March 2025 arXiv papers — page 185

Nicolo' Fontana, Francesco Corso, Enrico Zuccolotto, Francesco Pierri

The increasing prevalence of online misinformation has heightened the demand for automated fact-checking solutions. Large Language Models (LLMs) have emerged as potential tools for assisting in this task, but their effectiveness remains uncertain. This study evaluates the fact-checking capabilities of various open-source LLMs, focusing on their ability to as

Alexander Pei

Neural encoding of objects and cognitive states remains an elusive yet crucial aspect of brain function. While traditional feed-forward machine learning neural networks have enormous potential to encode information, modern architectures provide little insight into the brain's mechanisms. In this work, a Jansen and Rit neural mass model was constructed to enc

Julie Alhosh, Harley Wiltzer, David Meger

In reinforcement learning (RL), the long-term behavior of decision-making policies is evaluated based on their average returns. Distributional RL has emerged, presenting techniques for learning return distributions, which provide additional statistics for evaluating policies, incorporating risk-sensitive considerations. When the passage of time cannot natura

Marthe Bonamy, Mónika Csikós, Anna Gujgiczer, Yelena Yuditsky

The dominating number $\gamma(G)$ of a graph $G$ is the minimum size of a vertex set whose closed neighborhood covers all the vertices of the graph. The packing number $\rho(G)$ of $G$ is the maximum size of a vertex set whose closed neighborhoods are pairwise disjoint. In this paper we study graph classes ${\cal G}$ such that $\gamma(G)/\rho(G)$ is bounded

Rocco Gianni Rapisarda, Davide Ginelli, Diego Clerissi, Leonardo Mariani

Chatbots are software typically embedded in Web and Mobile applications designed to assist the user in a plethora of activities, from chit-chatting to task completion. They enable diverse forms of interactions, like text and voice commands. As any software, even chatbots are susceptible to bugs, and their pervasiveness in our lives, as well as the underlying

Mirja Granfors, Jesús Pineda, Blanca Zufiria Gerbolés, Joana B. Pereira

Graphs provide a powerful framework for modeling complex systems, but their structural variability poses significant challenges for analysis and classification. To address these challenges, we introduce GAUDI (Graph Autoencoder Uncovering Descriptive Information), a novel unsupervised geometric deep learning framework designed to capture both local details a

Á. Sánchez-Monge, C. L. Brogan, T. R. Hunter, A. Ahmadi

The ALMAGAL Large Program has observed 1017 high-mass star-forming regions distributed throughout the Galaxy, sampling different evolutionary stages and environmental conditions. In this work, we present the acquisition and processing of the ALMAGAL data. The main goal is to set up a robust pipeline that generates science-ready products, with a good and unif

Michael R. Douglas, Kit Fraser-Taliente

We review the problem of finding paths in Cayley graphs of groups and group actions, using the Rubik's cube as an example, and we list several more examples of significant mathematical interest. We then show how to formulate these problems in the framework of diffusion models. The exploration of the graph is carried out by the forward process, while finding

Charles Naudet, Brian Taylor, Matthew J. Zahr

We present a high-order, sharp-interface method for simulation of two-phase flow of real gases using implicit shock tracking. The method is based on a phase-field formulation of two-phase, compressible, inviscid flow with a trivial mixture model. Implicit shock tracking is a high-order, optimization-based discontinuous Galerkin method that automatically alig

Coralie Neiner, Adrien Girardot, Jean-Michel Reess

Several space missions are proposed or planned for the coming two decades dedicated or including mid- to high-resolution spectropolarimetry on a wide UV band. This includes the European instrument Pollux for the NASA HWO flagship mission, the NASA SMEX candidate Polstar, and the French nanosatellite demonstrator CASSTOR. We are developing UV polarimeters for

S. Molinari, P. Schilke, C. Battersby, P. T. P. Ho

Fundamental questions about the physics responsible for fragmenting molecular parsec-scale clumps into cores of ~1000 au are still open, that only a statistically significant investigation with ALMA is able to address: what are the dominant agents that determine the core demographics, mass, and spatial distribution as a function of the physical properties of

Justus Tobias Tsang

In recent years there has been impressive progress in quark flavour physics, with current efforts tackling complicated quantities such as for example inclusive decays, decays to QCD-unstable final states and radiative decays. At the same time current lattice flavour physics results are receiving a lot of attention from outside the lattice community. This req

Michael P. Tuite, Michael Welby

For a simple, self-dual, strong CFT-type vertex operator algebra (VOA) of central charge $c$, we describe the Virasoro $n$-point correlation function on a genus $g$ marked Riemann surface in the Schottky uniformisation. We show that this $n$-point function determines the correlation functions for all Virasoro vacuum descendants. Using our recent work on genu

Rémi Perrier, Laura Hernández, J. Ignacio Alvarez-Hamelin, Mariano G. Beiró Dimitris Kotzinos

We present an analysis of the dynamics of discussions in Twitter (before it became X) among supporters of various candidates in the 2022 French presidential election, and followers of different types of media. Our study demonstrates that we can automatically detect the synchronization of interest among different groups around specific topics at particular ti

Bryan Etzine, Masoud Hashemi, Nishanth Madhusudhan, Sagar Davasam

Existing benchmarks are becoming saturated and struggle to separate model performances due to factors like data contamination and advancing LLM capabilities. This paper introduces EMDM (Enhanced Model Differentiation Metric), a novel weighted metric that revitalizes benchmarks by enhancing model separation. EMDM integrates final answer and Chain-of-Thought (

Shi-Shun Chen

Google Scholar is a vital tool for engineering scholars, enabling efficient literature searches and facilitating academic dissemination. Elsevier, as one of the largest publishers of engineering journals, produces essential research that scholars rely on. The pre-proof policy, adopted by Elsevier for certain journals, allows articles to be published online i

Junpeng Jing, Weixun Luo, Ye Mao, Krystian Mikolajczyk

This paper introduces Stereo Any Video, a powerful framework for video stereo matching. It can estimate spatially accurate and temporally consistent disparities without relying on auxiliary information such as camera poses or optical flow. The strong capability is driven by rich priors from monocular video depth models, which are integrated with convolutiona

Alexandra Lassota, Koen Ligthart

We study integer linear programs (ILP) of the form $\min\{c^\top x\ \vert\ Ax=b,l\le x\le u,x\in\mathbb Z^n\}$ and analyze their parameterized complexity with respect to their distance to the generalized matching problem, following the well-established approach of capturing the hardness of a problem by the distance to triviality. The generalized matching pro

A. Hati, M. Pomponio, N. V. Nardelli, T. Grogan

This paper presents a frequency synthesis that achieves exceptional stability by transferring optical signals to the radio frequency (RF) domain at 100 MHz. We describe and characterize two synthesis chains composed of a cryogenic silicon cavity-stabilized laser at 1542 nm and an ultra-low expansion (ULE) glass cavity at 1157 nm, both converted to 10 GHz sig

Raphael Trumpp, Ansgar Schäfftlein, Mirco Theile, Marco Caccamo

As image-based deep reinforcement learning tackles more challenging tasks, increasing model size has become an important factor in improving performance. Recent studies achieved this by focusing on the parameter efficiency of scaled networks, typically using Impala-CNN, a 15-layer ResNet-inspired network, as the image encoder. However, while Impala-CNN evide

Xianghe Han, Hui Chen, Zhongyi Cao, Jingwen Guo

The symmetry breaking and its interplay among spin, charge, and lattice degrees of freedom is crucial for understanding correlated quantum states such as charge density waves (CDWs) and unconventional superconductivity. Here, we report the discovery by low-temperature scanning tunneling microscopy/spectroscopy of unconventional charge-spin-intertwined densit

Mengting Zhao, James Blyth, Grace L. Causer, Hongrun Zhang

Flat bands with small energy dispersion can give rise to strongly correlated electronic and topological phases, especially when located at the Fermi level. Whilst flat bands have been experimentally realized in two-dimensional (2D) twisted van der Waals heterostructures, they are highly sensitive to twist angle, necessitating complex fabrication techniques.

Junbo Zhao, Ting Zhang, Jiayu Sun, Mi Tian

Geometry problem solving has garnered increasing attention due to its potential applications in intelligent education field. Inspired by the observation that text often introduces ambiguities that diagrams can clarify, this paper presents Pi-GPS, a novel framework that unleashes the power of diagrammatic information to resolve textual ambiguities, an aspect

Laura Hucker, Markus Reiß, Thomas Stark

We consider standard gradient descent, gradient flow and conjugate gradients as iterative algorithms for minimising a penalised ridge criterion in linear regression. While it is well known that conjugate gradients exhibit fast numerical convergence, the statistical properties of their iterates are more difficult to assess due to inherent non-linearities and

Juan Miguel Valverde, Maja Østergaard, Adrian Rodriguez-Palomo, Peter Alling Strange Vibe

Accurate segmentation of thin, tubular structures (e.g., blood vessels) is challenging for deep neural networks. These networks classify individual pixels, and even minor misclassifications can break the thin connections within these structures. Existing methods for improving topology accuracy, such as topology loss functions, rely on very precise, topologic

Leonel Rozo, Miguel González-Duque, Noémie Jaquier, Søren Hauberg

Latent variable models are powerful tools for learning low-dimensional manifolds from high-dimensional data. However, when dealing with constrained data such as unit-norm vectors or symmetric positive-definite matrices, existing approaches ignore the underlying geometric constraints or fail to provide meaningful metrics in the latent space. To address these

Julius Franke, Akmaral Moldagalieva, Pia Hanfeld, Wolfgang Hönig

We present a novel approach for generating motion primitives for kinodynamic motion planning using diffusion models. The motions generated by our approach are adapted to each problem instance by utilizing problem-specific parameters, allowing for finding solutions faster and of better quality. The diffusion models used in our approach are trained on randomly

Rickmer Schulte, David Rügamer

Additive models (AMs) have sparked a lot of interest in machine learning recently, allowing the incorporation of interpretable structures into a wide range of model classes. Many commonly used approaches to fit a wide variety of potentially complex additive models build on the idea of boosting additive models. While boosted additive models (BAMs) work well i

Kohei Inayoshi

The James Webb Space Telescope has detected massive black holes (BHs) with masses of $\sim 10^{6-8}~M_\odot$ within the first billion years of the universe. One of the remarkable findings is the identification of "Little Red Dots" (LRDs), a unique class of active galactic nuclei (AGNs) with distinct characteristics representing a key phase in the formation a

Takeaki Miyamura, Yoshiki Sunada, Zhiling Wang, Jesper Ilves

Scaling up a superconducting quantum computer will likely require quantum communication between remote chips, which can be implemented using an itinerant microwave photon in a transmission line. To realize high-fidelity communication, it is essential to control the frequency and temporal shape of the microwave photon. In this work, we demonstrate the generat

Payal D. Solanki, Anh Pham

Quantum Extreme Learning Machine (QELM) is an emerging hybrid quantum machine learning framework that leverages quantum system dynamics to enhance classical models. However, QELM can suffer from the exponential concentration problem, where excessive entanglement reduces model expressivity. In this work, we gain insight into this challenge and demonstrate how

Josef Martínek, Erin Carson, Robert Scheichl

Multilevel sampling methods, such as multilevel and multifidelity Monte Carlo, multilevel stochastic collocation, or delayed acceptance Markov chain Monte Carlo, have become standard uncertainty quantification (UQ) tools for a wide class of forward and inverse problems. The underlying idea is to achieve faster convergence by leveraging a hierarchy of models,

Ahmed Ayad, Dominik J. Schwarz

Axions and axion-like particles (ALPs) have gained substantial attention as potential candidates for cold dark matter. The ALP field can exhibit fluctuations stemming from initial conditions. These initial field fluctuations hold the potential to give rise to gravitationally bound configurations known as axion miniclusters (AMC). While this proposition is wi

Alex Fedorov, Yutong Bu, Xiao Hu, Chris Rorden

Efficient and accurate whole-brain lesion segmentation remains a challenge in medical image analysis. In this work, we revisit MeshNet, a parameter-efficient segmentation model, and introduce a novel multi-scale dilation pattern with an encoder-decoder structure. This innovation enables capturing broad contextual information and fine-grained details without

Shai Bergman, Anne-Marie Kermarrec, Diana Petrescu, Rafael Pires

Retrieval-augmented generation (RAG) improves the reliability of large language model (LLM) answers by integrating external knowledge. However, RAG increases the end-to-end inference time since looking for relevant documents from large vector databases is computationally expensive. To address this, we introduce Proximity, an approximate key-value cache that

Roberto Cerina

This paper introduces PoSSUM, an open-source protocol for unobtrusive polling of social-media users via multimodal Large Language Models (LLMs). PoSSUM leverages users' real-time posts, images, and other digital traces to create silicon samples that capture information not present in the LLM's training data. To obtain representative estimates, PoSSUM employs

Jakob Miller, Martin Sandfuchs, Carla Ferradini

Two-source extractors aim to extract randomness from two independent sources of weak randomness. It has been shown that any two-source extractor which is secure against classical side information remains secure against quantum side information. Unfortunately, this generic reduction comes with a significant penalty to the performance of the extractor. In this

Gabriel Corrigan

We define the symmetric (outer) automorphism group of a right-angled Artin group and construct for it a (spine of) Outer space. This `symmetric spine' is a contractible cube complex upon which the symmetric outer automorphism group acts properly and cocompactly. One artefact of our technique is a strengthening of the proof of contractibility of the untwisted

C. S. Jeffery

The recent discovery of large-amplitude pulsations in faint blue stars (BLAPs) provides both challenges for stellar pulsation theory and opportunities to explore the late evolution of low-mass stars. This paper explores the radial-mode stability of stars across parameter space occupied by BLAPs. Models are constructed for homogeneous stellar envelopes and ar

Sergei I. Simak, Erna K. Delczeg-Czirjak, Olle Eriksson

The predictive accuracy of density functional theory (DFT) for alloy formation enthalpies is often limited by intrinsic energy resolution errors, particularly in ternary phase stability calculations. In this work, we present a machine learning (ML) approach to systematically correct these errors, improving the reliability of first-principles predictions. A n

Ryosuke Nogami, Jaeha Lee

We establish a necessary and sufficient condition for the existence of a quantum state that reproduces given correlation values in the Clauser--Horne--Shimony--Holt (CHSH) setup for any fixed normalized observables. This result addresses a fundamental question shared by both local realism and quantum mechanics: under what conditions a given set of observed d

Eren Erogullari, Sebastian Lapuschkin, Wojciech Samek, Frederik Pahde

Concept Activation Vectors (CAVs) are widely used to model human-understandable concepts as directions within the latent space of neural networks. They are trained by identifying directions from the activations of concept samples to those of non-concept samples. However, this method often produces similar, non-orthogonal directions for correlated concepts, s

Zuhair Al-Johar

When working in NF, [1] there is a sense that there are more non-Cantorian sets than Cantorian sets. But it is not that immediate result as one expects, since they are externally equinumerous, and the qualification "Cantorian" is not stratified and so not easy to spell internally. This account stipulates a fairly simple criterion to phrase such problems and

Christian Ketterer

This is a survey about the contruction of warped products between (semi-)Riemannian manifolds and metric (measure) spaces. The resulting spaces will be semi-Riemannian manifolds, metric (measure) spaces or Lorentzian metric and metric measure spaces. We present details of the contruction in each case and we will highlight important properties like fiber inde

Romain Hermary, Vincent Gaudillière, Abd El Rahman Shabayek, Djamila Aouada

One-class anomaly detection aims to detect objects that do not belong to a predefined normal class. In practice training data lack those anomalous samples; hence state-of-the-art methods are trained to discriminate between normal and synthetically-generated pseudo-anomalous data. Most methods use data augmentation techniques on normal images to simulate anom

Andrii Anataichuk, Sabine Harribey

This paper studies generic surface defects for multiscalar critical models using a perturbative $\epsilon$ expansion in $4-\epsilon$ dimensions. The beta functions of the defect couplings for a generic multiscalar bulk with quartic interactions are computed at first non-trivial order in $\epsilon$. Specific bulks of interest are then considered: $O(N)$, hype

Krystian Roslon

This proceeding provides an expanded overview of the Fast Interaction Trigger (FIT) system performance, focusing on new developments such as the prospective integration of the ALICE Low-Level Front-End Device (ALFRED) into the Detector Control System (DCS) and an upgraded Front-End Electronics (FEE) approach to enhance dynamic range and operational reliabili

Jameel-Un Nabi, Abdel Nasser Tawfik, Nada Ezzelarab, Ali Abas Khan

We assume nuclear statistical equilibrium (NSE) conditions and use Saha Equation to calculate mass fractions in stellar interior during presupernova evolution of massive stars. Our ensemble contains 728 nuclei. The distinguishing feature of our calculation is a state by state summation of nuclear level densities up to 10 MeV for all nuclei considered in our

Frederic Lemieux, Aisha Behr, Clara Kellermann-Bryant, Zaki Mohammed

Cognitive biases, systematic deviations from rationality in judgment, pose significant challenges in generating objective content. This paper introduces a novel approach for real-time cognitive bias detection in user-generated text using large language models (LLMs) and advanced prompt engineering techniques. The proposed system analyzes textual data to iden

Cixiao Zhang, Yin Xu, Size Peng, Xinghao Guo

This paper leverages fluid antenna (FA) and rate-splitting multiple access (RSMA) to enhance the physical layer security (PLS) of an integrated sensing and communication (ISAC) system. We consider a practical multi-user multi-input single-output (MU-MISO) system, where a base station (BS) equipped with fixed position antennas (FPAs) employs RSMA to communica

Guolin Yin, Junqing Zhang, Yuan Ding, Simon Cotton

Securing Internet of Things (IoT) devices presents increasing challenges due to their limited computational and energy resources. Radio Frequency Fingerprint Identification (RFFI) emerges as a promising authentication technique to identify wireless devices through hardware impairments. RFFI performance under low signal-to-noise ratio (SNR) scenarios is signi

Walter Gubler, Joseph Rabinoff

In this paper, we generalize results on Zhang's semipositive model metrics from the algebraic setting to strictly analytic spaces over a non-trivially valued non-Archimedean field. We prove stability under pointwise limits and under forming the maximum. We also prove the maximum principle. As tools, we use a local lifting theorem from the special fiber to th

Sigmund Kohler

We study the stationary state of an ac-driven two-level system under particle exchange with a fermionic environment. A particular question addressed is whether there exist limits in which the populations of the Floquet states are determined by their quasienergies or their mean energies, respectively. The focus lies on parameters in the vicinity of conical in

Jiahui Fan, Fujun Luan, Jian Yang, Miloš Hašan

Novel view synthesis (NVS) from multiple captured photos of an object is a widely studied problem. Achieving high quality typically requires dense sampling of input views, which can lead to frustrating manual labor. Manually positioning cameras to maintain an optimal desired distribution can be difficult for humans, and if a good distribution is found, it is

Segei Kucherenko, Nilay Shah, Oleksiy Klymenko

The primary objective of flexibility analysis is to identify and define the feasibility region, which represents the range of operational conditions (e.g., variations in process parameters) that ensure safe, reliable, and feasible process performance. This work introduces a novel flexibility analysis method that requires only that model constraints (e.g., de

Akash Dhasade, Anne-Marie Kermarrec, Erick Lavoie, Johan Pouwelse

Federated Learning (FL) enables end-user devices to collaboratively train ML models without sharing raw data, thereby preserving data privacy. In FL, a central parameter server coordinates the learning process by iteratively aggregating the trained models received from clients. Yet, deploying a central server is not always feasible due to hardware unavailabi

Yunsong Zhang, Xinyu Zhou, Feitian Zhang

Artificial muscle-driven modular soft robots exhibit significant potential for executing complex tasks. However, their broader applicability remains constrained by the lack of dynamic model-based control strategies tailored for multi-degree-of-freedom (DOF) configurations. This paper presents a novel design of a 2-DOF robotic wrist, envisioned as a fundament

Qingyuan Liang, Zhao Zhang, Zeyu Sun, Zheng Lin

Grammar serves as a cornerstone in programming languages and software engineering, providing frameworks to define the syntactic space and program structure. Existing research demonstrates the effectiveness of grammar-based code representations in small-scale models, showing their ability to reduce syntax errors and enhance performance. However, as language m

Xiamiao Zhao, Yuxuan Yang

A graph $G$ of order $n$ is called edge-pancyclic if, for every integer $k$ with $3 \leq k \leq n$, every edge of $G$ lies in a cycle of length $k$. Determining the minimum size $f(n)$ of a simple edge-pancyclic graph with $n$ vertices seems difficult. Recently, Li, Liu and Zhan \cite{li2024minimum} gave both a lower bound and an upper bound of $f(n)$. In th

Yusong Ke, Hongru Lin, Yuting Ruan, Junya Tang

Large language models (LLMs) are increasingly adopted in medical question-answering (QA) scenarios. However, LLMs can generate hallucinations and nonfactual information, undermining their trustworthiness in high-stakes medical tasks. Conformal Prediction (CP) provides a statistically rigorous framework for marginal (average) coverage guarantees but has limit

Carl B. Rosenkvist, Hannah Elfner

This study optimizes resonance parameters responsible for strangeness production in the SMASH (Simulating Many Accelerated Strongly-interacting Hadrons) transport model using a genetic algorithm. By fitting resonance parameters to experimental data on exclusive strangeness cross-sections at low energies, we significantly improve the model's accuracy, especia

David Meyer, Lukas Niebel, Christian Seis

We construct steady non-spherical bubbles and drops, which are traveling wave solutions to the axisymmetric two-phase Euler equations with surface tension, whose inner phase is a bounded connected domain. The solutions have a uniform vorticity distribution in this inner phase and they have a vortex sheet on its surface. Our construction relies on a perturbat

Miguel Lloret-Climent, Andrés Montoyo-Guijarro, Yoan Gutierrez-Vázquez, Rafael Muñoz-Guillena

Purpose - The purpose of this paper is to propose a mathematical model to determine invariant sets, set covering, orbits and, in particular, attractors in the set of tourism variables. Analysis was carried out based on an algorithm and applying an interpretation of chaos theory developed in the context of General Systems Theory and Big Data. Design/methodolo

José A. Caballero, Walter Seifert, Andreas Quirrenbach, Pedro J. Amado

CARMENES stands for Calar Alto high-Resolution search for M dwarfs with Exoearths with Near-infrared and optical \'Echelle Spectrographs. CARMENES took six years from a concept to the start of operations, and a couple more years of initial data collection until the first science publication, but now is revolutionising our knowledge on exoplanets and their st

Md Atik Ahamed, Qiang Ye, Qiang Cheng

The design of novel molecules with desired properties is a key challenge in drug discovery and materials science. Traditional methods rely on trial-and-error, while recent deep learning approaches have accelerated molecular generation. However, existing models struggle with generating molecules based on specific textual descriptions. We introduce Mol-CADiff,

S. Naresh Chockalingam, Narayan K. Sundaram

Hierarchically structured cellular solids have attracted increasing attention for their superior mass-specific mechanical properties. Using a remeshing-based continuum finite element (FE) framework, we reveal that two-scale metallic hierarchical solids exhibit a distinct, localized deformation mode that involves necking and fracture of microscale tension mem

W. H. Ma, D. Y. Tao, B. Zhou, J. S. Wang

The study of $t$+$t$ cluster states in $^{6}$He provides valuable insights into exotic nuclear structures and the behavior of fermionic cluster systems. This study shows rich cluster resonant state structures above the threshold, identified by experimental reconstruction and theoretical calculations. The excitation energy spectrum above the $t$+$t$ threshold

Gianluca Bande

In this short note we recall the definition of intrinsically harmonic forms, some known results and some open problems.

Mohammadreza Malekabbasi, Tobias Pfandzelter, David Bermbach

Application users react negatively to performance regressions or availability issues across software releases. To address this, modern cloud-based applications with their multiple daily releases rely on live testing techniques such as A/B testing or canary releases. In edge-to-cloud applications, however, which have similar problems, developers currently sti

Joshua Althüser, Götz S. Uhrig

We numerically study the collective excitations present in isotropic superconductors including a screened Coulomb interaction. By varying the screening strength, we analyze its impact on the system. We use a formulation of the effective phonon-mediated interaction between electrons that depends on the energy transfer between particles, rather than being a co

Qijiong Liu, Jieming Zhu, Lu Fan, Kun Wang

In recent years, integrating large language models (LLMs) into recommender systems has created new opportunities for improving recommendation quality. However, a comprehensive benchmark is needed to thoroughly evaluate and compare the recommendation capabilities of LLMs with traditional recommender systems. In this paper, we introduce RecBench, which systema

Haotian Hu, Jingwei Xu, Fanyi Wang, Toyota Li

Reconstruction of high-definition maps is a crucial task in perceiving the autonomous driving environment, as its accuracy directly impacts the reliability of prediction and planning capabilities in downstream modules. Current vectorized map reconstruction methods based on the DETR framework encounter limitations due to the redundancy in the decoder structur

Loïc Fosse, Frédéric Béchet, Benoît Favre, Géraldine Damnati

Tasks are central in machine learning, as they are the most natural objects to assess the capabilities of current models. The trend is to build general models able to address any task. Even though transfer learning and multitask learning try to leverage the underlying task space, no well-founded tools are available to study its structure. This study proposes

Amit Levy, Itzik Klein

The unscented Kalman filter is an algorithm capable of handling nonlinear scenarios. Uncertainty in process noise covariance may decrease the filter estimation performance or even lead to its divergence. Therefore, it is important to adjust the process noise covariance matrix in real time. In this paper, we developed an adaptive neural unscented Kalman filte

Jaewoo Song, Fangzhen Lin

The quantization of large language models (LLMs) is crucial for deploying them on devices with limited computational resources. While advanced quantization algorithms offer improved performance compared to the basic linear quantization, they typically require high-end graphics processing units (GPUs), are often restricted to specific deep neural network (DNN

Leo P. Mulholland, Steven J. Bromley, Connor P. Ballance, Stuart A. Sim

The Axelrod approximation is widely used in astrophysical modelling codes to evaluate electron-impact excitation effective collision strengths for forbidden transitions. Approximate methods such as this are a necessity for many heavy elements with open shells where collisional data is either non existent or sparse as the use of more robust methods prove proh

Minsoo Khang, Sang Chul Jung, Sungrae Park, Teakgyu Hong

Document Key Information Extraction (KIE) is a technology that transforms valuable information in document images into structured data, and it has become an essential function in industrial settings. However, current evaluation metrics of this technology do not accurately reflect the critical attributes of its industrial applications. In this paper, we prese

Thibault Damour, Alessandro Nagar, Andrea Placidi, Piero Rettegno

We present a new approach to the conservative dynamics of binary systems, within the effective one-body (EOB) framework, based on the use of a Lagrange multiplier to impose the mass-shell constraint. When applied to the post-Minkowskian (PM) description of the two-body problem in Einsteinian gravity, this Lagrange-EOB (LEOB) approach allows for a new formula

Ruxin Zheng, Shunqiao Sun, Hongshan Liu

Sparse arrays have been widely exploited in radar systems because of their advantages in achieving large array aperture at low hardware cost, while significantly reducing mutual coupling. However, sparse arrays suffer from high sidelobes which may lead to false detections. Missing elements in sparse arrays can be interpolated using the sparse array measureme

Marco Geraci

The generalized Laplace (GL) distribution, which falls in the larger family of generalized hyperbolic distributions, provides a versatile model to deal with a variety of applications thanks to its shape parameters. The elliptically symmetric GL admits a polar representation that can be used to yield a circular distribution, which we call projected GL (PGL) d

Miaowei Wang, Yibo Zhang, Rui Ma, Weiwei Xu

We present DecoupledGaussian, a novel system that decouples static objects from their contacted surfaces captured in-the-wild videos, a key prerequisite for realistic Newtonian-based physical simulations. Unlike prior methods focused on synthetic data or elastic jittering along the contact surface, which prevent objects from fully detaching or moving indepen

George Mihailescu, Saubhik Sarkar, Abolfazl Bayat, Steve Campbell

The theoretical foundation of quantum sensing is rooted in the Cram\'er-Rao formalism, which establishes quantitative precision bounds for a given quantum probe. In many practical scenarios, where more than one parameter is unknown, the multi-parameter Cram\'er-Rao bound (CRB) applies. Since this is a matrix inequality involving the inverse of the quantum Fi

Christofer Fellicious, Hans P. Reiser, Michael Granitzer

Forensic Memory Analysis (FMA) and Virtual Machine Introspection (VMI) are critical tools for security in a virtualization-based approach. VMI and FMA involves using digital forensic methods to extract information from the system to identify and explain security incidents. A key challenge in both FMA and VMI is the "Semantic Gap", which is the difficulty of

Tingmingke Lu

Large language models (LLMs) often generate inaccurate yet credible-sounding content, known as hallucinations. This inherent feature of LLMs poses significant risks, especially in critical domains. I analyze LLMs as a new class of engineering products, treating hallucinations as a product attribute. I demonstrate that, in the presence of imperfect awareness

Guillermo Encinas-Lago, Vincenzo Sciancalepore, Henk Wymeersch, Marco Di Renzo

The advance towards 6G networks comes with the promise of unprecedented performance in sensing and communication capabilities. The feat of achieving those, while satisfying the ever-growing demands placed on wireless networks, promises revolutionary advancements in sensing and communication technologies. As 6G aims to cater to the growing demands of wireless

Sergio Cobos, Javier Luis Cánovas Izquierdo

The development of Open-Source Software (OSS) projects relies on the collaborative work of contributors, generally scattered around the world. To enable this collaboration, OSS projects are hosted on social-coding platforms like GitHub, which provide the infrastructure to host the code as well as the support for enabling the participation of the community. T

Yu Rong, Peng Wang, Xiao-xiao Tang

We investigate the alignment between the angular momenta of galaxy groups and the spines of their associated cosmic filaments. Our results demonstrate a significant tendency for these two orientations to be perpendicular, indicating that the rotation of a galaxy group does not originate from the spin of cosmic filaments. Instead, it is driven by the orbital

Nizo Jaman Shohan, Gazi Tanbhir, Faria Elahi, Ahsan Ullah

The distributed denial-of-service (DDoS) attack stands out as a highly formidable cyber threat, representing an advanced form of the denial-of-service (DoS) attack. A DDoS attack involves multiple computers working together to overwhelm a system, making it unavailable. On the other hand, a DoS attack is a one-on-one attempt to make a system or website inacce

Gerold Alsmeyer, Konrad Kolesko, Matthias Meiners, Jakob Stonner

We study the phenomenon of explosion in general (Crump-Mode-Jagers) branching processes, which refers to the event where an infinite number of individuals are born in finite time. In a critical setting where the expected number of immediate offspring per individual is exactly one, whether or not explosion occurs depends on the fine properties of the reproduc

Jonas Schäfer, Benjamin A. Stickler, Klaus Hornberger

We derive the quantum master equation predicting how the translational and rotational dynamics of a nanoparticle is affected by the emission of surface adsorbates. This is motivated by recent experiments which prepared the motion of internally hot silica particles in the deep quantum regime. In the limit of a well localized nanoparticle the ro-translational

Ziran Zhou, Guanyu Gao, Xiaohu Wu, Yan Lyu

Personalized Federated Learning (PFL) aims to train a personalized model for each client that is tailored to its local data distribution, learning fails to perform well on individual clients due to variations in their local data distributions. Most existing PFL methods focus on personalizing the aggregated global model for each client, neglecting the fundame

Lorenz Wolf, Sangwoong Yoon, Ilija Bogunovic

Mixture of large language model (LLMs) Agents (MoA) architectures achieve state-of-the-art performance on prominent benchmarks like AlpacaEval 2.0 by leveraging the collaboration of multiple LLMs at inference time. Despite these successes, an evaluation of the safety and reliability of MoA is missing. We present the first comprehensive study of MoA's robustn

Noah Mamie, Susie Xi Rao

Multi-agent systems address issues of accessibility and scalability of artificial intelligence (AI) foundation models, which are often represented by large language models. We develop a framework - the "Society of HiveMind" (SOHM) - that orchestrates the interaction between multiple AI foundation models, imitating the observed behavior of animal swarms in na

Tano Kim Kender, Marco Corrias, Cesare Franchini

Quasicrystals are aperiodically ordered solids that exhibit long-range order without translational periodicity, bridging the gap between crystalline and amorphous materials. Due to their lack of translational periodicity, information on atomic arrangements in quasicrystals cannot be extracted by current crystalline lattice recognition softwares. This work in

Changjia Ma, Yi Zhao, Zhongxue Gan, Bingzhao Gao

Trajectory optimization in multi-vehicle scenarios faces challenges due to its non-linear, non-convex properties and sensitivity to initial values, making interactions between vehicles difficult to control. In this paper, inspired by topological planning, we propose a differentiable local homotopy invariant metric to model the interactions. By incorporating

Joan Giner-Miguelez, Sergio Morales, Sergio Cobos, Javier Luis Canovas Izquierdo

Context: Interest in diversity in software development has significantly increased in recent years. Reporting on diversity in software projects can enhance user trust and assist regulators in evaluating adoption. Recent AI directives include clauses that mandate diversity information during development, highlighting the growing interest of public regulators.

Peter Mörters, Nick Schleicher

We identify the size of the largest connected component in a subcritical inhomogeneous random graph with a kernel of preferential attachment type. The component is polynomial in the graph size with an explicitly given exponent, which is strictly larger than the exponent for the largest degree in the graph. This is in stark contrast to the behaviour of inhomo

Konrad Kolesko, Matthias Meiners, Ivana Tomic

We consider the Markov renewal equation $F(t) = f(t) + \boldsymbol{\mu}*F(t)$ for vector-valued functions $f,F: \mathbb{R} \to \mathbb{R}^{p}$ and a $p \times p$ matrix $\boldsymbol{\mu}$ of locally finite measures $\mu^{i,j}$ on $[0,\infty)$, $i,j=1,\ldots,p$. Sgibnev [Semimultiplicative estimates for the solution of the multidimensional renewal equation. {

Christian Ikenmeyer, Jakob Moosbauer

We give a short proof for Strassen's result that the rank of the 2 by 2 matrix multiplication tensor is at most 7. The proof requires no calculations and also no pattern matching or other type of nontrivial verification, and is based solely on properties of a specific order 6 group action. Our proof is based on the recent combination of flip graph algorithms

Mingming Wang, Guowu Yuan, Hailan He, Chengming Tan

Accurate identification of solar radio bursts (SRBs) is essential for advancing research in solar physics and predicting space weather. However, the majority of current studies mainly concentrate on detecting whether SRBs are present or absent, often focusing on only one particular type of burst. Moreover, the neural network models used for SRB detection are