March 2025 arXiv papers — page 13

F. Thelen, F. Lourens, A. Ludwig

Efficient characterization of surface compositions across high-dimensional materials spaces is critical for accelerating the discovery of surface-dominated functional materials. While X-ray photoelectron spectroscopy allows detailed surface composition investigation, it remains a time-intensive technique. In this work, it is demonstrated that Gaussian proces

Dim Shaiakhmetov, Gulnaz Gimaletdinova, Kadyrmamat Momunov, Selcuk Cankurt

Proper recitation of the Quran, adhering to the rules of Tajweed, is crucial for preventing mistakes during recitation and requires significant effort to master. Traditional methods of teaching these rules are limited by the availability of qualified instructors and time constraints. Automatic evaluation of recitation can address these challenges by providin

Andrew Pocklington, Aashish A. Clerk

We present a new technique for efficiently simulating (in polynomial time) a class of one-dimensional (1D) dissipative spin chains that, when mapped to fermions, have quadratic Hamiltonians, with the only nonlinearity coming from Jordan-Wigner strings appearing in the jump operators, despite the fact that these models cannot be mapped to quadratic fermionic

Eytan Kats, Kai Geißler, Jochen G. Hirsch, Stefan Heldman

Automated patient positioning is a crucial step in streamlining MRI workflows and enhancing patient throughput. RGB-D camera-based systems offer a promising approach to automate this process by leveraging depth information to estimate internal organ positions. This paper investigates the feasibility of a learning-based framework to infer approximate internal

Taeyoung Jeong, Kun Hee Ye, Seungjae Yoon, Dohyun Kim

Multiscale modeling, which integrates material properties from ab initio calculations into continuum-scale simulations, is a promising strategy for optimizing semiconductor devices. However, a key challenge remains: while ab initio methods provide diffusion parameters specific to individual migration paths, continuum equations require a single effective set

Max Hort, Leon Moonen

Software is used in critical applications in our day-to-day life and it is important to ensure its correctness. One popular approach to assess correctness is to evaluate software on tests. If a test fails, it indicates a fault in the software under test; if all tests pass correctly, one may assume that the software is correct. However, the reliability of the

Pranjal Paul, Vineeth Bhat, Tejas Salian, Mohammad Omama

Global localization is a critical problem in autonomous navigation, enabling precise positioning without reliance on GPS. Modern global localization techniques often depend on dense LiDAR maps, which, while precise, require extensive storage and computational resources. Recent approaches have explored alternative methods, such as sparse maps and learned feat

Javier Argüello-Luengo, Alejandro González-Tudela, J. Ignacio Cirac

Here, we propose a platform based on ultra-cold fermionic molecules trapped in optical lattices to simulate nonadiabatic effects, as they appear in certain molecular dynamical problems. The idea consists of a judicious choice of two rotational states as the simulated electronic or nuclear degrees of freedom, in which their dipolar interactions induce the req

Xingcheng Zhou, Xuyuan Han, Feng Yang, Yunpu Ma

We present OpenDriveVLA, a Vision Language Action model designed for end-to-end autonomous driving, built upon open-source large language models. OpenDriveVLA generates spatially grounded driving actions by leveraging multimodal inputs, including 2D and 3D instance-aware visual representations, ego vehicle states, and language commands. To bridge the modalit

Viktor Stein, Wuchen Li

Stein variational gradient descent (SVGD) is a kernel-based particle method for sampling from a target distribution, e.g., in generative modeling and Bayesian inference. SVGD does not require estimating the gradient of the log-density, which is called score estimation. In practice, SVGD can be slow compared to score-estimation based sampling algorithms. To d

Ying Tai, Nikai Du, Rui Xie, Zhennan Chen

In this paper, we present TextCrafter, a Complex Visual Text Generation (CVTG) framework inspired by selective visual attention in cognitive science, and introduce the "Text Insulation-and-Attention" mechanisms. To implement the selective-attention principle that selection operates on discrete objects, we propose a novel Bottleneck-aware Constrained Reinforc

Jun Hu, Mengru Xue, Cheng Yao, Yuan Feng

Connectivity enabled by technologies such as the Internet of Things, Artificial Intelligence, Big Data, and Cloud Computing is rapidly transforming our interactions with the world and with each other. It reshapes social interactions, fostering collaboration, creativity, and unprecedented access to information and resources. However, this connected world and

Chenglong Lu, Shen Liang, Xuewei Wang, Wei Wang

Vision Transformers (ViTs) have computational costs scaling quadratically with the number of tokens, calling for effective token pruning policies. Most existing policies are handcrafted, lacking adaptivity to varying inputs. Moreover, they fail to consider the sequential nature of token pruning across multiple layers. In this work, for the first time (as far

Karan Mukhi, Georg Loho, Alessandro Abate

It is well established that the aggregate flexibility inherent in populations of distributed energy resources (DERs) can be leveraged to mitigate the intermittency and uncertainty associated with renewable generation, while also providing ancillary grid services. To enable this, aggregators must effectively represent the flexibility in the populations they c

Daniel J. Gauthier, Andrew Pomerance, Erik Bollt

We extend an advanced variation of a machine learning algorithm, next-generation reservoir Computing (NGRC), to forecast the dynamics of the Ikeda map of a chaotic laser. The machine learning model is created by observing time-series data generated by the Ikeda map, and the trained model is used to forecast the behavior without any input from the map. The Ik

Maofu Liu, Xin Jiang, Xiaokang Zhang

Referring Remote Sensing Image Segmentation (RRSIS) is a challenging task, aiming to segment specific target objects in remote sensing (RS) images based on a given language expression. Existing RRSIS methods typically employ coarse-grained unidirectional alignment approaches to obtain multimodal features, and they often overlook the critical role of language

Junzhu Mao, Yang Shen, Jinyang Guo, Yazhou Yao

Token compression is essential for reducing the computational and memory requirements of transformer models, enabling their deployment in resource-constrained environments. In this work, we propose an efficient and hardware-compatible token compression method called Prune and Merge. Our approach integrates token pruning and merging operations within transfor

Dennis Müller

This article analyzes the emerging ethical turn in mathematics education, arguing that it is a nuanced extension of the sociopolitical turn. While sociopolitical studies of mathematics have highlighted systemic issues and group concerns (e.g., equity, diversity, exclusion), the newer scholarship on ethics in mathematics presents a sharpened focus on the indi

Maofu Liu, Jiahui Liu, Xiaokang Zhang

Remote sensing image captioning aims to generate semantically accurate descriptions that are closely linked to the visual features of remote sensing images. Existing approaches typically emphasize fine-grained extraction of visual features and capturing global information. However, they often overlook the complementary role of textual information in enhancin

Yuhang Yang, Ke Fan, Shangkun Sun, Hongxiang Li

The rapid advancement of video generation has rendered existing evaluation systems inadequate for assessing state-of-the-art models, primarily due to simple prompts that cannot showcase the model's capabilities, fixed evaluation operators struggling with Out-of-Distribution (OOD) cases, and misalignment between computed metrics and human preferences. To brid

Aimira Baitieva, Yacine Bouaouni, Alexandre Briot, Dick Ameln

Anomaly detection (AD) is essential for automating visual inspection in manufacturing. This field of computer vision is rapidly evolving, with increasing attention towards real-world applications. Meanwhile, popular datasets are typically produced in controlled lab environments with artificially created defects, unable to capture the diversity of real produc

Bo Zhang, Xiubo Liang, Wei Song, Yulu Chen

Within modern warehouse scenarios, the rapid expansion of e-commerce and increasingly complex, multi-level storage environments have exposed the limitations of traditional AGV (Automated Guided Vehicle) path planning methods--often reliant on static 2D models and expert-tuned heuristics that struggle to handle dynamic traffic and congestion. Addressing these

Bohao Xing, Kaishen Yuan, Zitong Yu, Xin Liu

Facial Action Units (AUs) detection is a cornerstone of objective facial expression analysis and a critical focus in affective computing. Despite its importance, AU detection faces significant challenges, such as the high cost of AU annotation and the limited availability of datasets. These constraints often lead to overfitting in existing methods, resulting

Abhishek Mund, Arindam Das

This article outlines a thorough stability analysis by means of both theoretical and experimental modeling for Omni-phobic Lubricant Impregnated Surfaces (LIS). The liquid-repellent properties, particularly with regard to water and oil, have gained substantial attention due to their numerous potential applications. The theoretical section of this study focus

Max Hort, Linas Vidziunas, Leon Moonen

Recent advances in defect detection use language models. Existing works enhanced the training data to improve the models' robustness when applied to semantically identical code (i.e., predictions should be the same). However, the use of semantically identical code has not been considered for improving the tools during their application - a concept closely re

Jongseo Lee, Joohyun Chang, Dongho Lee, Jinwoo Choi

We propose Cross-Attention in Audio, Space, and Time (CA^2ST), a transformer-based method for holistic video recognition. Recognizing actions in videos requires both spatial and temporal understanding, yet most existing models lack a balanced spatio-temporal understanding of videos. To address this, we propose a novel two-stream architecture, called Cross-At

Hanlin Cai, Houtianfu Wang, Haofan Dong, Ozgur B. Akan

The expansion of sixth-generation (6G) wireless networks into space introduces technical challenges that conventional bit-oriented communication approaches cannot efficiently address, including intermittent connectivity, severe latency, limited bandwidth, and constrained onboard resources. To overcome these limitations, semantic communication has emerged as

Samuel Cohen-Salmon, Itzik Klein

This paper presents the design, development, and experimental validation of MARVEL, an autonomous unmanned surface vehicle built for real-world testing of sensor fusion-based navigation algorithms in GNSS-denied environments. MARVEL was developed under strict constraints of cost-efficiency, portability, and seaworthiness, with the goal of creating a modular,

Rainer Rehak, Christian R. Kuehne

Since SARS-CoV-2 started spreading in Europe in early 2020, there has been a strong call for technical solutions to combat or contain the pandemic, with contact tracing apps at the heart of the debates. The EU's General Data Protection Regulation (GDPR) requires controllers to carry out a data protection impact assessment (DPIA) where their data processing i

M. Mahdian, Z. Mousavi

Detecting and quantifying quantum entanglement remain significant challenges in the noisy intermediate-scale quantum (NISQ) era. This study presents the implementation of quantum support vector machines (QSVMs) on IBM quantum devices to identify and classify entangled states. By employing quantum variational circuits, the proposed framework achieves a runtim

Prim Plansangkate, Lenka Zalabová

We give conserved quantities of two generalizations of conformal circles on the conformal sphere. One generalization concerns curves carrying a distinguished parallel tractor along them, which can be used to construct the conserved quantities. The other is the class of curves satisfying the conformal Mercator equation, the conserved quantities of which are c

Roberta Mota, Ehud Sharlin, Usman Alim

Lens visualization has been a prominent research area in the visualization community, fueled by the continuous need to mitigate visual clutter and occlusion resulting from the increase in data volume. Interactive lenses for spatial data, particularly, challenge designers to conceive design strategies to support the analysis of high-density, multifaceted data

Cong Zhang, Yisheng Yang, Shilong Mu, Chuqiao Lyu

In the pursuit of deeper immersion in human-machine interaction, achieving higher-dimensional tactile input and output on a single interface has become a key research focus. This study introduces the Visual-Electronic Tactile (VET) System, which builds upon vision-based tactile sensors (VBTS) and integrates electrical stimulation feedback to enable bidirecti

Hyunjong Ok, Suho Yoo, Jaeho Lee

Spoken dialogue systems powered by large language models have demonstrated remarkable abilities in understanding human speech and generating appropriate spoken responses. However, these systems struggle with end-turn detection (ETD) -- the ability to distinguish between user turn completion and hesitation. This limitation often leads to premature or delayed

Loc Hoang Tran, Luong Anh Tuan Nguyen

Face recognition is a crucial topic in data science and biometric security, with applications spanning military, finance, and retail industries. This paper explores the implementation of sparse Principal Component Analysis (PCA) using the Proximal Gradient method (also known as ISTA) and the Runge-Kutta numerical methods. To address the face recognition prob

Lami Chan, Xianyong Bai, Hui Tian, Yufei Feng

The spectra of coronal mass ejections (CMEs) in the low corona play a crucial role in understanding their origins and physical mechanism, and enhancing space weather forecasting. However, capturing these spectra faces significant challenges. This paper introduces a scheme of a multi-slit spectrometer design with five slits, acquiring the global spectra of th

Ran Eilat, Zvika Neeman, Eilon Solan

In this note, we propose an alternative definition of strategies and histories in opportunity hunting games, as introduced in Eilat, Neeman and Solan (2025). The advantage of the formulation presented here is that it encompasses a broader class of strategies than the inertial strategies analyzed in Eilat, Neeman and Solan (2025), including those that allow f

Sheng Lu, Mingxi Ge, Jiuyi Zhang, Wanli Zhu

Sequential Recommender Systems (SRS) aim to model sequential behaviors of users to capture their interests which usually evolve over time. Transformer-based SRS have achieved distinguished successes recently. However, studies reveal self-attention mechanism in Transformer-based models is essentially a low-pass filter and ignores high frequency information po

Xuan Kien Phung

Gottschalk's surjunctivity conjecture for a group $G$ states that it is impossible for cellular automata (CA) over the universe $G$ with finite alphabet to produce strict embeddings of the full shift into itself. A group universe $G$ satisfying Gottschalk's surjunctivity conjecture is called a surjunctive group. The surjunctivity theorem of Gromov and Weiss

Yucheng Shi, Wenhao Yu, Jingyuan Huang, Wenlin Yao

Graphical User Interface (GUI) agents extend large language models from text generation to action execution in real-world digital environments. Unlike conversational systems, GUI agents perform irreversible operations such as submitting forms, granting permissions, or deleting data, making trustworthiness a core requirement. This survey identifies the execut

Harman Deep Kaur, Mariagrazia Trapanese, Kirill Zatrimaylov

We present the first--ever example of a macroscopic system in a quantum superposition. The system in question is a Siamese cat known as Lola; however, on a time scale of about 12 hours it oscillates into a different state that we refer to as "Mola". In the "Lola" state, the system is sweet and friendly and allows to cuddle itself, but in the "Mola" state, it

Pompeu Casanovas

The title of this essay, Induced perversity, La perversidad inducida (in Spanish), refers to the side effects of sharing information on social media when it turns out to be false; or when judgments are made about people, events, companies, or facts that can produce effects often leading to consequences that were not initially foreseen. In digital environment

L. A. Mazhorina, N. D. Korolev, N. V. Morozov, E. Yu. Egorova

Measures characterizing the non-Markovianity degree of the quantum dynamics have several drawbacks when applied to real devices. They depend on the chosen measurement time interval and are highly sensitive to experimental noise and errors. We propose several techniques to enhance the practical applicability of the measures and verify our findings experimenta

Youngjun Song, Youngsik Hwang, Jonghun Lee, Heechang Lee

Domain generalization (DG) aims to learn models that perform well on unseen target domains by training on multiple source domains. Sharpness-Aware Minimization (SAM), known for finding flat minima that improve generalization, has therefore been widely adopted in DG. However, our analysis reveals that SAM in DG may converge to \textit{fake flat minima}, where

Weilong Sun, Yumin Zhang, Boren Wei

The Visual-Inertial Simultaneous Localization and Mapping (VI-SLAM) algorithms which are mostly based on static assumption are widely used in fields such as robotics, UAVs, VR, and autonomous driving. To overcome the localization risks caused by dynamic landmarks in most VI-SLAM systems, a robust visual-inertial motion prior SLAM system, named IDY-VINS, is p

Qiang Zhai, Hai-Yao Deng, Xin-Yuan Gao, Leo S. I. Lam

A free surface induces enhanced dynamics in glass formers. We study the dynamical enhancement of glassy films with a distinguishable-particle lattice model of glass free of elastic effects. We demonstrate that the thickness of the surface mobile layer depends on temperature differently under different definitions, although all are based on local structure re

Hido Pinto, Eran Segal

Machine learning has been successfully used in critical domains, such as medicine. However, extracting meaningful insights from biomedical data is often constrained by the lack of their available disease labels. In this research, we demonstrate how machine learning can be leveraged to enhance explainability and uncover biologically meaningful associations, e

Wenhan Liu, Xinyu Ma, Yutao Zhu, Lixin Su

Large Language Models (LLMs) have demonstrated superior listwise ranking performance. However, their superior performance often relies on large-scale parameters (\eg, GPT-4) and a repetitive sliding window process, which introduces significant efficiency challenges. In this paper, we propose \textbf{CoRanking}, a novel collaborative ranking framework that co

Haonan Wang, Xinlei Yi, Yiguang Hong

This paper studies the stochastic distributed nonconvex optimization problem over a network of agents, where agents only access stochastic zeroth-order information about their local cost functions and collaboratively optimize the global objective over bandwidth-limited communication networks. To mitigate communication overhead and handle the unavailability o

Rasim Yılmaz

It has been shown that double field theory (DFT) action can be constructed in a perturbative manner up to cubic order by using double copy of Yang-Mills. This construction can also be extended by starting with higher-derivative Yang-Mills theory to obtain higher-derivative DFT action. In this work, we extend this classical double copy formulation to the Abel

Gil Gekker, Meirav Segal, Dan Lahav, Omer Nevo

Following the rapid increase in Artificial Intelligence (AI) capabilities in recent years, the AI community has voiced concerns regarding possible safety risks. To support decision-making on the safe use and development of AI systems, there is a growing need for high-quality evaluations of dangerous model capabilities. While several attempts to provide such

Kazuki Okamura

We present a construction of graph-directed invariant sets of weak contractions in the sense of Matkowski-Rus on semi-metric spaces. We follow the approach by Bessenyei and P\'enzes, which applies the Kuratowski noncompactness measure without relying on Blascke's completeness theorem. We also establish a relationship between this approach and a generalized d

Xin Zuo, Jiaran Jiang, Jifeng Shen, Wankou Yang

Underwater image understanding is crucial for both submarine navigation and seabed exploration. However, the low illumination in underwater environments degrades the imaging quality, which in turn seriously deteriorates the performance of underwater semantic segmentation, particularly for outlining the object region boundaries. To tackle this issue, we prese

Andrei Cristian Ionescu, Sergey Titov, Maliheh Izadi

Effective onboarding in software engineering is crucial but difficult due to the fast-paced evolution of technologies. Traditional methods, like exploration and workshops, are costly, time-consuming, and quickly outdated in large projects. We propose the Onboarding Buddy system, which leverages large language models, retrieval augmented generation, and an au

Yifan Feng, Hao Hu, Xingliang Hou, Shiquan Liu

Large language models (LLMs) have transformed various sectors, including education, finance, and medicine, by enhancing content generation and decision-making processes. However, their integration into the medical field is cautious due to hallucinations, instances where generated content deviates from factual accuracy, potentially leading to adverse outcomes

Massimo Pappalardo, Nguyen Nang Thieu, Nguyen Dong Yen

Preliminary results of our investigations on solving indefinite qua\-dra\-tic programs by dynamical systems are given. First, dynamical systems corresponding to two fundamental DC programming algorithms to deal with indefinite quadratic programs are considered. Second, the existence and the uniqueness of the global solution of the dynamical system are proved

Riccardo Della Monica, Richard Brito

Rotating black holes can amplify ultralight bosonic fields through superradiance, forming macroscopic clouds known as gravitational atoms. When the cloud forms around one of the components of a binary system, it can undergo a series of distinctive interactions, comprising both secular effects, such as dynamical friction or accretion, and resonant behaviour.

Fenton John Doolan

The spokes observed in Saturn's rings have been a subject of scientific debate since their discovery by Stephen J. O'Meara in the 1970s and their confirmation by the Voyager flybys in the early 1980s (Smith et al., 1982). While the Cassini spacecraft confirmed that the spokes are linked to Saturn's magnetosphere, their exact formation mechanism remains uncer

Xinyu Wang, Xiang Chen, Junping Yang, Ying Cui

In this work, we present a model-independent method to quantify the non-Gaussian fluctuations in the observable distributions, which are assessed by the difference between the measured observable distributions and reconstructed observable distributions via the Bayesian method. Our results indicate that the strength of non-Gaussian fluctuation increases with

Leonardo Pedroso, Pedro Batista, W. P. M. H. Heemels

The transition from large centralized complex control systems to distributed configurations that rely on a network of a very large number of interconnected simpler subsystems is ongoing and inevitable in many applications. It is attributed to the quest for resilience, flexibility, and scalability in a multitude of engineering fields with far-reaching societa

Alexander Murphy, Mohd Sanad Zaki Rizvi, Aden Haussmann, Ping Nie

Large Language Models (LLMs) frequently produce factually inaccurate outputs - a phenomenon known as hallucination - which limits their accuracy in knowledge-intensive NLP tasks. Retrieval-augmented generation and agentic frameworks such as Reasoning and Acting (ReAct) can address this issue by giving the model access to external knowledge. However, LLMs oft

Haitham S. Al-Sinani, Chris J. Mitchell

This paper investigates the presence and impact of questionable, AI-generated academic papers on widely used preprint repositories, with a focus on their role in citation manipulation. Motivated by suspicious patterns observed in publications related to our ongoing research on GenAI-enhanced cybersecurity, we identify clusters of questionable papers and prof

Stanislav Semenov

This paper develops a technical and practical reinterpretation of the real interval [a,b] under the paradigm of fractal countability. Instead of assuming the continuum as a completed uncountable totality, we model [a,b] as a layered structure of constructively definable points, indexed by a hierarchy of formal systems. We reformulate classical notions from r

Rajesh Kumar, Veronica Dexheimer, Johannes Jahan

Neutron stars provide a natural laboratory for studying the properties of dense nuclear matter under extreme conditions. In this proceeding, we review our current understanding of dense isospin symmetric and asymmetric matter and neutron star physics. We focus on modern theoretical, experimental, and observational constraints, including first-principle calcu

Fujia Su, Bingxuan Li, Qingyang Yin, Yanchen Zhang

Robust light transport algorithms, particularly bidirectional path tracing (BDPT), face significant challenges when dealing with specular or highly glossy involved paths. BDPT constructs the full path by connecting sub-paths traced individually from the light source and camera. However, it remains difficult to sample by connecting vertices on specular and gl

Luis Nasi, Gergő Illés, János Hebling, György Tóth

Numerical investigations of multicycle terahertz pulse generation based on wafer stack and tilted pulse front pumped lithium niobate setups concerning the conversion efficiency and the resulting temporal shape and spectrum were performed. Pumping by Fourier-limited-, chirped-, and intensity-modulated pulses, as well as pulse sequences was considered. Wafer s

Zhi Zhang, Meng Gai, Sheng Li

Prior foveated rendering methods often suffer from a limitation where the shading load escalates with increasing display resolution, leading to decreased efficiency, particularly when dealing with retinal-level resolutions. To tackle this challenge, we begin with the essence of the human visual system (HVS) perception and present visual acuity-consistent fov

Ximu Zeng, Liwei Deng, Penghao Chen, Xu Chen

Approximate nearest neighbor search is fundamental in information retrieval. Previous partition-based methods enhance search efficiency by probing partial partitions, yet they face two common issues. In the query phase, a common strategy is to probe partitions based on the distance ranks of a query to partition centroids, which inevitably probes irrelevant p

Saiyam Sakhuja, Shivanshu Siyanwal, Abhishek Tiwari, Britant

Quantum Machine Learning (QML) presents as a revolutionary approach to weather forecasting by using quantum computing to improve predictive modeling capabilities. In this study, we apply QML models, including Quantum Gated Recurrent Units (QGRUs), Quantum Neural Networks (QNNs), Quantum Long Short-Term Memory(QLSTM), Variational Quantum Circuits(VQCs), and Q

Wei Zeng, Xuebin Chang, Jianghao Su, Xiang Gu

Cross-domain generative models based on encoder-decoder AI architectures have attracted much attention in generating realistic images, where domain alignment is crucial for generation accuracy. Domain alignment methods usually deal directly with the initial distribution; however, mismatched or mixed clusters can lead to mode collapse and mixture problems in

Anbang Du, Michael Head, Markus Brede

Interdisciplinary research, a process of knowledge integration, is vital for scientific advancements. It remains unclear whether prestigious journals that are highly impactful lead in disseminating interdisciplinary knowledge. In this paper, by constructing topic-level correlation networks based on publications, we evaluated the interdisciplinarity of more a

I. Cherkaoui, S. Belabssir, J. Horgan, I. Dey

Shor algorithm led to the discovery of multiple vulnerabilities in a number of cryptosystems. As a result, post-quantum cryptography attempts to provide cryptographic solutions that can face these attacks, ensuring the security of sensitive data in a future where quantum computers are assumed to exist. Error correcting codes are a source for efficiency when

Yumou Fei, Dor Minzer, Shuo Wang

In the Max-Cut problem in the streaming model, an algorithm is given the edges of an unknown graph $G = (V,E)$ in some fixed order, and its goal is to approximate the size of the largest cut in $G$. Improving upon an earlier result of Kapralov, Khanna and Sudan, it was shown by Kapralov and Krachun that for all $\varepsilon>0$, no $o(n)$ memory streaming alg

Leonie Neufeld

Using the subordination approach, we provide a new Berry-Esseen-type estimate in the free central limit theorem in terms of the fourth Lyapunov fraction. In the special case of identical distributions, our result implies a rate of order $n^{-1/2 + \varepsilon}$ for any $\varepsilon>0$, thus almost leading to the optimal rate of order $n^{-1/2}$.

Junsu Kim, Yunhoe Ku, Dongyoon Han, Seungryul Baek

Few-shot class-incremental learning (FSCIL) is challenging due to extremely limited training data while requiring models to acquire new knowledge without catastrophic forgetting. Recent works have explored generative models, particularly Stable Diffusion (SD), to address these challenges. However, existing approaches use SD mainly as a replay generator, wher

Sourav Dey, Gonzalo Rivero-Carracedo, Andrei Shumilin, Carlos Gonzalez-Ballestero

Magnonics is an emerging field widely considered as a paradigm shift in information technology that uses spin waves for data storage, processing and transmission. However, the coherent control of spin waves in 2D magnets still remains a challenge. Herein, we investigate the interplay between molecular spins and magnons in hybrid heterostructures formed by [C

Yu Liao, Li Li, Zhesi Shen

Probably Not. Journal Citation Indicator (JCI) was introduced to address the limitations of traditional metrics like the Journal Impact Factor (JIF), particularly its inability to normalize citation impact across different disciplines. This study reveals that JCI faces significant challenges in field normalization for Art & Humanities journals, as evidenced

Feifei Fan

We study the mod $p$ cohomology ring of the classifying space $BPU(p)$ of the projective unitary group $PU(p)$, when $p$ is an odd prime. We prove a mod $p$ formula analogous to a formula of Vistoli for the integral cohomology ring of $BPU(p)$. As an application, we give a simple topological proof of Vistoli's formula.

Leander Girrbach, Stephan Alaniz, Genevieve Smith, Zeynep Akata

With the increasing use of image generation technology, understanding its social biases, including gender bias, is essential. This paper presents a large-scale study on gender bias in text-to-image (T2I) models, focusing on everyday situations. While previous research has examined biases in occupations, we extend this analysis to gender associations in daily

Yuan Chen, Ao Li, Wenhai Li, Lingfeng Deng

B$^+$-trees are prevalent in traditional database systems due to their versatility and balanced structure. While binary search is typically utilized for branch operations, it may lead to inefficient cache utilization in main-memory scenarios. In contrast, trie-based index structures drive branch operations through prefix matching. While these structures gene

Jianhua Zhang, Yansong He, Hao Chen

Koopman operator has been recognized as an ongoing data-driven modeling method for vehicle dynamics which lifts the original state space into a high-dimensional linear state space. The deep neural networks (DNNs) are verified to be useful for the approximation of Koopman operator. To further improve the accuracy of Koopman operator approximation, this paper

Ting Dang, Yan Gao, Hong Jia

Large language models (LLMs) have shown exceptional versatility in natural language processing, prompting recent efforts to extend their multimodal capabilities to speech processing through the development of audio large language models (Audio LLMs). While Audio LLMs excel in tasks such as speech recognition and synthesis, it remains unclear how they perform

Sangyoon Bae, Junbeom Kwon, Shinjae Yoo, Jiook Cha

Understanding how the brain's complex nonlinear dynamics give rise to cognitive function remains a central challenge in neuroscience. While brain functional dynamics exhibits scale-free and multifractal properties across temporal scales, conventional neuroimaging analytics assume linearity and stationarity, failing to capture frequency-specific neural comput

Yadong Xie, Fan Li, Yue Wu, Song Yang

Since the number of cars has grown rapidly in recent years, driving safety draws more and more public attention. Drowsy driving is one of the biggest threatens to driving safety. Therefore, a simple but robust system that can detect drowsy driving with commercial off-the-shelf devices (such as smartphones) is very necessary. With this motivation, we explore

Shay Rabani, Kirill Shvalb, Boris Malomed, Tal Carmon

We introduce a new type of lens that focuses a plane wave into a spherical one, where light comes from all directions. Our method also suggests the design of ideal optical tweezers or, in the reverse direction, photo-detection of nearly all of the light emitted from an omnidirectional source. Our design trades off simple isotropic non-magnetic materials for

Yadong Xie, Fan Li, Yue Wu, Song Yang

Driving safety has drawn much public attention in recent years due to the fast-growing number of cars. Smoking is one of the threats to driving safety but is often ignored by drivers. Existing works on smoking detection either work in contact manner or need additional devices. This motivates us to explore the practicability of using smartphones to detect smo

Song Lai, Zhe Zhao, Fei Zhu, Xi Lin

Continual learning aims to learn multiple tasks sequentially. A key challenge in continual learning is balancing between two objectives: retaining knowledge from old tasks (stability) and adapting to new tasks (plasticity). Experience replay methods, which store and replay past data alongside new data, have become a widely adopted approach to mitigate catast

Hugo de Souza Oliveira, Niloofar Saeedzadeh Khaanghah, Martijn Oetelmans, Niko Münzenrieder

The technological transition from soft machines to soft robots necessarily passes through the integration of soft electronics and sensors. This allows for the establishment of feedback control systems while preserving the softness of the robot embodiment. Multistable mechanical metamaterials are excellent building blocks of soft machines, as their nonlinear

Fanding Huang, Jingyan Jiang, Qinting Jiang, Hebei Li

Recent vision-language models (VLMs) face significant challenges in test-time adaptation to novel domains. While cache-based methods show promise by leveraging historical information, they struggle with both caching unreliable feature-label pairs and indiscriminately using single-class information during querying, significantly compromising adaptation accura

Yadong Xie, Fan Li, Yue Wu, Yu Wang

Fitness can help to strengthen muscles, increase resistance to diseases, and improve body shape. Nowadays, a great number of people choose to exercise at home/office rather than at the gym due to lack of time. However, it is difficult for them to get good fitness effects without professional guidance. Motivated by this, we propose the first personalized fitn

V. I. Yukalov, E. P. Yukalova, D. Sornette

In statistical and nonlinear systems, two qualitatively distinct parameter regions are typically identified: the regular region, characterized by smooth behavior of key quantities, and the critical region, where these quantities exhibit singularities or strong fluctuations. Due to their starkly different properties, these regions are often perceived as being

Dorde Zivanovic

This paper introduces the implementation of the Figaro-GPU algorithm for computing a QR and SVD decomposition over a join matrix defined by the natural join over two tables on GPUs. Figaro-GPU's main novelty is a GPU implementation of the Figaro algorithm \cite{olteanu2022givens, vzivanovic2022linear,olteanu2024givens}: symbolical transformations combined wi

MicroBooNE collaboration, P. Abratenko, D. Andrade Aldana, L. Arellano

Understanding electron neutrino interactions is crucial for measurements of neutrino oscillations and searches for new physics in neutrino experiments. We present the first measurement of the flux-averaged $\nu_e$ + $\bar{\nu}_e$ charged current single charged pion production cross section on argon using the MicroBooNE detector and data from the NuMI neutrin

Xuefeng Li, Haoyang Zou, Pengfei Liu

We introduce ToRL (Tool-Integrated Reinforcement Learning), a framework for training large language models (LLMs) to autonomously use computational tools via reinforcement learning. Unlike supervised fine-tuning, ToRL allows models to explore and discover optimal strategies for tool use. Experiments with Qwen2.5-Math models show significant improvements: ToR

Jiexin Wang, Wenwen Qiang, Zhao Yang, Bing Su

Expressive representation of pose sequences is crucial for accurate motion modeling in human motion prediction (HMP). While recent deep learning-based methods have shown promise in learning motion representations, these methods tend to overlook the varying relevance and dependencies between historical information and future moments, with a stronger correlati

Roman V. Dribas, Andrew S. Golovnev, Nikolay A. Gusev

If $f\colon [0,1]^2 \to \mathbb{R}$ is of class $C^2$ then Sard's theorem implies that $f$ has the following relaxed Sard property: the image under $f$ of the Lebesgue measure restricted to the critical set of $f$ is a singular measure. We show that for $C^{1,\alpha}$ functions with $\alpha<1$ this property is strictly stronger than the weak Sard property in

Haiduo Huang, Yadong Zhang, Yinghui Xu, Pengju Ren

Dynamic convolution enhances model capacity by adaptively combining multiple kernels, yet faces critical trade-offs: prior works either (1) incur significant parameter overhead by scaling kernel numbers linearly, (2) compromise inference speed through complex kernel interactions, or (3) struggle to jointly optimize dynamic attention and static kernels. We ob

Yuqin Yin, Zonghua Shi, Hao Liu, Xiurong Guo

Based on the developed Bethe-Salpeter theory for dealing with unstable state, we investigate unstable meson-meson molecular state in which at least one of the constituents is an unstable meson and provide a reasonable and feasible scheme to deal with this interesting problem in the framework of relativistic quantum field theory. The developed Bethe-Salpeter

Kai Liu, Wei Li, Lai Chen, Shengqiong Wu

This paper introduces JavisDiT, a novel Joint Audio-Video Diffusion Transformer designed for synchronized audio-video generation (JAVG). Based on the powerful Diffusion Transformer (DiT) architecture, JavisDiT simultaneously generates high-quality audio and video content from open-ended user prompts in a unified framework. To ensure audio-video synchronizati

Luyuan Zhang, An Liu, Kexuan Wang

In the forthcoming 6G era, extend reality (XR) has been regarded as an emerging application for ultra-reliable and low latency communications (URLLC) with new traffic characteristics and more stringent requirements. In addition to the quasi-periodical traffic in XR, burst traffic with both large frame size and random arrivals in some real world low latency c