February 2025 arXiv papers — page 12

Jens Høigaard Jensen, Kristoffer Plagborg Bak Sørensen, Jonas le Fevre Sejersen, Andriy Sarabakha

Multi-agent path planning is a critical challenge in robotics, requiring agents to navigate complex environments while avoiding collisions and optimizing travel efficiency. This work addresses the limitations of existing approaches by combining Gaussian belief propagation with path integration and introducing a novel tracking factor to ensure strict adherenc

Sichong Zhang, Xiong Wang, Fei Lu

Learning kernels in operators from data lies at the intersection of inverse problems and statistical learning, providing a powerful framework for capturing non-local dependencies in function spaces and high-dimensional settings. In contrast to classical nonparametric regression, where the inverse problem is well-posed, kernel estimation involves a compact no

Boya Zhang, Iris Andrussow, Andreas Zell, Georg Martius

Stable and robust robotic grasping is essential for current and future robot applications. In recent works, the use of large datasets and supervised learning has enhanced speed and precision in antipodal grasping. However, these methods struggle with perception and calibration errors due to large planning horizons. To obtain more robust and reactive grasping

Leticia Bertuzzi, João P. Engster, Evandro C. R. da Rosa, Eduardo I. Duzzioni

Improving the performance of quantum algorithms is a fundamental task to achieve quantum advantage. In many cases, extracting information from quantum systems poses an important challenge for practical implementations in real-world quantum computers, given the high resource cost of performing state tomography. In this scenario, randomized measurements emerge

Jonas le Fevre Sejersen, Erdal Kayacan

This study presents the conflict-aware multi-agent estimated time of arrival (CAMETA) framework, a novel approach for predicting the arrival times of multiple agents in unstructured environments without predefined road infrastructure. The CAMETA framework consists of three components: a path planning layer generating potential path suggestions, a multi-agent

Nicholas A. Lanzillo, Sergey Faleev, Aakash Pushp

Magnetic Tunnel Junction (MTJ) based Spin-Transfer Torque Magnetic Random Access Memory (STT-MRAM) is poised to replace embedded Flash for advanced applications such as automotive microcontroller units. To achieve deeper technological adoption, MTJ needs to exhibit three key features: low magnetization (Ms), high perpendicular magnetic anisotropy (PMA) and h

Ryan C. Barron, Maksim E. Eren, Olga M. Serafimova, Cynthia Matuszek

Agentic Generative AI, powered by Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG), Knowledge Graphs (KGs), and Vector Stores (VSs), represents a transformative technology applicable to specialized domains such as legal systems, research, recommender systems, cybersecurity, and global security, including proliferation research. This tec

Antoine Belley, Jose M. Munoz, Ronald F. Garcia Ruiz

We introduce a hierarchical framework that combines ab initio many-body calculations with a Bayesian neural network, developing emulators capable of accurately predicting nuclear properties across isotopic chains simultaneously and being applicable to different regions of the nuclear chart. We benchmark our developments using the oxygen isotopic chain, achie

Chi-Hang Ng, Pin-Gao Gu, Thiem Hoang

In this paper, we conduct a detailed study on the effect of Radiative Torque Disruption (RATD) mechanism on the fragmentation of micrometer-sized dust grains into nanoparticles within the heliosphere. We start by estimating the disruption timescales for dust grains under various centrifugal stresses. Our numerical calculations demonstrate that RATD is a high

Shuming Liu, Chen Zhao, Fatimah Zohra, Mattia Soldan

Temporal action detection (TAD) is a fundamental video understanding task that aims to identify human actions and localize their temporal boundaries in videos. Although this field has achieved remarkable progress in recent years, further progress and real-world applications are impeded by the absence of a standardized framework. Currently, different methods

Maryam Bahrani, Michael Neuder, S. Matthew Weinberg

Selfish miners selectively withhold blocks to earn disproportionately high revenue. The vast majority of the selfish mining literature focuses exclusively on block rewards. Carlsten et al. [2016] is a notable exception, observing that similar strategic behavior is profitable in a zero-block-reward regime (the endgame for Bitcoin's quadrennial halving schedul

Sai Ganesh Grandhi, Saeed Samet

Traditional authentication methods, such as passwords and biometrics, verify a user's identity only at the start of a session, leaving systems vulnerable to session hijacking. Continuous authentication, however, ensures ongoing verification by monitoring user behavior. This study investigates the long-term feasibility of eye-tracking as a behavioral biometri

Wuzhen Li, Zhiyuan Zhou, Li Chen, Yinhai Li

Linear optics-based multi-beam interference (MBI), like the Fabry-Perot interferometer, plays an important role in precision optical metrology applications such as laser stabilization in optical clocks, precision spectroscopy, and gravitational wave detection. Here, we propose and experimentally verify a nonlinear optics-based MBI principle with the assistan

Wojciech Kotlarski, Gregory Patellis

The idea of reduction of couplings provides a systematic procedure to search for relations among seemingly unrelated parameters of a renormalizable theory. As a consequence, such reduced theories exhibit more constrained parameter spaces. Motivated by this, in this work we perform a precise phenomenological analysis of the Higgs sector of a version of the Ty

Kuan Lok Zhou, Jiayi Chen, Siddharth Suresh, Reuben Narad

Large Language Models (LLMs) have shown significant limitations in understanding creative content, as demonstrated by Hessel et al. (2023)'s influential work on the New Yorker Cartoon Caption Contest (NYCCC). Their study exposed a substantial gap between LLMs and humans in humor comprehension, establishing that understanding and evaluating creative content i

Tobias Boege

A definable set $X$ in the first-order language of rings defines a family of random vectors: for each finite field $\mathbb{F}_q$, let the distribution be supported and uniform on the $\mathbb{F}_q$-rational points of $X$. We employ results from the model theory of finite fields to show that their entropy profiles settle into one of finitely many stable asym

Nazarii Drushchak, Vladyslava Tyshchenko, Nataliya Polyakovska

The growth of Educational Technology (EdTech) has enabled highly personalized learning experiences through Artificial Intelligence (AI)-based recommendation systems tailored to each student needs. However, these systems can unintentionally introduce biases, potentially limiting fair access to learning resources. This study presents a recommendation system fo

Aron Harder, Madhur Behl

Scenario Description Languages (SDLs) provide structured, interpretable embeddings that represent traffic scenarios encountered by autonomous vehicles (AVs), supporting key tasks such as scenario similarity searches and edge case detection for safety analysis. This paper introduces the Trajectory-to-Action Pipeline (TAP), a scalable and automated method for

Sourav Dey, Amaresh Jaiswal

We consider the rotational Brownian motion of heavy quark in QCD medium and provide analytical results for polarization of open heavy-flavor hadrons. We calculate expressions for vector and tensor polarization, corresponding to baryon spin polarization and vector meson spin alignment, respectively. Assuming that heavy quarks are initially fully spin polarize

Dawid Przyczyna, Grzegorz Hess, Konrad Szaciłowski

Nanodevices that show the potential for non-linear transformation of electrical signals and various forms of memory can be successfully used in new computational paradigms, such as neuromorphic or reservoir computing (RC). Dedicated hardware implementations based on functional neuromorphic structures significantly reduce energy consumption and/or increase co

Kai Zhang, Rui Zhu, Shutian Ma, Jingwei Xiong

Drug discovery is a critical task in biomedical natural language processing (NLP), yet explainable drug discovery remains underexplored. Meanwhile, large language models (LLMs) have shown remarkable abilities in natural language understanding and generation. Leveraging LLMs for explainable drug discovery has the potential to improve downstream tasks and real

Thomas Budiarjo, Santana Yuda Pradata, Kadek Gemilang Santiyuda, Muhammad Alfian Amrizal

High energy consumption remains a key challenge in high-performance computing (HPC) systems, which often feature hundreds or thousands of nodes drawing substantial power even in idle or standby modes. Although powering down unused nodes can improve energy efficiency, choosing the wrong time to do so can degrade quality of service by delaying job execution. M

Mehmet Fatih Ozkan, Jeff Chrstos, Marcello Canova, Stephanie Stockar

Accurate driver behavior modeling is essential for improving the interaction and cooperation of the human driver with the driver assistance system. This paper presents a novel approach for modeling the response of human drivers to visual cues provided by a speed advisory system using a Koopman-based method with online updates. The proposed method utilizes th

Kshipra Bhawalkar, Jeff Dean, Christopher Liaw, Aranyak Mehta

We envision a marketplace where diverse entities offer specialized "modules" through APIs, allowing users to compose the outputs of these modules for complex tasks within a given budget. This paper studies the market design problem in such an ecosystem, where module owners strategically set prices for their APIs (to maximize their profit) and a central platf

Diluka Galappaththige, Mohammadali Mohammadi, Gayan Aruma Baduge, Chintha Tellambura

Cell-free (CF) integrated sensing and communication (ISAC) combines CF architecture with ISAC. CF employs distributed access points, eliminates cell boundaries, and enhances coverage, spectral efficiency, and reliability. ISAC unifies radar sensing and communication, enabling simultaneous data transmission and environmental sensing within shared spectral and

Yi Jing, Zijun Yao, Hongzhu Guo, Lingxu Ran

Large language models (LLMs) demonstrate exceptional performance on tasks requiring complex linguistic abilities, such as reference disambiguation and metaphor recognition/generation. Although LLMs possess impressive capabilities, their internal mechanisms for processing and representing linguistic knowledge remain largely opaque. Prior research on linguisti

Seungheon Shin, Byeonghyeon Goh, Youngtaek Oh, Hayoung Chung

Topology optimization produces designs with intricate geometries and complex topologies that require advanced manufacturing techniques such as additive manufacturing (AM). However, insufficient consideration of manufacturability during the optimization process often results in design modifications that compromise the optimality of the design. While multi-axi

Gisya Abdi, Lulu Alluhaibi, Ewelina Kowalewska, Tomasz Mazur

Sensing technology is an important aspect of information processing. Current development in artificial intelligence systems (especially those aimed at medical and environmental applications) requires a lot of data on the chemical composition of biological fluids or environmental samples. These complex matrices require advanced sensing devices, and photoelect

Kaustubh Mani, Vincent Mai, Charlie Gauthier, Annie Chen

Reinforcement learning algorithms typically necessitate extensive exploration of the state space to find optimal policies. However, in safety-critical applications, the risks associated with such exploration can lead to catastrophic consequences. Existing safe exploration methods attempt to mitigate this by imposing constraints, which often result in overly

David Hoffman, Francisco Martin, Brian White

We prove existence for many examples of shrinkers by producing compact, smoothly embedded surfaces that, under mean curvature flow, develop singularities at which the shrinkers occur as blowups.

Daniele Paliotta, Junxiong Wang, Matteo Pagliardini, Kevin Y. Li

Recent advancements have demonstrated that the performance of large language models (LLMs) can be significantly enhanced by scaling computational resources at test time. A common strategy involves generating multiple Chain-of-Thought (CoT) trajectories and aggregating their outputs through various selection mechanisms. This raises a fundamental question: can

Nathaniel K. Brown, Lore Depuydt, Mohsen Zakeri, Anas Alhadi

Long maximal exact matches (MEMs) are used in many genomics applications such as read classification and sequence alignment. Li's ropebwt3 finds long MEMs quickly because it can often ignore much of its input. In this paper we show that a fast and space efficient $k$-mer filtration step using a Bloom filter speeds up MEM-finders such as ropebwt3 even further

Anna C. Childs, Alexa P. S. Hua, Rebecca G. Martin, Chao-Chin Yang

We identify observational signatures suggesting a history of dynamical instability in 26 out of 34 M-dwarf multi-planet systems containing no large planets. These systems may have primarily formed in a gas-rich environment, potentially hosted more planets and were more compact. We extend previous simulations of the formation of the TRAPPIST-1 system to 100 M

Alan Ashworth, Munir Al-Dajani, Keegan Duchicela, Kiril Kafadarov

Clinical decision-making depends on expert reasoning, which is guided by standardized, evidence-based guidelines. However, translating these guidelines into automated clinical decision support systems risks inaccuracy and importantly, loss of nuance. We share an application architecture, the Large Language Expert (LLE), that combines the flexibility and powe

Ben Berger, Edward W. Felten, Akaki Mamageishvili, Benny Sudakov

Optimistic rollups rely on fraud proofs -- interactive protocols executed on Ethereum to resolve conflicting claims about the rollup's state -- to scale Ethereum securely. To mitigate against potential censorship of protocol moves, fraud proofs grant participants a significant time window, known as the challenge period, to ensure their moves are processed on

Tamir Shor, Moti Freiman, Chaim Baskin, Alex Bronstein

Cardiac T1 mapping provides critical quantitative insights into myocardial tissue composition, enabling the assessment of pathologies such as fibrosis, inflammation, and edema. However, the inherently dynamic nature of the heart imposes strict limits on acquisition times, making high-resolution T1 mapping a persistent challenge. Compressed sensing (CS) appro

Yukang Yang, Declan Campbell, Kaixuan Huang, Mengdi Wang

Many recent studies have found evidence for emergent reasoning capabilities in large language models (LLMs), but debate persists concerning the robustness of these capabilities, and the extent to which they depend on structured reasoning mechanisms. To shed light on these issues, we study the internal mechanisms that support abstract reasoning in LLMs. We id

Arunima Bhattacharya, Chinmoy Dey, M. C. Kumar, Vaibhav Pandey

We study the threshold effects for the associated production of a Higgs boson with a massive vector boson $(V=Z,W)$ in the $q\bar{q} \rightarrow V^\star \rightarrow VH$ process at the LHC. By leveraging the universality of threshold logarithms and employing soft-virtual (SV) and next-to-soft virtual (NSV) resummation techniques, we compute threshold correcti

Guanzheng Chen, Qilong Feng, Jinjie Ni, Xin Li

The emergence of long-context large language models (LLMs) offers a promising alternative to traditional retrieval-augmented generation (RAG) for processing extensive documents. However, the computational overhead of long-context inference presents significant efficiency challenges. While Speculative Decoding (SD) traditionally accelerates inference using sm

Alexander Immer, Jan-Matthis Lueckmann, Alex Bo-Yuan Chen, Peter H. Li

Large-scale neuronal activity recordings with fluorescent calcium indicators are increasingly common, yielding high-resolution 2D or 3D videos. Traditional analysis pipelines reduce this data to 1D traces by segmenting regions of interest, leading to inevitable information loss. Inspired by the success of deep learning on minimally processed data in other do

Muhammad Naeem Anwar

At the large distances compared to the chiral symmetry breaking scale, a four-quark system $\bar Q \bar Q qq$ (with $Q$ as heavy and $q$ as light quarks) can be treated as two asymptotic mesons interacting via strong residual forces. The static heavy quark assumption enables using the Born-Oppenheimer approximation, where one can compute the potential betwee

Albert Georg Passegger, Rainer Verch

We calculate the transition rate of an Unruh-DeWitt detector coupled to a non-equilibrium steady state (NESS) of a free massless scalar field on four-dimensional Minkowski spacetime. Bringing two semi-infinite heat baths at different temperatures into thermal contact along a surface, the NESS arises at asymptotically late times as a stationary state that has

Camilla Felisetti, Andras Szenes, Olga Trapeznikova

The study of the intersection cohomology of moduli spaces of semistable bundles was initiated by Frances Kirwan in the 1980's. In this paper, we give a complete geometric proof of a recursive formula, which reduces the calculation of the intersection Betti numbers of the moduli spaces of semistable bundles on Riemann surfaces in degree-0 and arbitrary rank t

Thomas Hickling, Maxwell Hogan, Abdulla Tammam, Nabil Aouf

This paper presents the first end-to-end framework that combines guidance, navigation, and centralised task allocation for multiple UAVs performing autonomous search-and-rescue (SAR) in GNSS-denied indoor environments. A Twin Delayed Deep Deterministic Policy Gradient controller is trained with an Artificial Potential Field (APF) reward that blends attractiv

Tamir Shor, Chaim Baskin, Alex Bronstein

Multi-rotor aerial autonomous vehicles (MAVs, more widely known as "drones") have been generating increased interest in recent years due to their growing applicability in a vast and diverse range of fields (e.g., agriculture, commercial delivery, search and rescue). The sensitivity of visual-based methods to lighting conditions and occlusions had prompted gr

Karolina Gutmańska, Piotr Szweda, Marek Daszkiewicz, Konrad Szaciłowski

In the present study, the structure, thermal stability, conductive properties, and antimicrobial activity of silver(I) complexes with nitrile ligands were investigated. For the construction of the materials, 2 cyanopyridine (2-cpy), 4-cyanopyridine (4-cpy), 1,2-dicyanobenzene (1,2-dcb), and 1,3 dicyanobenzene (1,3-dcb) were used in addition to the silver nit

Xuangeng Chu, Nabarun Goswami, Ziteng Cui, Hanqin Wang

Speech-driven 3D facial animation aims to generate realistic lip movements and facial expressions for 3D head models from arbitrary audio clips. Although existing diffusion-based methods are capable of producing natural motions, their slow generation speed limits their application potential. In this paper, we introduce a novel autoregressive model that achie

Genheng Zhao

Let $s\geq 8$ be an integer and $P$ be a set of primes with relative lower density greater than $\sqrt{1-\min\{s,16\}/32}$. We prove that every sufficiently large integer $n\equiv s({\rm mod}24)$ can be represented by a sum of $s$ squares of primes in $P$.

Chuofan Ma, Yi Jiang, Junfeng Wu, Jihan Yang

Visual generative and understanding models typically rely on distinct tokenizers to process images, presenting a key challenge for unifying them within a single framework. Recent studies attempt to address this by connecting the training of VQVAE (for autoregressive generation) and CLIP (for understanding) to build a unified tokenizer. However, directly comb

Azuka Chiejina, Subhramoy Mohanti, Vijay K. Shah

Recent advancements in AI and edge computing have accelerated the development of machine-centric applications (MCAs), such as smart surveillance systems. In these applications, video cameras and sensors offload inference tasks like license plate recognition and vehicle tracking to remote servers due to local computing and energy constraints. However, legacy

Mehrdad Anvari, Hamidreza Marasi, Hossein Kheiri

Identifying governing equations in physical and biological systems from datasets remains a long-standing challenge across various scientific disciplines, providing mechanistic insights into complex system evolution. Common methods like sparse identification of nonlinear dynamics (SINDy) often rely on precise derivative estimations, making them vulnerable to

Matteo Gallone, Alessandro Michelangeli, Diego Noja

We characterise the families of self-adjoint Dirac and Schr\"{o}dinger operators with Aharonov-Bohm magnetic field, and we exploit the non-relativistic limit of infinite light speed to connect the former to the latter. The limit consists of the customary removal of the rest energy and of a suitable scaling, with the light speed, of the short-scale boundary c

Yongjia Lei, Haoyu Han, Ryan A. Rossi, Franck Dernoncourt

Text-rich Graph Knowledge Bases (TG-KBs) have become increasingly crucial for answering queries by providing textual and structural knowledge. However, current retrieval methods often retrieve these two types of knowledge in isolation without considering their mutual reinforcement and some hybrid methods even bypass structural retrieval entirely after neighb

Mohamed Abdelsamad, Michael Ulrich, Claudius Gläser, Abhinav Valada

Masked autoencoders (MAE) have shown tremendous potential for self-supervised learning (SSL) in vision and beyond. However, point clouds from LiDARs used in automated driving are particularly challenging for MAEs since large areas of the 3D volume are empty. Consequently, existing work suffers from leaking occupancy information into the decoder and has signi

Shangyin Tan, Lakshya A Agrawal, Arnav Singhvi, Liheng Lai

Composing language models (LMs) into multi-step language programs and automatically optimizing their modular prompts is now a mainstream paradigm for building AI systems, but the tradeoffs in this space have only scarcely been studied before. We introduce LangProBe, the first large-scale benchmark for evaluating the architectures and optimization strategies

Tamir Shor, Ethan Fetaya, Chaim Baskin, Alex Bronstein

Implicit Neural Representations (INRs) have been recently garnering increasing interest in various research fields, mainly due to their ability to represent large, complex data in a compact, continuous manner. Past work further showed that numerous popular downstream tasks can be performed directly in the INR parameter-space. Doing so can substantially reduc

Siyu Jiao, Gengwei Zhang, Yinlong Qian, Jiancheng Huang

This work challenges the residual prediction paradigm in visual autoregressive modeling and presents FlexVAR, a new Flexible Visual AutoRegressive image generation paradigm. FlexVAR facilitates autoregressive learning with ground-truth prediction, enabling each step to independently produce plausible images. This simple, intuitive approach swiftly learns vis

Lennox S. Keeble, Alejandro Cárdenas-Avendaño, Daniel C. M. Palumbo

Accurately inferring black hole spin is crucial for understanding black hole dynamics and their astrophysical environments. In this work, we outline a geometric method for spin estimation by using the interferometric shape of the first photon ring ($n=1$) as an approximation to the critical curve, which, given an assumed value of the black hole inclination,

Marcus Yu Zhe Wee, Justin Juin Hng Wong, Lynus Lim, Joe Yu Wei Tan

Effective communication in Air Traffic Control (ATC) is critical to maintaining aviation safety, yet the challenges posed by accented English remain largely unaddressed in Automatic Speech Recognition (ASR) systems. Existing models struggle with transcription accuracy for Southeast Asian-accented (SEA-accented) speech, particularly in noisy ATC environments.

Chris D. A. Blair, Johannes Lahnsteiner, Niels A. Obers, Ziqi Yan

We study properties of non-Lorentzian geometries arising from BPS decoupling limits of string theory that are central to matrix theory and the AdS/CFT correspondence. We focus on duality transformations between ten-dimensional non-Lorentzian geometries coupled to matrix theory on D-branes. We demonstrate that T- and S-duality transformations exhibit novel as

Franck Cappello, Sandeep Madireddy, Robert Underwood, Neil Getty

Recent advancements have positioned AI, and particularly Large Language Models (LLMs), as transformative tools for scientific research, capable of addressing complex tasks that require reasoning, problem-solving, and decision-making. Their exceptional capabilities suggest their potential as scientific research assistants but also highlight the need for holis

Ricardo Alonso, Milana Colic

This paper reviews results on the scalar Boltzmann equation for a single-component polyatomic gas with continuous internal energy. For the space homogeneous problem, $L^1$-theory is established, for solutions with initial strictly positive mass and bounded energy, which enables to solve the Cauchy problem for initial data with $L^1_{2^+}$-moments using the c

Xiuli Bi, Jianfei Yuan, Bo Liu, Yong Zhang

We present Mobius, a novel method to generate seamlessly looping videos from text descriptions directly without any user annotations, thereby creating new visual materials for the multi-media presentation. Our method repurposes the pre-trained video latent diffusion model for generating looping videos from text prompts without any training. During inference,

Lingyu Du, Yupei Liu, Jinyuan Jia, Guohao Lan

Gaze estimation models are widely used in applications such as driver attention monitoring and human-computer interaction. While many methods for gaze estimation exist, they rely heavily on data-hungry deep learning to achieve high performance. This reliance often forces practitioners to harvest training data from unverified public datasets, outsource model

Francesco Hoch, Eugenio Caruccio, Giovanni Rodari, Tommaso Francalanci

The implementation of large-scale universal quantum computation represents a challenging and ambitious task on the road to quantum processing of information. In recent years, an intermediate approach has been pursued to demonstrate quantum computational advantage via non-universal computational models. A relevant example for photonic platforms has been provi

Jack Michael Solomon, Rosemary Renaut, Matthias Chung

Electroencephalograms (EEG) are invaluable for treating neurological disorders, however, mapping EEG electrode readings to brain activity requires solving a challenging inverse problem. Due to the time series data, the use of $\ell_1$ regularization quickly becomes intractable for many solvers, and, despite the reconstruction advantages of $\ell_1$ regulariz

Alberto Cerezo

We construct a countable collection of one-parameter families of non-rotational minimal annuli with free boundary in geodesic balls of hyperbolic 3-space. Every surface within a given family shares a common prismatic symmetry group, and they appear as bifurcations from certain free boundary hyperbolic catenoids.

Alireza Gharahighehi, Achilleas Ghinis, Michela Venturini, Frederik Cornillie

Massive Open Online Courses (MOOCs) are emerging as a popular alternative to traditional education, offering learners the flexibility to access a wide range of courses from various disciplines, anytime and anywhere. Despite this accessibility, a significant number of enrollments in MOOCs result in dropouts. To enhance learner engagement, it is crucial to rec

Volodymyr Denysiuk, Lydmila Rybachuk

Finite trigonometric Fourier series on a set of discrete equidistant points are considered. A finite system of orthogonal functions that have interpolation and certain differential properties on the period is introduced. Finite Fourier series based on this system of functions form a class of trigonometric splines, which includes polynomial periodic simple sp

Jinghao Feng, Qiaoyu Zheng, Chaoyi Wu, Ziheng Zhao

Agentic AI systems have gained significant attention for their ability to autonomously perform complex tasks. However, their reliance on well-prepared tools limits their applicability in the medical domain, which requires to train specialized models. In this paper, we make three contributions: (i) We present M3Builder, a novel multi-agent system designed to

Hong Deng, Yu Zhang, Xiangfeng Chen, Wim Bogaerts

Electro-optical modulators are essential components in optical communication systems. They encode an electrical waveform onto an optical carrier. However, their performance is often limited by inherent electro-optic processes and imperfections in existing integrated designs, which limit their adaptability to diverse applications. This paper presents a circui

Esam Ghaleb, Bulat Khaertdinov, Aslı Özyürek, Raquel Fernández

In face-to-face interaction, we use multiple modalities, including speech and gestures, to communicate information and resolve references to objects. However, how representational co-speech gestures refer to objects remains understudied from a computational perspective. In this work, we address this gap by introducing a multimodal reference resolution task c

Nathaniel Hoy, Theodora Koulouri

Fake news poses global risks by influencing elections and spreading misinformation, making detection critical. Existing NLP and supervised Machine Learning methods perform well under cross-validation but struggle to generalise across datasets, even within the same domain. This issue stems from coarsely labelled training data, where articles are labelled base

Björn Altenburger, Joachim Fritzsche, Christoph Langhammer

UV-Vis spectroscopy is a workhorse in analytical chemistry that finds application in life science, organic synthesis and energy technologies like photocatalysis. In its traditional implementation with cuvettes, it requires sample volumes in the milliliter range. Here, we show how Nanofluidic Scattering Spectroscopy, NSS, which measures visible light scattere

Virgile Guémard, Gilles Zémor

We introduce new families of quantum Tanner codes, a class of quantum codes that first appeared in the work of Leverrier and Z\'emor (FOCS 2022). These codes are built from two classical Tanner codes, for which the underlying graphs are extracted from coverings of 2D geometrical complexes, and the local linear codes are tensor-products of cyclic or double-ci

Dipankar Das, Miguel Levy, Anugrah M. Prasad

We study a variant of the 3HDM, referred to as the BGL-3HDM, incorporating a $U(1)_1\times U(1)_2$ symmetry, which can distinguish the primary sources of mass for different fermion generations. In the version considered here, the Yukawa matrices in the down-quark and charged lepton sectors are diagonal, thereby eliminating tree-level FCNCs in these sectors.

Hiro J Farre-Kaga, Andrew Rothstein, Rohit Sonker, SangKyeun Kim

AI models have demonstrated strong predictive capabilities for various tokamak instabilities--including tearing modes (TM), ELMs, and disruptive event--but their opaque nature raises concerns about safety and trustworthiness when applied to fusion power plants. Here, we present a physics-based interpretation framework using a TM prediction model as a first d

Zahiriddin Rustamov, Ayham Zaitouny, Nazar Zaki

Instance selection (IS) addresses the critical challenge of reducing dataset size while keeping informative characteristics, becoming increasingly important as datasets grow to millions of instances. Current IS methods often struggle with capturing complex relationships in high-dimensional spaces and scale with large datasets. This paper introduces a graph a

Kyle Stein, Arash Mahyari, Guillermo Francia, Eman El-Sheikh

Vision-Language Models (VLMs) have demonstrated impressive multimodal capabilities in learning joint representations of visual and textual data, making them powerful tools for tasks such as Compositional Zero-Shot Learning (CZSL). CZSL requires models to generalize to novel combinations of visual primitives--such as attributes and objects--that were not expl

Tarannum Shaila Zaman, Macharla Hemanth Kishan, Lutfun Nahar Lota

Reproducing concurrency bugs is a complex task due to their unpredictable behavior. Researchers, regardless of gender, are contributing to automating this complex task to aid software developers. While some studies have investigated gender roles in the broader software industry, limited research exists on gender representation specifically among researchers

Federico Dell'Anna, Rafael Gomez-Lurbe, Armando Perez, Elisa Ercolessi

We investigate the performance of the Quantum Natural Gradient (QNG) optimizer in the presence of noise. Specifically, we evaluate the efficacy of QNG within the Quantum Approximate Optimization Algorithm (QAOA) for finding the ground state of the Transverse Field Ising Model (TFIM). Its performance is benchmarked against the Vanilla Gradient Descent optimiz

Efthymios Karangelos, Louis Wehenkel

This paper concerns the security of the electric power transmission grid facing the threat of malicious cyber-physical attackers. We posit that there is no such thing as perfectly effective cyber-security. Rather, any cyber-security measure comes with the possibility that a highly skilled attacker could (eventually find a way to) bypass it. On these grounds,

Zhiyu Kang, Raghavendra B. Rao, Eric F. Lock

In biomedical research and other fields, it is now common to generate high content data that are both multi-source and multi-way. Multi-source data are collected from different high-throughput technologies while multi-way data are collected over multiple dimensions, yielding multiple tensor arrays. Integrative analysis of these data sets is needed, e.g., to

Catherine Yu-Chi Chen, Jingyan Shen, Zhun Deng, Lihua Lei

Recent developments in large language models (LLMs) have led to their widespread usage for various tasks. The prevalence of LLMs in society implores the assurance on the reliability of their performance. In particular, risk-sensitive applications demand meticulous attention to unexpectedly poor outcomes, i.e., tail events, for instance, toxic answers, humili

Shenghui Chen, Yunhao Yang, Kayla Boggess, Seongkook Heo

Large Language Models (LLMs) are increasingly used for planning tasks, offering unique capabilities not found in classical planners such as generating explanations and iterative refinement. However, trust--a critical factor in the adoption of planning systems--remains underexplored in the context of LLM-based planning tasks. This study bridges this gap by co

Grant Z. X. Yang, Zi-Ting Sun, Ying-Ming Xie, K. T. Law

Planar Josephson junctions are pivotal for engineering topological superconductivity, yet are severely hindered by orbital effects induced by in-plane magnetic fields. In this work, we introduce the generic topological altermagnetic Josephson junctions (TAJJs) by leveraging the intrinsic spin-polarized band splitting and zero net magnetization attributes of

William K. Black, David Neilsen, Eric W. Hirschmann, David F. Van Komen

Adaptive mesh refinement efficiently facilitates the computation of gravitational waveforms in numerical relativity. However, determining precisely when, where, and to what extent to refine when solving the Einstein equations poses challenges; several ad hoc refinement criteria have been explored in the literature. This work introduces an optimized resolutio

Lucy D'Agostino McGowan, Shannon Tass, Sam Tyner, HaiYing Wang

The evolving focus in statistics and data science education highlights the growing importance of computing. This paper presents the Data Jamboree, a live event that combines computational methods with traditional statistical techniques to address real-world data science problems. Participants, ranging from novices to experienced users, followed workshop lead

Jeff D. Eldredge, Hanieh Mousavi

Many applications in aerodynamics, particularly in closed-loop control, depend on sensors to estimate the evolving state of the flow. This estimation task is inherently accompanied by uncertainty due to the noisy measurements of sensors or the non-uniqueness of the underlying mapping. Knowledge of this uncertainty can be as important for decision-making as t

Mia Gerber, Anna Sergeevna Bosman, Johan Pieter de Villiers

Automated machine learning (AutoML) is a research area focusing on using optimisation techniques to design machine learning (ML) algorithms, alleviating the need for a human to perform manual algorithm design. Real-time AutoML enables the design process to happen while the ML algorithm is being applied to a task. Real-time AutoML is an emerging research area

Palawat Busaranuvong, Emmanuel Agu, Reza Saadati Fard, Deepak Kumar

Infections in Diabetic Foot Ulcers (DFUs) can cause severe complications, including tissue death and limb amputation, highlighting the need for accurate, timely diagnosis. Previous machine learning methods have focused on identifying infections by analyzing wound images alone, without utilizing additional metadata such as medical notes. In this study, we aim

G. De Vecchi, M. Buzzi, G. Jotzu, S. Fava

Photoexcited K$_3$C$_{60}$ displays several properties reminiscent of equilibrium superconductivity, including transient optical spectra, pressure dependence, and I-V characteristics. However, these observations do not decisively establish non-equilibrium superconductivity, which would be conclusively evidenced by transient Meissner diamagnetism, as shown re

Jiahe Wang, Yan Wu, Yuke Hou, Yang Li

Cancer is a complex disease driven by dynamic regulatory shifts that cannot be fully captured by individual molecular profiling. We employ a data-driven approach to construct a coarse-grained dynamic network model based on hallmark interactions, integrating stochastic differential equations with gene regulatory network data to explore key macroscopic dynamic

Mengjiao Guo, Zhe Zhang, Ronggang Ping, Jianbin Jiao

This study introduces a complete joint angular distribution analysis for the process $e^{+}e^{-}\to J/\psi\to B(\to B_1\pi)\bar{B}(\to \bar{B}_1\pi)$ with polarized beams, where $B$ and $B_{1}$ are hyperons. We consider both transverse and longitudinal beam polarization and analyze $CP$ violation in the production and decay of hyperons. We present the comple

Varshini Reddy, Craig W. Schmidt, Yuval Pinter, Chris Tanner

Tokenization, a crucial initial step in natural language processing, is governed by several key parameters, such as the tokenization algorithm, vocabulary size, pre-tokenization strategy, inference strategy, and training data corpus. This paper investigates the impact of an often-overlooked hyperparameter, tokenizer training data size. We train BPE, UnigramL

Qingsen Yan, Yixu Feng, Cheng Zhang, Guansong Pang

Low-Light Image Enhancement (LLIE) is a crucial computer vision task that aims to restore detailed visual information from corrupted low-light images. Many existing LLIE methods are based on standard RGB (sRGB) space, which often produce color bias and brightness artifacts due to inherent high color sensitivity in sRGB. While converting the images using Hue,

Finn Orson Koepke

Let $(X, \mathcal{F})$ be a hypergraph. The Maker-Breaker game on $(X, \mathcal{F})$ is a combinatorial game between two players, Maker and Breaker. Beginning with Maker, the players take turns claiming vertices from $X$ that have not yet been claimed. Maker wins if she manages to claim all vertices of some hyperedge $F \in \mathcal{F}$. Breaker wins if he c

Felipe Castañeda-Ramírez, Moisés Martínez-Mares

Quantum transport through left-right symmetric chaotic cavities in the presence of the symplectic symmetry, is studied through the statistical distribution of the dimensionless conductance. With this particular point symmetry, their associated scattering matrices are blocky diagonalized by a rotation by an angle of $\pi/4$. Although the formulation is establ

Lukas Bödeker, Luc J. B. Kusters, Markus Müller

Neural-network (NN) based decoders are becoming increasingly popular in the field of quantum error correction (QEC), including for decoding of state-of-the-art quantum computation experiments. In this work, we make use of established interpretability methods from the field of machine learning, to introduce a toolbox to achieve an understanding of the underly

Davor Vukadin, Marin Šilić, Goran Delač

Large Language Models (LLMs) have demonstrated remarkable performance across diverse domains. However, effectively leveraging their vast knowledge for training smaller downstream models remains an open challenge, especially in domains like tabular data learning, where simpler models are often preferred due to interpretability and efficiency. In this paper, w