March 2025 arXiv papers — page 147

Pierre Sermanet, Anirudha Majumdar, Vikas Sindhwani

Given the recent rate of progress in artificial intelligence (AI) and robotics, a tantalizing question is emerging: would robots controlled by emerging AI systems be strongly aligned with human values? In this work, we propose a scalable way to probe this question by generating a benchmark spanning the key moments in 824 major pieces of science fiction liter

Beatrice Achilli, Luca Ambrogioni, Carlo Lucibello, Marc Mézard

We generalize the computation of the capacity of exponential Hopfield model from Lucibello and M\'ezard (2024) to more generic pattern ensembles, including binary patterns and patterns generated from a hidden manifold model.

Ignacio F. Graña, Savvas Varsamopoulos, Tatsuhito Ando, Hiroyuki Maeshima

Materials discovery is a computationally intensive process that requires exploring vast chemical spaces to identify promising candidates with desirable properties. In this work, we propose using quantum-enhanced machine learning algorithms following the extremal learning framework to predict novel heteroacene structures with low hole reorganization energy $\

Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon

Efficiently acquiring external knowledge and up-to-date information is essential for effective reasoning and text generation in large language models (LLMs). Prompting advanced LLMs with reasoning capabilities to use search engines during inference is often suboptimal, as the LLM might not fully possess the capability on how to interact optimally with the se

Aykut İşleyen, René van de Molengraft, Ömür Arslan

Safe autonomous exploration of unknown environments is an essential skill for mobile robots to effectively and adaptively perform environmental mapping for diverse critical tasks. Due to its simplicity, most existing exploration methods rely on the standard frontier-based exploration strategy, which directs a robot to the boundary between the known safe and

Bin Hu, Chenqiang Gao, Shurui Liu, Junjie Guo

Image translation is one of the crucial approaches for mitigating information deficiencies in the infrared and visible modalities, while also facilitating the enhancement of modality-specific datasets. However, existing methods for infrared and visible image translation either achieve unidirectional modality translation or rely on cycle consistency for bidir

Md Morshed Alam, Lokesh Chandra Das, Sandip Roy, Sachin Shetty

Internet of Things (IoT) platforms with trigger-action capability allow event conditions to trigger actions in IoT devices autonomously by creating a chain of interactions. Adversaries exploit this chain of interactions to maliciously inject fake event conditions into IoT hubs, triggering unauthorized actions on target IoT devices to implement remote injecti

Yongsheng Lian

Inspired by the success of DeepSeek R1 in reasoning via reinforcement learning without human feedback, we train a 3B language model using the Countdown Game with pure reinforcement learning. Our model outperforms baselines on four of five benchmarks, demonstrating improved generalization beyond its training data. Notably, response length does not correlate w

Hannah VanderHoeven, Brady Bhalla, Ibrahim Khebour, Austin Youngren

We present TRACE, a novel system for live *common ground* tracking in situated collaborative tasks. With a focus on fast, real-time performance, TRACE tracks the speech, actions, gestures, and visual attention of participants, uses these multimodal inputs to determine the set of task-relevant propositions that have been raised as the dialogue progresses, and

Rosalia Tufano, Gabriele Bavota

Code Review consists in assessing the code written by teammates with the goal of increasing code quality. Empirical studies documented the benefits brought by such a practice that, however, has its cost to pay in terms of developers' time. For this reason, researchers have proposed techniques and tools to automate code review tasks such as the reviewers sele

Juncan Deng, Shuaiting Li, Zeyu Wang, Kedong Xu

Visual Mamba networks (ViMs) extend the selective state space model (Mamba) to various vision tasks and demonstrate significant potential. As a promising compression technique, vector quantization (VQ) decomposes network weights into codebooks and assignments, significantly reducing memory usage and computational latency, thereby enabling the deployment of V

Pan Xu

The online randomized primal-dual method has widespread applications in online algorithm design and analysis. A key challenge is identifying an appropriate function space, $F$, in which we search for an optimal updating function $f \in F$ that yields the best possible lower bound on the competitiveness of a given algorithm. The choice of $F$ must balance two

Josef Janák, Enrico Priola

We deal with a class of semilinear SPDEs driven by space-time white noise that includes the one dimensional stochastic Burgers equation. Such equations can have nonlocal and quadratic nonlinearities. We consider the problem of estimation of the diffusivity parameter in front of the second-order spatial derivative. Based on local observations in space, we stu

Philipp Schwenke, Ephraim Spindler, Vitaliy I. Vasyuchka, Alexandre Abbass Hamadeh

The manipulation of skyrmions by surface acoustic waves (SAW) has garnered significant interest in the field of spintronic devices. Previous studies established that skyrmions can be generated and moved by strain pulses. In this study, we propose that sawtooth-SAWs can be used to drive a ratchet motion of magnetic skyrmions in the presence of pinning centers

Maxim N. Cherkashin, Jan Laufer, Thomas Kirchner

Photoacoustic (PA) imaging of deep tissue tends to employ Q-switched lasers with high pulse energy to generate high optical fluence and therefore high PA signal. Compared to Q-switched lasers, pulsed laser diodes (PLDs) typically generate low pulse energy. In PA imaging applications with strong acoustic attenuation, such as through human skull bone, the broa

Bakary Badjie, José Cecílio, António Casimiro

The Mixture-of-Experts (MoE) model has succeeded in deep learning (DL). However, its complex architecture and advantages over dense models in image classification remain unclear. In previous studies, MoE performance has often been affected by noise and outliers in the input space. Some approaches incorporate input clustering for training MoE models, but most

Spencer Leslie, Jingwei Xiao, Wei Zhang

In a companion paper, we formulated a global conjecture for the automorphic period integral associated to the symmetric pairs defined by unitary groups over number fields, generalizing a theorem of Waldspurger's toric period for $\mathrm{GL}(2)$. In this paper, we introduce a new relative trace formula to prove our global conjecture under some local hypothes

Lennart Maximilian Seifert, Connor T. Hann, Kyungjoo Noh

We theoretically study the influence of frequency uncertainties on the operation of a Kerr-cat qubit. As the mean photon number increases, Kerr-cat qubits provide an increasing level of protection against phase errors induced by unknown frequency shifts during idling and X rotations. However, realizing rotations about the other principal axes (e.g., Y and Z

Juan Carlos López Vieyra, Alexander V Turbiner

An infinite 3-parametric family of superintegrable and exactly-solvable quantum models on a plane, admitting separation of variables in polar coordinates, marked by integer index $k$ was introduced in Journ Phys A 42 (2009) 242001 and was called in literature the TTW system. In this paper it is conjectured that the Hamiltonian and both integrals of TTW syste

Ziyu Wan, Yunxiang Li, Xiaoyu Wen, Yan Song

Recent research on Reasoning of Large Language Models (LLMs) has sought to further enhance their performance by integrating meta-thinking -- enabling models to monitor, evaluate, and control their reasoning processes for more adaptive and effective problem-solving. However, current single-agent work lacks a specialized design for acquiring meta-thinking, res

Spencer Leslie, Jingwei Xiao, Wei Zhang

We formulate a global conjecture for the automorphic period integral associated to the symmetric pairs defined by unitary groups over number fields, generalizing a theorem of Waldspurger's toric period for $\mathrm{GL}(2)$. We introduce a new relative trace formula to prove our global conjecture under some local hypotheses. A new feature is the presence of t

Zhe Xu, Daoyuan Chen, Zhenqing Ling, Yaliang Li

Large foundation models face challenges in acquiring transferable, structured thinking abilities, especially when supervised with rigid templates or crowd-annotated instruction datasets. Unlike prior approaches, we focus on a thinking-centric data synthesis paradigm that enables models to evolve through self-generated, cognitively guided data. We propose Min

Nazanin Moradinasab, Saurav Sengupta, Jiebei Liu, Sana Syed

Healthcare relies on multiple types of data, such as medical images, genetic information, and clinical records, to improve diagnosis and treatment. However, missing data is a common challenge due to privacy restrictions, cost, and technical issues, making many existing multi-modal models unreliable. To address this, we propose a new multi-model model called

Reza Mirzaeifard, Ashkan Moradi, Masahiro Yukawa, Stefan Werner

This paper addresses the challenge of localization in federated settings, which are characterized by distributed data, non-convexity, and non-smoothness. To tackle the scalability and outlier issues inherent in such environments, we propose a robust algorithm that employs an $\ell_1$-norm formulation within a novel federated ADMM framework. This approach add

Junjie Zhou, Jiao Tang, Yingli Zuo, Peng Wan

The integrative analysis of histopathological images and genomic data has received increasing attention for survival prediction of human cancers. However, the existing studies always hold the assumption that full modalities are available. As a matter of fact, the cost for collecting genomic data is high, which sometimes makes genomic data unavailable in test

M. Kalliokoski, G. Levi, A. Maulik, I. Ostrovskiy

The calibration of the CR39 and Makrofol Nuclear Track Detectors of the MoEDAL experiment at the CERN-LHC was performed by exposing stacks of detector foils to heavy ion beams with energies ranging from 340 MeV/nucleon to 150 GeV/nucleon. After chemical etching, the base areas and lengths of etch-pit cones were measured using automatic and manual optical mic

Qi Xu, Annie Qu

In the era of big data, large-scale, multi-source, multi-modality datasets are increasingly ubiquitous, offering unprecedented opportunities for predictive modeling and scientific discovery. However, these datasets often exhibit complex heterogeneity, such as covariates shift, posterior drift, and blockwise missingness, which worsen predictive performance of

Andy Buschmann

This paper introduces a simple JavaScript-based web application designed to assist educators in detecting AI-generated content in student essays and written assignments. Unlike existing AI detection tools that rely on obfuscated machine learning models, AIDetection.info employs a heuristic-based approach to identify common syntactic traces left by generative

Romain Thoreau, Valerio Marsocci, Dawa Derksen

As large-scale heterogeneous data sets become increasingly available, adapting foundation models at low cost has become a key issue. Seminal works in natural language processing, e.g. Low-Rank Adaptation (LoRA), leverage the low "intrinsic rank" of parameter updates during adaptation. In this paper, we argue that incorporating stronger inductive biases in bo

Yunli Wang, Zhen Zhang, Zhiqiang Wang, Zixuan Yang

Cascade Ranking is a prevalent architecture in large-scale top-k selection systems like recommendation and advertising platforms. Traditional training methods focus on single-stage optimization, neglecting interactions between stages. Recent advances have introduced interaction-aware training paradigms, but still struggle to 1) align training objectives with

Junjie Zhou, Shouju Wang, Yuxia Tang, Qi Zhu

The prediction of nanoparticles (NPs) distribution is crucial for the diagnosis and treatment of tumors. Recent studies indicate that the heterogeneity of tumor microenvironment (TME) highly affects the distribution of NPs across tumors. Hence, it has become a research hotspot to generate the NPs distribution by the aid of multi-modal TME components. However

Hongyu Lin, Yuchen Li, Haoran Luo, Kaichun Yao

Operating system (OS) kernel tuning is a critical yet challenging problem for performance optimization, due to the large configuration space, complex interdependencies among configuration options, and the rapid evolution of kernel versions. Recent work has explored large language models (LLMs) for automated kernel tuning, but existing approaches often suffer

Haoming Shen, Yang Zeng, Baoyu Zhou

A sequential quadratic programming method is designed for solving general smooth nonlinear stochastic optimization problems subject to expectation equality constraints. We consider the setting where the objective and constraint function values, as well as their derivatives, are not directly available. The algorithm applies an adaptive step size policy and on

Tianyu Fang, Mengyuan Ma, Markku Juntti, Nir Shlezinger

Integrated sensing and communications (ISAC) has emerged as a promising paradigm to unify wireless communications and radar sensing, enabling efficient spectrum and hardware utilization. A core challenge with realizing the gains of ISAC stems from the unique challenges of dual purpose beamforming design due to the highly non-convex nature of key performance

Haoyuan Gao, Zicong Zhang, Yuqi Wei, Linglan Zhao

Vision-Language Models (VLMs) represent a significant breakthrough in artificial intelligence by integrating visual and textual modalities to achieve impressive zero-shot capabilities. However, VLMs are susceptible to catastrophic forgetting when sequentially fine-tuned on multiple downstream tasks. Existing continual learning methods for VLMs face various l

Oliver Lindström

We construct a model for the (non-unital) S^1-framed little 2d-dimensional disks operad for any positive integer d using logarithmic geometry. We also show that the unframed little 2d-dimensional disks operad has a model which can be constructed using log schemes with virtual morphisms.

Beier Zhu, Jiequan Cui, Hanwang Zhang, Chi Zhang

While image-text foundation models have succeeded across diverse downstream tasks, they still face challenges in the presence of spurious correlations between the input and label. To address this issue, we propose a simple three-step approach,Project-Probe-Aggregate (PPA), that enables parameter-efficient fine-tuning for foundation models without relying on

Daniel Spivak

The directed landscape, the central object in the Kardar-Parisi-Zhang universality class, is shown to be the scaling limit of various models by Dauvergne and Vir\'ag (2022) and Dauvergne, Ortmann and Vir\'ag (2018). In his study of geodesics in upper tail deviations of the directed landscape, Liu (2022) put forward a conjecture about the rate of the lowest r

Kadir Özçoban, Murat Manguoğlu, Emrullah Fatih Yetkin

The real-life data have a complex and non-linear structure due to their nature. These non-linearities and the large number of features can usually cause problems such as the empty-space phenomenon and the well-known curse of dimensionality. Finding the nearly optimal representation of the dataset in a lower-dimensional space (i.e. dimensionality reduction) o

Aasish Kumar Sharma, Michael Bidollahkhani, Julian Martin Kunkel

The rapid adoption of AI-driven automation in IoT environments, particularly in smart cities and industrial systems, necessitates a standardized approach to quantify AIs computational workload. Existing methodologies lack a consistent framework for measuring AI computational effort across diverse architectures, posing challenges in fair taxation models and e

Ethan Y. H. Li

In 2020, Dahlberg, She, and van Willigenburg conjectured that the chromatic symmetric function of any tree with maximum degree at least 4 is not e-positive. Zheng and Tom verified this conjecture for all trees with maximum degree at least 5 and spiders with maximum degree 4, and in their proofs the following necessary condition given by Wolfgang plays an imp

Andreas Kofler, Luca Calatroni, Christoph Kolbitsch, Kostas Papafitsoros

We propose an unrolled algorithm approach for learning spatially adaptive parameter maps in the framework of convolutional synthesis-based $\ell_1$ regularization. More precisely, we consider a family of pre-trained convolutional filters and estimate deeply parametrized spatially varying parameters applied to the sparse feature maps by means of unrolling a F

Luke A. F. Brunswick, Luke Hallacy, René Dost, Edmund Clarke

On-chip micro-cavities with embedded quantum emitters provide an excellent platform for high-performance quantum technologies. A major difficulty for such devices is overcoming the detrimental effects of fluctuations in the device dimensions caused by the limitations of the fabrication processes. We present a fully tunable system based on a 1D photonic-cryst

Alice Suozzi, Luca Capone, Gianluca E. Lebani, Alessandro Lenci

This paper presents BAMBI (BAby language Models Boostrapped for Italian), a series of Baby Language Models (BabyLMs) trained on data that mimic the linguistic input received by a five-year-old Italian-speaking child. The BAMBI models are tested using a benchmark specifically designed to evaluate language models, which takes into account the amount of trainin

Xiang Zhou, Zhen-Peng Xu, Liang-Liang Sun, Chunfeng Wu

Quantum networks with bipartite resources and shared randomness present the simplest infrastructure for implementing a future quantum internet. Here, we shall investigate which kinds of entanglement can or cannot be generated from this kind of quantum network by examining their fidelity with different graph states. On the one hand, based on a standard form o

M. Uehlein, H. T. Snowden, C. Seibel, T. Held

The simulation of non-equilibrium electron distributions is essential for capturing light-metal interactions and therefore the study of photoabsorption, photocatalysis, laser ablation, and many other phenomena. Current methodologies, such as the Boltzmann equation using full collision integrals, describe non-equilibrium electron dynamics in great detail but

Jing Wang, Fengzhuo Zhang, Xiaoli Li, Vincent Y. F. Tan

Auto-Regressive Video Diffusion Models (AR-VDMs) have shown strong capabilities in generating long, photorealistic videos, but suffer from two key limitations: (i) history forgetting, where the model loses track of previously generated content, and (ii) temporal degradation, where frame quality deteriorates over time. Yet a rigorous theoretical analysis of t

Xiangmin Jiao, Hongji Gao

Measuring how quickly iterative methods converge is essential in computational mathematics, but current approaches have significant limitations. Q-order analysis requires strict smoothness conditions, while R-order analysis lacks precision and creates ambiguity, especially when analyzing convergence rates close to linear. We introduce P-order, a new framewor

Noel Naughton, Arman Tekinalp, Keshav Shivam, Seung Hung Kim

A long-standing engineering problem, the control of soft robots is difficult because of their highly non-linear, heterogeneous, anisotropic, and distributed nature. Here, bridging engineering and biology, a neural reservoir is employed for the dynamic control of a bio-hybrid model arm made of multiple muscle-tendon groups enveloping an elastic spine. We show

Zanyang Kong

In this paper,we propose a Multi-Objective Sequential Quadratic Programming (MOSQP) algorithm for constrained multi-objective optimization problems,basd on a low-order smooth penalty function as the merit function for line search. The algorithm constructs single-objective optimization subproblems based on each objective function, solves quadratic programming

Alexander Von Moll, Dejan Milutinović, Isaac Weintraub, David W. Casbeer

In this paper we address the problem of 'weaponeering', i.e., placing the weapon engagement zone (WEZ) of a vehicle on a moving target, while simultaneously avoiding the target's WEZ. A WEZ describes the lethality region of a range-limited weapon considering both the range of the weapon along with the state of the target. The weapons are assumed to have simp

Jiayuan Huang, Runlong He, Danyal Zaman Khan, Evangelos B. Mazomenos

Image-guided surgery demands adaptive, real-time decision support, yet static AI models struggle with structured task planning and providing interactive guidance. Large language models (LLMs)-powered agents offer a promising solution by enabling dynamic task planning and predictive decision support. Despite recent advances, the absence of surgical agent data

Justus Neumann, Tulja Varun Kondra, Kiara Hansenne, Lisa T. Weinbrenner

Quantum networks connect systems at separate locations via quantum links, enabling a wide range of quantum information tasks between distant parties. Large-scale networks have the potential to enable global secure communication, distributed quantum computation, enhanced clock synchronization, and high-precision multiparameter metrology. For the optimal devel

Yang Liu

This paper conducts research on the established model and presents the main conclusions . Firstly, by separately considering the infectivity of each of the two infectious diseases and the infectivity of the population simultaneously infected with the two infectious diseases, the existence of three types of boundary equilibrium points is determined, as well a

Ivan Atamas, Sergey Dashkovskiy, Vitalii Slynko

In this work we consider linear non-autonomous systems of Wazewski type on Hilbert spaces and provide a new approach to study their stability properties by means of a decomposition into subsystems and conditions implied on the interconnection properties. These conditions are of the small-gain type but the appoach is based on a conceptually new notion which w

Immo C. Reis

Radiative corrections are playing an increasingly significant role in low-energy nuclear physics. We investigate the influence of photons as explicit degrees of freedom within nuclear EFT on the deuteron binding energy, charge form factor, and the radiative capture process $ np \to d\gamma $ using the velocity renormalization group. In each case, evolving th

Alireza Entezari, Arunava Banerjee

Current performance bounds for randomized iterative methods are often considered tight under per-iteration analyses, yet they are notoriously loose in practice. We derive asymptotic performance bounds that narrow this theory-practice gap, leveraging a new technique for bounding the spectral radii of operators arising in randomized iterations and a connection

Ce Jin, Yael Kirkpatrick, Virginia Vassilevska Williams, Nicole Wein

We consider the classical $k$-Center problem in undirected graphs. The problem is known to have a polynomial-time 2-approximation. There are even $(2+\varepsilon)$-approximations running in near-linear time. The conventional wisdom is that the problem is closed, as $(2-\varepsilon)$-approximation is NP-hard when $k$ is part of the input, and for constant $k\

Michael Lönne, Matteo Penegini

In this work, we establish two main results in the context of arithmetic and geometric properties of plane curves. First, we construct numerous new examples of arithmetic Zariski pairs and multiplets, where only a few ones were previously available. Second, we describe a faithful action of the absolute Galois group on the equisingular strata of plane curves,

K. Larsen, R. Lukyanenko, Roland M. Mueller, V. Storey

Researchers must ensure that the claims about the knowledge produced by their work are valid. However, validity is neither well-understood nor consistently established in design science, which involves the development and evaluation of artifacts (models, methods, instantiations, and theories) to solve problems. As a result, it is challenging to demonstrate a

Jean-Marcel Rax

The impact of earth's gravity on neutral mesons dynamics is analyzed. The main effect of a Newtonian potential is to couple the strangeness and bottomness oscillations with the quark zitterbewegung oscillations. This coupling is responsible for the observed CP violation in the three types of experiments analyzed here: (i) indirect violation in the mixing, (i

Máté Tóth, Péter Kovács, Réka Bencses, Zoltán Bendefy

Neural reconstruction models for autonomous driving simulation have made significant strides in recent years, with dynamic models becoming increasingly prevalent. However, these models are typically limited to handling in-domain objects closely following their original trajectories. We introduce a hybrid approach that combines the strengths of neural reconst

Shitong Shao, Zikai Zhou, Dian Xie, Yuetong Fang

Making text-to-image (T2I) generative model sample both fast and well represents a promising research direction. Previous studies have typically focused on either enhancing the visual quality of synthesized images at the expense of sampling efficiency or dramatically accelerating sampling without improving the base model's generative capacity. Moreover, near

Benedict Short, Ian McInerney, John Wickerson

Co-developing scientific algorithms and hardware accelerators requires domain-specific knowledge and large engineering resources. This leads to a slow development pace and high project complexity, which creates a barrier to entry that is too high for the majority of developers to overcome. We are developing a reusable end-to-end compiler toolchain for the Ju

Meng Zhou, Paul La Plante, Adam Lidz, Yi Mao

The 21\,cm line and the patchy kinetic Sunyaev-Zel'dovich (kSZ) effect are promising and complementary probes of the Epoch of Reionization (EoR). A challenge for cross-correlating these two signals is that foreground avoidance or removal algorithms applied to the 21\,cm data inevitably sacrifice Fourier modes with long wavelengths along the line-of-sight (i.

Florian Rein, Laure Carénini, Florian Fichot, Benjamin Favier

We investigate convection in a thin cylindrical gas layer with an imposed flux at the bottom and a fixed temperature along the side, using a combination of direct numerical simulations and laboratory experiments. The experimental approach allows us to extend by two orders of magnitude the explored range in terms of flux Rayleigh number. We identify a scaling

Nadia Bihari Padhan, Rahul Pandit

The Cahn-Hilliard-Navier-Stokes (CHNS) partial differential equations (PDEs) provide a powerful framework for the study of the statistical mechanics and fluid dynamics of multiphase fluids. We provide an introduction to the equilibrium and nonequilibrium statistical mechanics of systems in which coexisting phases, distinguished from each other by scalar orde

Julia Freund, Francesco Basso Basset, Tobias M. Krieger, Alessandro Laneve

Measurements are central in all quantitative sciences, and a fundamental challenge is to make observations without systematic measurement errors. This holds in particular for quantum information processing, where other error sources, such as noise and decoherence, are unavoidable. Consequently, methods for detecting systematic errors have been developed, but

Viktor Harangi

A $k$-star decomposition of a graph is a partition of its edges into $k$-stars (i.e., $k$ edges with a common vertex). The paper studies the following problem: for what values of $k>d/2$ does the random $d$-regular graph have a $k$-star decomposition (asymptotically almost surely, provided that the number of edges is divisible by $k$)? Delcourt, Greenhill, I

Johnson Loh, Lyubov Dudchenko, Justus Viga, Tobias Gemmeke

Deep neural network (DNN) models have shown remarkable success in many real-world scenarios, such as object detection and classification. Unfortunately, these models are not yet widely adopted in health monitoring due to exceptionally high requirements for model robustness and deployment in highly resource-constrained devices. In particular, the acquisition

Athokpam Langlen Chanu, Youngjai Park, Jaesung Choi, Younghwa Cha

We analyze electroencephalography (EEG) signals using the ordinal pattern framework to investigate whether different human brain states can be distinguished based on the disorder of EEG dynamics. Rather than analyzing raw EEG signals, we focus on the principal mode of EEG phase dynamics, reflecting anterior-posterior information flow, and quantify disorder u

I. Colecchio, E. Le Gall, B. Noetinger

The effective conductivity ($T^{eff}$) of 2D and 3D Random Resistor Networks (RRNs) with random edge conductivity are studied. The combined influence of geometrical disorder, which controls the overall connectivity of the medium, and leads to percolation effects, and conductivity randomness is investigated. A formula incorporating connectivity aspects and se

Francesco Ballerin, Nello Blaser, Erlend Grong

Analyzing scalar and vector fields on the sphere, such as temperature or wind speed and direction on Earth, is a difficult task. Models should respect both the rotational symmetries of the sphere and the inherent symmetries of the vector fields. A class of equivariant models has emerged, which process these spherical signals by applying group convolutions in

Morimichi Kawasaki, Mitsuaki Kimura, Shuhei Maruyama, Takahiro Matsushita

Shelukhin constructed a quasimorphism on the universal covering of the group of Hamiltonian diffeomorphisms for a general closed symplectic manifold. In the present paper, we prove the non-extendability of that quasimorphism for certain symplectic manifolds, such as a blow-up of torus and the product of a surface of genus at least two and a closed symplectic

Malik Marmonier, Rachel Bawden, Benoît Sagot

This study explores an LLM's ability to learn new languages using explanations found in a grammar book, a process we term "explicit learning." To rigorously assess this ability, we design controlled translation experiments between English and constructed languages generated, through specific cryptographic means, from Latin or French. Contrary to previous stu

Xiangjian Jiang, Nikola Simidjievski, Mateja Jamnik

Heterogeneous tabular data poses unique challenges in generative modelling due to its fundamentally different underlying data structure compared to homogeneous modalities, such as images and text. Although previous research has sought to adapt the successes of generative modelling in homogeneous modalities to the tabular domain, defining an effective generat

Sreemayee Aditya

This paper explores the effect of strong-to-weak fragmentation transition, namely freezing transition, and its rich characteristics in a family of one-dimensional spinless fermionic models involving short-to-long-range facilitated hoppings with an East constraint. Focusing on this family of models with range-$q$ terms, our investigation furnishes an exhausti

Rufeng Liu, Claudia Wehrhahn, Andrés F. Barrientos, Alejandro Jara

We propose a Bayesian nonparametric model for mixed-type bounded data, where some variables are compositional and others are interval-bounded. Compositional variables are non-negative and sum to a given constant, such as the proportion of time an individual spends on different activities during the day or the fraction of different types of nutrients in a per

Klervie Toczé, Simin Nadjm-Tehrani

Microservices are a way of splitting the logic of an application into small blocks that can be run on different computing units and used by other applications. It has been successful for cloud applications and is now increasingly used for edge applications. This new architecture brings many benefits but it makes deciding where a given service request should

Viktor Nevelius Wernholm, Alfred Wärnsäter, Axel Ringh

In multiple target tracking, it is important to be able to evaluate the performance of different tracking algorithms. The trajectory generalized optimal sub-pattern assignment metric (TGOSPA) is a recently proposed metric for such evaluations. The TGOSPA metric is computed as the solution to an optimization problem, but for large tracking scenarios, solving

Xing Wei, Shengqian Han, Chenyang Yang, Chengjian Sun

Proactive virtual reality (VR) streaming requires users to upload viewpoint-related information, raising significant privacy concerns. Existing strategies preserve privacy by introducing errors to viewpoints, which, however, compromises the quality of experience (QoE) of users. In this paper, we first delve into the analysis of the viewpoint leakage probabil

Saimouli Katragadda, Cho-Ying Wu, Yuliang Guo, Xinyu Huang

To enable AI agents to interact seamlessly with both humans and 3D environments, they must not only perceive the 3D world accurately but also align human language with 3D spatial representations. While prior work has made significant progress by integrating language features into geometrically detailed 3D scene representations using 3D Gaussian Splatting (GS

Zhihua Tian, Sirun Nan, Ming Xu, Shengfang Zhai

Text-to-image (T2I) diffusion models have achieved remarkable progress in generating high-quality images but also raise people's concerns about generating harmful or misleading content. While extensive approaches have been proposed to erase unwanted concepts without requiring retraining from scratch, they inadvertently degrade performance on normal generatio

Xiaoda Yang, JunYu Lu, Hongshun Qiu, Sijing Li

Vision-Language Models (VLMs) based on Mixture-of-Experts (MoE) architectures have emerged as a pivotal paradigm in multimodal understanding, offering a powerful framework for integrating visual and linguistic information. However, the increasing complexity and diversity of tasks present significant challenges in coordinating load balancing across heterogene

Irina Ya. Aref'eva, Ali Hajilou, Pavel Slepov, Marina Usova

In a previous paper [arXiv:2402.14512v3], we investigated the dependence of the running coupling constant on temperature and chemical potential for holographic models of the light and heavy quarks, supported by an Einstein-dilaton-Maxwell action. In this paper, we study the dependence of the corresponding $\beta$-functions on the temperature and the chemical

Julian Spravil, Sebastian Houben, Sven Behnke

Cross-lingual, cross-task transfer is challenged by task-specific data scarcity, which becomes more severe as language support grows and is further amplified in vision-language models (VLMs). We investigate multilingual generalization in encoder-decoder transformer VLMs to enable zero-shot image captioning in languages encountered only in the translation tas

Yunfeng Zhang

We prove new local well-posedness results for nonlinear Schr\"odinger equations posed on a general product of spheres and tori, by the standard approach of multi-linear Strichartz estimates. To prove these estimates, we establish and utilize multi-linear bounds for the joint spectral projector associated to the Laplace--Beltrami operators on the individual s

Eckart Cobo-Briesewitz, Khaled Wahba, Wolfgang Hönig

The increasing complexity of multirotor applications demands flight controllers that can accurately account for all forces acting on the vehicle. Conventional controllers model most aerodynamic and dynamic effects but often neglect higher-order forces, as their accurate estimation is computationally expensive. Incremental Nonlinear Dynamic Inversion (INDI) o

Therese Biedl

In his Ph.D. thesis, Farber proved that every strongly chordal graph can be represented as intersection graph of subtrees of a weighted tree, and these subtrees are ``compatible''. Moreover, this is an equivalent characterization of strongly chordal graphs. To my knowledge, Farber never published his results in a conference or a journal, and the thesis is no

Sowad Rahman

The detection of cardiac abnormalities using electrocardiogram (ECG) signals is crucial for early diagnosis and intervention in cardiovascular diseases. Traditional deep learning models often lack adaptability to varying signal patterns. This study introduces the Neuro-Informed Adaptive Learning (NIAL) algorithm, a hybrid approach integrating convolutional n

Yuki Osada, Alessio Pomponio

In this paper we deal with the following weakly coupled nonlinear Schr\"{o}dinger system \begin{align*} \begin{cases} - \Delta_\alpha u + \omega u = |u|^2 u + \beta u |v|^2&\quad \mathrm{in}\ \mathbb{R}^2,\\ - \Delta v + \tilde{\omega} v = |v|^2 v + \beta |u|^2 v&\quad \mathrm{in}\ \mathbb{R}^2, \end{cases} %\tag{$\mathcal{P}_\beta$} \end{align*} where $-\De

Zhizhen Chen, Zi Cai

We propose a periodically driven system whose dimensionality is an emergent property that can be tunable, thus enables us to realize not only many-body phases with arbitrary dimensions, but also phase transitions, instead of crossovers, between phases with various dimensions. We study an interacting rotor model whose instantaneous Hamiltonian keeps the one-d

Krzysztof Adamkiewicz, Paweł W. Woźniak, Julia Dominiak, Andrzej Romanowski

Recent technological advances popularized the use of image generation among the general public. Crafting effective prompts can, however, be difficult for novice users. To tackle this challenge, we developed PromptMap, a new interaction style for text-to-image AI that allows users to freely explore a vast collection of synthetic prompts through a map-like vie

Pierluigi Francesco De Paola, Alessandro Borri, Alessia Paglialonga, Pasquale Palumbo

The role played by physical activity in slowing down the progression of type-2 diabetes is well recognized. However, except for general clinical guidelines, quantitative real-time estimates of the recommended amount of physical activity, based on the evolving individual conditions, are {still missing} in the literature. The aim of this work is to provide a c

Marta Ghirardelli, Brynjulf Owren, Elena Celledoni

We derive nonlinear stability results for numerical integrators on Riemannian manifolds, by imposing conditions on the ODE vector field and the step size that makes the numerical solution non-expansive whenever the exact solution is non-expansive over the same time step. Our model case is a geodesic version of the explicit Euler method. Precise bounds are ob

Richard A. Dubniczky, Krisztofer Zoltán Horvát, Tamás Bisztray, Mohamed Amine Ferrag

Identifying vulnerabilities in source code is crucial, especially in critical software components. Existing methods such as static analysis, dynamic analysis, formal verification, and recently Large Language Models are widely used to detect security flaws. This paper introduces CASTLE (CWE Automated Security Testing and Low-Level Evaluation), a benchmarking

Fei Hu, Tuyen Trung Truong, Junyi Xie

Let $X$ be a smooth projective variety of dimension $n$ over the algebraic closure of a finite field $\mathbb{F}_p$. Assuming the standard conjecture $D$, we prove a weaker form of the Dynamical Degree Comparison conjecture; equivalence of semisimplicity of Frobenius endomorphism and of any polarized endomorphism (a more general result, in terms of the bigge

Honglin Chen, Wei Xie, Yingqi Yu, Hao Fu

With the rapid advancement of quantum information technology, designing efficient distributed quantum algorithms to perform various information processing tasks remains challenging. In this paper, we consider a distributed scenario where two parties, Alice and Bob, given access to matrices $A$ and $B$ respectively, aim to estimate ${\rm Tr}(f(A)g(B))$, where

Chieh-Chun Chen, Chia-Yu Chang, Navid Nikaein

Mobile networks evolve on a regular basis to meet the requirements of a rapidly changing application ecosystem; hence, a future-proof design is key to getting the most out of their lifecycle. In comparison to other access networks, one major issue with the 5G Radio Access Network (RAN) is that it behaves as a "fat Layer 2" entity, resulting in disparities in

Chenxi Zhao, Haiyang Yu, Zixin Zhang, Qinghua Lei

The technology of lined rock cavern (LRC) with great geographical flexibility is a promising, cost-effective solution to underground hydrogen storage. However, the air-tight steel tanks used in this technology are susceptible to material degradation due to hydrogen embrittlement (HE), potentially leading to leakage and structural failure, especial for LRCs c