties remain largely unexplored. Here, we focus on th

Yurii Belov, Aleksei Kulikov

Let $g\in L^2(\mathbb{R})$ be a strictly decreasing continuous function supported on $\mathbb{R}_+$ such that for all $t > 0$ we have $g(x+t)\le q(t)g(x)$ for some $q(t)<1$. We prove that the Gabor system $$\mathcal{G}(g;\alpha,\beta):=\{g_{m,n}\}_{m,n\in\mathbb{Z}}=\{e^{2\pi i \beta m x}g(x-\alpha n)\}_{m,n\in\mathbb{Z}}$$ always forms a frame in $L^2(\math

Flavio Di Martino, Franca Delmastro

The widespread adoption of wearable sensors has the potential to provide massive and heterogeneous time series data, driving the use of Artificial Intelligence in human sensing applications. However, data collection remains limited due to stringent ethical regulations, privacy concerns, and other constraints, hindering progress in the field. Synthetic data g

Song Shao, Hui Xu

In this paper, we develop several structure theorems concerning commuting transformations and minimal $\mathbb{R}$-flows. Specifically, we show that if $(X,S)$, $(X,T)$ are minimal systems with $S$ and $T$ being commutative, then they possess an identical higher-order regionally proximal relation. Consequently, both $(X, S)$ and $(X, T)$ share the same incre

Yang Hu, Runchen Wang, Stephen Chong Zhao, Xuhui Zhan

We introduce Perceptual-Initialization (PI), a paradigm shift in visual representation learning that incorporates human perceptual structure during the initialization phase rather than as a downstream fine-tuning step. By integrating human-derived triplet embeddings from the NIGHTS dataset to initialize a CLIP vision encoder, followed by self-supervised lear

Ingo Bloch, Ben Bruers, Heiko Lacker, Peilin Li

Radiation-hard silicon sensors used in high-energy physics require a high electric field and are susceptible to surface breakdown. This study aims to improve the understanding of the underlying mechanisms by developing new methods to probe the electric field at surface near the sensor's edge. For planar sensors, avalanche breakdown primarily occurs at the Si

Shibo Feng, Zhicheng Chen, Xi Xiao, Zhong Zhang

Discrete Token Modeling (DTM), which employs vector quantization techniques, has demonstrated remarkable success in modeling non-natural language modalities, particularly in time series generation. While our prior work SDformer established the first DTM-based framework to achieve state-of-the-art performance in this domain, two critical limitations persist i

Kosmas Alexandridis, Vasileios Titopoulos, Giorgos Dimitrakopoulos

The transformer's attention mechanism has revolutionized AI and machine learning, with its efficient computation being crucial to its performance. However, calculating attention involves matrix operations interspersed with softmax rescaling, which inherently slows down computation and requires processing the entire input sequence. Building on online softmax

Aftab Hussain, Sadegh AlMahdi Kazemi Zarkouei, Md Rafiqul Islam Rabin, Mohammad Amin Alipour

Large language models of code exhibit high capability in performing diverse software engineering tasks, such as code translation, defect detection, text-to-code generation, and code summarization. While their ability to enhance developer productivity has spurred widespread use, these models have also seen substantial growth in size, often reaching billions o

Fariba Osali, Khwaja Zubair Sediqi, Oliver Gasser

Since the standardization of IPv6 in 1998, both versions of the Internet Protocol have coexisted in the Internet. Clients usually run algorithms such as Happy Eyeballs, to decide whether to connect to an IPv4 or IPv6 endpoint for dual-stack domains. To identify whether two addresses belong to the same device or service, researchers have proposed different fo

Svante Janson

We give bounds for (central) moments for balanced P\'olya urns under very general conditions. In some cases, these bounds imply that moment convergence holds in earlier known results on asymptotic distribution. The results overlap with previously known results, but are here given more generally and with a simpler proof.

Xinshen Zhang, Zhen Ye, Xu Zheng

Omnidirectional images (ODIs), with their 360{\deg} field of view, provide unparalleled spatial awareness for immersive applications like augmented reality and embodied AI. However, the capability of existing multi-modal large language models (MLLMs) to comprehend and reason about such panoramic scenes remains underexplored. This paper addresses this gap by

Thai-Hoc Vu, Ngo Hoang Tu, Thien Huynh-The, Kyungchun Lee

The evolution from fifth-generation (5G) to sixth-generation (6G) networks is driving an unprecedented demand for advanced machine learning (ML) solutions. Deep learning has already demonstrated significant impact across mobile networking and communication systems, enabling intelligent services such as smart healthcare, smart grids, autonomous vehicles, aeri

Sela Fried

Inspired by OEIS sequence A377912, which consists of the nonnegative integers in which every even digit (except possibly the last) is immediately followed by a strictly larger digit, we define even-up and odd-up words over an alphabet of size~$k$ via similar constraints. We introduce and analyze weak and cyclic variants of these words, deriving explicit gene

Tuc Nguyen, Yifan Hu, Thai Le

Recent advancements in large language models (LLMs) have been fueled by large scale training corpora drawn from diverse sources such as websites, news articles, and books. These datasets often contain explicit user information, such as person names and addresses, that LLMs may unintentionally reproduce in their generated outputs. Beyond such explicit content

Robin Yoël Engel, Patrick Lömker

POLEVAL provides a software toolbox for collaborative, persistent and reproducible analysis of XPS experiments. It allows to treat, analyse and visualise the results of an extended experimental campaign in a single python notebook in a consistent manner. Managing experimental data in adequate objects enables experimentalists to process and analyse measuremen

Abdallah Abuaisha, Bojie Shen, Daniel Harabor, Peter Stuckey

Delays in public transport are common, often impacting users through prolonged travel times and missed transfers. Existing solutions for handling delays remain limited; backup plans based on historical data miss opportunities for earlier arrivals, while snapshot planning accounts for current delays but not future ones. With the growing availability of live d

Bikash K. Behera, Saif Al-Kuwari, Ahmed Farouk

A brain-computer interface (BCI) system enables direct communication between the brain and external devices, offering significant potential for assistive technologies and advanced human-computer interaction. Despite progress, BCI systems face persistent challenges, including signal variability, classification inefficiency, and difficulty adapting to individu

Chunlin Yi, Jason A. Galvis, Gustavo Machado

African swine fever (ASF) is a highly contagious viral disease that poses a significant threat to the swine industry, requiring stringent control measures, including movement restrictions that delay pig placements, impacting business continuity. The number and economic impact of unplaced healthy animals due to control zone restrictions remains unmeasured. Th

Andre Catarino, Rui Melo, Rui Abreu, Luis Cruz

The widespread adoption of dynamic Time-of-Use (dToU) electricity tariffs requires accurately identifying households that would benefit from such pricing structures. However, the use of real consumption data poses serious privacy concerns, motivating the adoption of synthetic alternatives. In this study, we conduct a comparative evaluation of four synthetic

Ni Ding, Miao Qiao, Jiaxing Xu, Yiping Ke

This paper proposes $\alpha$-GAN, a generative adversarial network using R\'{e}nyi measures. The value function is formulated, by R\'{e}nyi cross entropy, as an expected certainty measure incurred by the discriminator's soft decision as to where the sample is from, true population or the generator. The discriminator tries to maximize the R\'{e}nyi certainty

Xavier Goaoc, Arnau Padrol

Let $P$ be a finite full-dimensional point configuration in $\mathbb{R}^d$. We show that if a point configuration $Q$ has the property that all finite chirotopes realizable by adding (generic) points to $P$ are also realizable by adding points to $Q$, then $P$ and $Q$ are equal up to a direct affine transform. We also show that for any point configuration $P

Viola Negroni, Davide Salvi, Paolo Bestagini, Stefano Tubaro

With the proliferation of speech deepfake generators, it becomes crucial not only to assess the authenticity of synthetic audio but also to trace its origin. While source attribution models attempt to address this challenge, they often struggle in open-set conditions against unseen generators. In this paper, we introduce the source verification task, which,

Noelia Sánchez González, Joseph P. Conlon, Edmund J. Copeland, Edward Hardy

We study the string theory dynamics of the volume scalar rolling down an exponential potential during the period between inflation and reheating, in a background of cosmic superstring loops. In the context of the LVS potential, we demonstrate the existence of a novel string loop attractor tracker solution, in which 75% of the energy density of the universe i

Adeline Guéret, Wolf-Peter Schill, Felix Schmidt

Prosumer households that generate and store electricity from rooftop PV installations play an increasing role in electricity markets around the world. As retail tariffs usually do not convey time-varying wholesale price signals to households and the rollout of smart meters is low in many countries, prosumers do not necessarily self-consume and feed-in solar

Kaustubh Ponkshe, Shaan Shah, Raghav Singhal, Praneeth Vepakomma

Large Language Models (LLMs) rely on safety alignment to produce socially acceptable responses. However, this behavior is known to be brittle: further fine-tuning, even on benign or lightly contaminated data, can degrade safety and reintroduce harmful behaviors. A growing body of work suggests that alignment may correspond to identifiable directions in weigh

Guosheng Liang, Longguang Zhong, Ziyi Yang, Xiaojun Quan

Large reasoning models (LRMs) excel at solving complex tasks by leveraging long chain-of-thought (CoT) reasoning. However, this often leads to overthinking on simple tasks, resulting in unnecessary computational overhead. We observe that LRMs inherently possess the capability for efficient short CoT reasoning, which can be reliably elicited through prompt de