May 2025 arXiv papers — page 118

Jean Tapie, Matteo Nerini, Bruno Clerckx, Philipp del Hougne

We present the first experimental prototype of a reflective beyond-diagonal reconfigurable intelligent surface (BD-RIS), i.e., a RIS with reconfigurable inter-element connections. Our BD-RIS consists of an antenna array whose ports are terminated by a tunable load network. The latter can terminate each antenna port with three distinct individual loads or con

Mikołaj Małkiński, Jacek Mańdziuk

The abstract visual reasoning (AVR) domain presents a diverse suite of analogy-based tasks devoted to studying model generalization. Recent years have brought dynamic progress in the field, particularly in i.i.d. scenarios, in which models are trained and evaluated on the same data distributions. Nevertheless, o.o.d. setups that assess model generalization t

Chunlei Li, Peng Yu, Tiantian Liu, Siyuan Yu

We introduce a novel Unsmoothed Aggregation (UA) Algebraic Multigrid (AMG) method combined with Preconditioned Conjugate Gradient (PCG) to overcome the limitations of Extended Position-Based Dynamics (XPBD) in high-resolution and high-stiffness simulations. While XPBD excels in simulating deformable objects due to its speed and simplicity, its nonlinear Gaus

Peiyuan Zhang, Yongqi Chen, Haofeng Huang, Will Lin

Scaling video diffusion transformers (DiTs) is limited by their quadratic 3D attention, even though most of the attention mass concentrates on a small subset of positions. We turn this observation into VSA, a trainable, hardware-efficient sparse attention that replaces full attention at \emph{both} training and inference. In VSA, a lightweight coarse stage p

David Anugraha, Zilu Tang, Lester James V. Miranda, Hanyang Zhao

Reward models are essential for aligning language model outputs with human preferences, yet existing approaches often lack both controllability and interpretability. These models are typically optimized for narrow objectives, limiting their generalizability to broader downstream tasks. Moreover, their scalar outputs are difficult to interpret without context

Debesh Bhattacharjee, Prasad Subramanian, Saikat majumder, Wageesh Mishra

As solar coronal mass ejections (CMEs) propagate through the heliosphere, they expend energy in heating protons to compensate for the cooling that occurs due to expansion. CME propagation models usually treat energy dissipation implicitly via a polytropic index ($\delta$). Here we calculate the power dissipation implied by a given $\delta$ and compare it wit

D. Rosales Herrera, J. R. Alvarado García, A. Fernández Téllez, E. Cuautle

In this paper, we conduct a data-driven study of the production of softened hadrons and their contribution to the transverse momentum spectrum. To this end, we assume that the production of charged particles at soft and hard scales fundamentally results from the fragmentation of color strings. We analyze the $p_\text{T}$-spectrum data from pp to AA collision

Timofei Leahu, Giorgio Picci

Empirical Bayes inference is based on estimation of the parameters of an a priori distribution from the observed data. The estimation technique of the parameters of the prior, called hyperparameters, is based on the marginal distribution obtained by integrating the joint density of the model with respect to the prior. This is a key step which needs to be pro

Hubert Lacoin

When $d\ge 3$, the directed polymer a in random environment on $\mathbb Z^d$ is known to display a phase transition from a diffusive phase, known as \textit{weak disorder} to a localized phase, referred to as \textit{strong disorder}. This transition is encoded by the behavior of the the free energy of the model, defined by $$\mathfrak f(\beta):=\lim_{N\to \

Mak Ahmad, Prerna Ravi, David Karger, Marc Facciotti

Providing personalized, detailed feedback at scale in large undergraduate STEM courses remains a persistent challenge. We present an empirically evaluated practice exam system that integrates AI generated feedback with targeted textbook references, deployed in a large introductory biology course. Our system encourages metacognitive behavior by asking student

Nam V. Nguyen, Huy Nguyen, Quang Pham, Van Nguyen

Sparse mixture of experts (SMoE) offers an appealing solution to scale up the model complexity beyond the mean of increasing the network's depth or width. However, we argue that effective SMoE training remains challenging because of the suboptimal routing process where experts that perform computation do not directly contribute to the routing process. In thi

Gongfan Fang, Xinyin Ma, Xinchao Wang

Reasoning Language Models, capable of extended chain-of-thought reasoning, have demonstrated remarkable performance on tasks requiring complex logical inference. However, applying elaborate reasoning for all queries often results in substantial computational inefficiencies, particularly when many problems admit straightforward solutions. This motivates an op

Mark A. Scheel, Michael Boyle, Keefe Mitman, Nils Deppe

We present a major update to the Simulating eXtreme Spacetimes (SXS) Collaboration's catalog of binary black hole simulations. Using highly efficient spectral methods implemented in the Spectral Einstein Code (SpEC), we have nearly doubled the total number of binary configurations from 2,018 to 3,756. The catalog now densely covers the parameter space with p

Yasi Zhang, Tianyu Chen, Zhendong Wang, Ying Nian Wu

Learning generative models directly from corrupted observations is a long standing challenge across natural and scientific domains. We introduce Restoration Score Distillation (RSD), a unified framework for learning high fidelity, one step generative models using only degraded data and the mapping $A$ may be the identity or a non invertible corruption operat

Dan BW Choe, Sundhar Vinodh Sangeetha, Steven Emanuel, Chih-Yuan Chiu

Increased robot deployment, such as in warehousing, has revealed a need for seamless collaboration among heterogeneous robot teams to resolve unforeseen conflicts. To address this challenge, we propose a novel, decentralized framework for robots to request and provide help. The framework begins with robots detecting conflicts using a Vision Language Model (V

Christopher Kolloff, Tobias Höppe, Emmanouil Angelis, Mathias Jacob Schreiner

We propose a regularization framework inspired by thermodynamic work for guiding pre-trained probability flow generative models (e.g., continuous normalizing flows or diffusion models) by minimizing excess work, a concept rooted in statistical mechanics and with strong conceptual connections to optimal transport. Our approach enables efficient guidance in sp

Kunal Bahuguna, Ramesh Kolluru, S. V. Raghurama Rao

This paper presents a novel structure-preserving scheme for Euler equations, focusing on the numerical conservation of entropy and kinetic energy. Explicit flux functions engineered to conserve entropy are introduced within the finite-volume framework. Further, discrete kinetic energy conservation too is introduced. A systematic inquiry is presented, commenc

Irene Brugnara, Alessandro Valentini, Andrea Micheli

Recent work investigated the use of Reinforcement Learning (RL) for the synthesis of heuristic guidance to improve the performance of temporal planners when a domain is fixed and a set of training problems (not plans) is given. The idea is to extract a heuristic from the value function of a particular (possibly infinite-state) MDP constructed over the traini

Marcelo Campos, Matthew Jenssen, Marcus Michelen, Julian Sahasrabudhe

We prove a new lower bound for the off-diagonal Ramsey numbers, \[ R(3,k) \geq \bigg( \frac{1}{3}+ o(1) \bigg) \frac{k^2}{\log k }\, , \] thereby narrowing the gap between the upper and lower bounds to a factor of $3+o(1)$. This improves the best known lower bound of $(1/4+o(1))k^2/\log k$ due, independently, to Bohman and Keevash, and Fiz Pontiveros, Griffi

Miguel de Carvalho, Clemente Ferrer, Ronny Vallejos

This paper addresses the growing concern of cascading extreme events, such as an extreme earthquake followed by a tsunami, by presenting a novel method for risk assessment focused on these domino effects. The proposed approach develops an extreme value theory framework within a Kolmogorov-Arnold network (KAN) to estimate the probability of one extreme event

Dmitrii Korikov, Alexey Kokotov

Let $X$ be a Riemann surface of genus $g\ge 1$ endowed with a flat conical metric $m$ and let ${\rm det}\,\Delta$ be the $\zeta$-regularized determinant of the Friedrichs Laplacian on $(X,m)$. We derive variational formulas for ${\rm det}\,\Delta$ with respect to conical points and conical angles within a given conformal class. Integration of them leads to a

Gianni Gagliardo, Christian Saemann, Roberto Tellez-Dominguez

We explore the notion of an adjusted connection for principal 3-bundles. We first derive the explicit form of an adjustment datum for 3-term $L_\infty$-algebras, which allows us to give a local description of such adjusted connections and their infinitesimal symmetries. We then integrate the corresponding action Lie 3-algebroid to an action Lie 3-groupoid, e

Carsten Henkel

The Bogoliubov-de Gennes equations are solved for an inhomogeneous condensate in the vicinity of a turning point, addressing the full continuous spectrum. A basis change in the space of the two Bogoliubov "particle" and "hole" amplitudes is introduced that decouples them approximately. We find a spatially extended mode that governs mainly excitations in the

Sarah Chehade, Andrea Delgado, Elaine Wong

Variational quantum algorithms (VQAs) offer a promising near-term approach to finding optimal quantum strategies for playing non-local games. These games test quantum correlations beyond classical limits and enable entanglement verification. In this work, we present a variational framework for the Magic Square Game (MSG), a two-player non-local game with per

Haimi Nguyen, Arkajit Mandal, Ankit Mahajan, David R. Reichman

In this work, using two distinct semiclassical approaches, namely the mean-field Ehrenfest (MFE) method and the mapping approach to surface hopping (MASH), we investigate the spectral function of a single charge interacting with phonons on a lattice. This quantity is relevant for the description of angle-resolved photoemission experiments. Focusing on the on

Giacomo Aletti, Irene Crimaldi, Andrea Ghiglietti, Federico Nutarelli

Innovation is cumulative and interdependent: successful inventions build on prior knowledge within technological fields and may also affect success across related ones. Yet these dimensions are often studied separately in the innovation literature. This paper asks whether patent success across technological categories can be represented within a single dynam

Adrian Pacheco-Pozo, Igor M. Sokolov, Ralf Metzler, Diego Krapf

Diffusion in heterogeneous energy and diffusivity landscapes is widespread in biological systems. However, solving the Langevin equation in such environments introduces ambiguity due to the interpretation parameter $\alpha$, which depends on the underlying physics and can take values in the range $0<\alpha<1$. The typical interpretations are It\^o ($\alpha=0

Javad Forough, Hamed Haddadi

Membership Inference Attacks (MIAs) expose privacy risks by determining whether a specific sample was part of a model's training set. These threats are especially serious in sensitive domains such as healthcare and finance. Traditional mitigation techniques, such as static differential privacy, rely on injecting a fixed amount of noise during training or inf

Kyle Cheng, Udathari Kumarasinghe, Cristian Staii

Despite significant advances in understanding neuronal development, a fully quantitative framework that integrates intracellular mechanisms with environmental cues during axonal growth remains incomplete. Here, we present a unified biophysical model that captures key mechanochemical processes governing axonal extension on micropatterned substrates. In these

Chenyang Yang, Yike Shi, Qianou Ma, Michael Xieyang Liu

Prompt underspecification is a common challenge when interacting with LLMs. In this paper, we present an in-depth analysis of this problem, showing that while LLMs can often infer unspecified requirements by default (41.1%), such behavior is fragile: Under-specified prompts are 2x as likely to regress across model or prompt changes, sometimes with accuracy d

Dipanjan Mitra, Rahul Basu, George I. Melikidze

Polarization behaviour shows a transition in the pulsar population, where energetic sources with higher spin-down energy loss, $\dot{E} > 10^{34}$ erg~s$^{-1}$, often have fractional linear polarisation ($L/I$) close to 100\%, while below this range $L/I$ is usually lower than 50\%. The polarisation behaviour has been primarily studied at higher frequencies

Nimrod Berman, Ilan Naiman, Moshe Eliasof, Hedi Zisling

Diffusion-based generative models have demonstrated exceptional performance, yet their iterative sampling procedures remain computationally expensive. A prominent strategy to mitigate this cost is distillation, with offline distillation offering particular advantages in terms of efficiency, modularity, and flexibility. In this work, we identify two key obser

Rebecca Pelke, Nils Bosbach, Lennart M. Reimann, Rainer Leupers

Accelerating Machine Learning (ML) workloads requires efficient methods due to their large optimization space. Autotuning has emerged as an effective approach for systematically evaluating variations of implementations. Traditionally, autotuning requires the workloads to be executed on the target hardware (HW). We present an interface that allows executing a

Djallel Bouneffouf, Raphael Feraud

Bandit algorithms and Large Language Models (LLMs) have emerged as powerful tools in artificial intelligence, each addressing distinct yet complementary challenges in decision-making and natural language processing. This survey explores the synergistic potential between these two fields, highlighting how bandit algorithms can enhance the performance of LLMs

Owen Lee, Kenneth Joseph

This paper presents the first large-scale analysis of public-facing chatbots on Character$.$AI, a rapidly growing social media platform where users create and interact with chatbots. Character$.$AI is distinctive in that it merges generative AI with user-generated content, enabling users to build bots for others to engage with. It is also popular, with over

Prakash Timsina, Andres Chappa, Deema Alyones, Igor Vasiliev

Frustrated magnets provide a platform for exploring exotic phases beyond conventional ordering, with potential relevance to functional materials and information technologies. In this work, we use Monte Carlo simulations to map the thermodynamic phase diagram of pyrochlore iridates R2Ir2O7 (R = Dy, Ho) with three stable magnetic ground-state stable phases: fr

Tomasz Goliński, Grzegorz Jakimowicz

We construct the linear Poisson bracket on the predual bundle of a Banach Lie algebroid. It is an alternative approach to the already known results on the linear sub-Poisson structure on the dual bundle. We also discuss the existence of queer Banach Lie algebroids. The example on a trivial bundle is presented and the situation on precotangent bundles is disc

Marcelo Botta Cantcheff

In holographic quantum gravity, Euclidean pieces of the spacetime appear in the large N limit as representing semi-classical states of the theory. In this essay, we argue that the duals of entangled states are spacetime geometries that contain Euclidean regions in order to preserve classical connectivity. Thereby, the proposal is to extend the ER-EPR conject

Narek Maloyan, Bislan Ashinov, Dmitry Namiot

Large Language Models (LLMs) are increasingly employed as evaluators (LLM-as-a-Judge) for assessing the quality of machine-generated text. This paradigm offers scalability and cost-effectiveness compared to human annotation. However, the reliability and security of such systems, particularly their robustness against adversarial manipulations, remain critical

Carsten Dietzel

In this work, we address a question posed by Dehornoy et al. in the book "Foundations of Garside Theory" that asks for a theory of groups of $\mathrm{I}_G$-type when $G$ is a Garside group. In this article, we introduce a broader notion than the one suggested by Dehornoy et al.: given a left-ordered group $G$, we define a group of $\mathrm{I}_G$-type as a le

Austin Xu, Yilun Zhou, Xuan-Phi Nguyen, Caiming Xiong

To keep pace with the increasing pace of large language models (LLM) development, model output evaluation has transitioned away from time-consuming human evaluation to automatic evaluation, where LLMs themselves are tasked with assessing and critiquing other model outputs. LLM-as-judge models are a class of generative evaluators that excel in evaluating rela

Shuqing Luo, Pingzhi Li, Jie Peng, Hanrui Wang

Mixture-of-experts (MoE) architectures could achieve impressive computational efficiency with expert parallelism, which relies heavily on all-to-all communication across devices. Unfortunately, such communication overhead typically constitutes a significant portion of the total runtime, hampering the scalability of distributed training and inference for mode

Ahmet Berke Gokmen, Yigit Ekin, Bahri Batuhan Bilecen, Aysegul Dundar

We propose RoPECraft, a training-free video motion transfer method for diffusion transformers that operates solely by modifying their rotary positional embeddings (RoPE). We first extract dense optical flow from a reference video, and utilize the resulting motion offsets to warp the complex-exponential tensors of RoPE, effectively encoding motion into the ge

Andrej Čop, Blaž Bertalanič, Marko Grobelnik, Carolina Fortuna

As the complexity and number of machine learning (ML) models grows, well-documented ML models are essential for developers and companies to use or adapt them to their specific use cases. Model metadata, already present in unstructured format as model cards in online repositories such as Hugging Face, could be more structured and machine readable while also i

Yuval Grinberg, Nimrod Harel, Jacob Goldberger, Ofir Lindenbaum

Falsely annotated samples, also known as noisy labels, can significantly harm the performance of deep learning models. Two main approaches for learning with noisy labels are global noise estimation and data filtering. Global noise estimation approximates the noise across the entire dataset using a noise transition matrix, but it can unnecessarily adjust corr

Ashish Gurung, Jionghao Lin, Zhongtian Huang, Conrad Borchers

Prior work has developed a range of automated measures ("detectors") of student self-regulation and engagement from student log data. These measures have been successfully used to make discoveries about student learning. Here, we extend this line of research to an underexplored aspect of self-regulation: students' decisions about when to start and stop worki

Hermine Biermé, Olivier Durieu, Donatas Surgailis

We study the limit distribution of the volume fraction estimator $\widehat p_{λ, A}$ (= the Lebesgue measure of the intersection $\mathcal{X}\cap (λA)$ of a random set $\mathcal{X}$ with a large observation set $λA$, divided by the Lebesgue measure of $λA$), as $λ\to \infty$, for a Boolean set $\mathcal{X}$ formed by uniformly scattered random grains $Ξ\subs

Jia-Hui Pan, Yeok Tatt Cheah, Zhengzhe Liu, Ka-Hei Hui

Robotic bin packing aids in a wide range of real-world scenarios such as e-commerce and warehouses. Yet, existing works focus mainly on considering the shape of objects to optimize packing compactness and neglect object properties such as fragility, edibility, and chemistry that humans typically consider when packing objects. This paper presents OPA-Pack (Ob

Qiongqiong Wang, Hardik B. Sailor, Tianchi Liu, Ai Ti Aw

Current speech-LLMs exhibit limited capability in contextual reasoning alongside paralinguistic understanding, primarily due to the lack of Question-Answer (QA) datasets that cover both aspects. We propose a novel framework for dataset generation from in-the-wild speech data, that integrates contextual reasoning with paralinguistic information. It consists o

Babak Badnava, Jacob Chakareski, Morteza Hashemi

We investigate multitask edge-user communication-computation resource allocation for $360^\circ$ video streaming in an edge-computing enabled millimeter wave (mmWave) multi-user virtual reality system. To balance the communication-computation trade-offs that arise herein, we formulate a video quality maximization problem that integrates interdependent multit

Julia Henninger, Sebastian Ohrem, Wolfgang Reichel

We prove existence of real-valued, time-periodic and spatially localized solutions (breathers) of semilinear wave equations $V(x)u_{tt} - u_{xx} = \Gamma(x) |u|^{p-1} u$ on $\mathbb{R}^2$ for all values of $p\in (1,\infty)$. Using tools from the calculus of variations our main result provides breathers as ground states of an indefinite functional under suita

Liam A. Kruse, Marc R. Schlichting, Mykel J. Kochenderfer

Importance sampling is a Monte Carlo technique for efficiently estimating the likelihood of rare events by biasing the sampling distribution towards the rare event of interest. By drawing weighted samples from a learned proposal distribution, importance sampling allows for more sample-efficient estimation of rare events or tails of distributions. A common ch

Ho-Chun Herbert Chang

Measuring social influence is difficult due to the lack of counter-factuals and comparisons. By combining machine learning-based modeling and network science, we present general properties of social value, a recent measure for social influence using synthetic control applicable to political behavior. Social value diverges from centrality measures on in that

A. Martínez-Henares, I. Jiménez-Serra, C. Vastel, T. Sakai

During the early stages of star formation, accretion processes such as infall from the envelope and molecular streamers, and ejection of matter through winds and jets take place simultaneously. The Class 0/I binary [BHB2007] 11 shows evidence for accretion and ejection at the scales of the circumbinary disk and the inner close binary. Recent H$_2$CO observat

Dylan G. L. Allegretti, Hyun Kyu Kim, Peng Shan

We prove a conjecture of the first and third named authors relating the Kauffman bracket skein algebra of a genus zero surface with boundary to a quantized $K$-theoretic Coulomb branch. As a consequence, we see that our skein algebra arises as the Grothendieck ring of the bounded derived category of equivariant coherent sheaves on the Braverman-Finkelberg-Na

Babak Badnava, Jacob Chakareski, Morteza Hashemi

In virtual reality (VR) environments, computational tasks exhibit an elastic nature, meaning they can dynamically adjust based on various user and system constraints. This elasticity is essential for maintaining immersive experiences; however, it also introduces challenges for communication and computing in VR systems. In this paper, we investigate elastic t

Junhao Cai, Yuhan Li, Yunqiao Liu, Benlong Wang

In this work, we present a unified experimental and simulation investigation of cavitation in aqueous electrolyte solutions, combining nanosecond laser-induced optical breakdown and all-atom molecular dynamics (MD) simulations under tensile stress. Across both cavitation scenarios, we find that cavitation inception and intensity (bubble nucleation count, cav

A'nya Buckner, Sarah Lang, Rafael Loureiro

The integration of plant-based bioregenerative life support systems is a central objective in NASA's Moon to Mars strategy. Arabidopsis thaliana, a model organism with extensive genomic resources, serves as a key species to investigate plant resilience in extraterrestrial environments. We assessed the physiological and gene expression responses of A. thalian

Frédéric Berdoz, Dustin Brunner, Yann Vonlanthen, Roger Wattenhofer

Voting advice applications (VAAs) help millions of voters understand which political parties or candidates best align with their views. This paper explores the potential risks these applications pose to the democratic process when targeted by adversarial entities. In particular, we expose 11 manipulation strategies and measure their impact using data from Sw

Hongru Wang, Wenyu Huang, Yufei Wang, Yuanhao Xi

Existing benchmarks that assess Language Models (LMs) as Language Agents (LAs) for tool use primarily focus on stateless, single-turn interactions or partial evaluations, such as tool selection in a single turn, overlooking the inherent stateful nature of interactions in multi-turn applications. To fulfill this gap, we propose \texttt{DialogTool}, a multi-tu

Ajian Liu, Haocheng Yuan, Xiao Guo, Hui Ma

PAD and FFD are proposed to protect face data from physical media-based Presentation Attacks and digital editing-based DeepFakes, respectively. However, isolated training of these two models significantly increases vulnerability towards unknown attacks, burdening deployment environments. The lack of a Unified Face Attack Detection model to simultaneously han

Yuhang Wang, Youhe Jiang, Bin Cui, Fangcheng Fu

Recent advances in test-time scaling suggest that Large Language Models (LLMs) can gain better capabilities by generating Chain-of-Thought reasoning (analogous to human thinking) to respond a given request, and meanwhile exploring more reasoning branches (i.e., generating multiple responses and ensembling them) can improve the final output quality. However,

Kara Schechtman, Benjamin Brandon, Jenise Stafford, Hannah Li

In higher education, many institutions use algorithmic alerts to flag at-risk students and deliver advising at scale. While much research has focused on evaluating algorithmic predictions, relatively little is known about how discretionary interventions by human experts shape outcomes in algorithm-assisted settings. We study this question using rich quantita

Galit Shmueli, David Martens, Jaewon Yoo, Travis Greene

Counterfactuals play a pivotal role in the two distinct data science fields of causal inference (CI) and explainable artificial intelligence (XAI). While the core idea behind counterfactuals remains the same in both fields--the examination of what would have happened under different circumstances--there are key differences in how they are used and interprete

C P Haritha, Karthik Jain, Barilang Mawlong

The extension of the standard model to include a single scalar or vector leptoquark has been shown to account for the observed deviations in the lepton flavor universality ratios in the $b \to c \ell \nu_{\ell}$ and $b \to s \ell \ell$ transitions. Exploring new physics in the $b \to c \tau \nu_{\tau}$ decays, in this work we analyze the baryonic decay chann

Andrés Rubiano, Armando Reyes

We investigate the differential smoothness of bi-quadratic algebras with PBW basis.

William Good, Fei Yao, Huey-Wen Lin

We report the first nucleon gluon parton distribution function (PDF) using Large-Momentum Effective Theory (LaMET). We focus on the gluon operator which was demonstrated to have the best signal-to-noise in the previous attempt [1] in computing gluon PDFs using LaMET. We compute the corresponding Wilson coefficients needed for the hybrid-renormalized matrix e

Christopher A. Bowers, Cass T. Miller

Transport during flow of generalized Newtonian fluids (GNFs) appears often in systems that can be treated in a simplified form as either cylindrical tubes or slit openings between parallel plates. Based on the pioneering work of Taylor, analytical solutions for transport in these simplified systems were derived generally. This includes analytical solutions f

Tian Wen, Sheng Sun, Yuwei Wang, Peiyan Chen

Secure Aggregation (SA) is an indispensable component of Federated Learning (FL) that concentrates on privacy preservation while allowing for robust aggregation. However, most SA designs rely heavily on the unrealistic assumption of homogeneous model architectures. Federated Distillation (FD), which aggregates locally computed logits instead of model paramet

Paula Feldman, Martin Sinnona, Claudio Delrieux, Viviana Siless

Anatomical trees are critical for clinical diagnosis and treatment planning, yet their complex and diverse geometry make accurate representation a significant challenge. Motivated by the latest advances in large language models, we introduce an autoregressive method for synthesizing anatomical trees. Our approach first embeds vessel structures into a learned

Song-Lin Lv, Rui Zhu, Tong Wei, Yu-Feng Li

Semi-supervised learning (SSL) alleviates the cost of data labeling process by exploiting unlabeled data and has achieved promising results. Meanwhile, with the development of large foundation models, exploiting pre-trained models becomes a promising way to address the label scarcity in the downstream tasks, such as various parameter-efficient fine-tuning te

Gabriele Spadaro, Alberto Presta, Jhony H. Giraldo, Marco Grangetto

Efficient compression of low-bit-rate point clouds is critical for bandwidth-constrained applications. However, existing techniques mainly focus on high-fidelity reconstruction, requiring many bits for compression. This paper proposes a "Denoising Diffusion Probabilistic Model" (DDPM) architecture for point cloud compression (DDPM-PCC) at low bit-rates. A Po

Reza T. Batley, Sourav Saha

Contemporary models of high dimensional physical systems are constrained by the curse of dimensionality and a reliance on dense data. We introduce KHRONOS (Kernel Expansion Hierarchy for Reduced Order, Neural Optimized Surrogates), an AI framework for model based, model free and model inversion tasks. KHRONOS constructs continuously differentiable target fie

Guangkun Zhang, Wen Fang, Mingliang Xiong, Qingwen Liu

High-precision positioning in GPS-denied environments is a demanding but challenging technology. Resonant Beam Positioning (RBP) utilizes a resonant beam with properties such as energy focusing, self-establishment, self-alignment, and passive operation, offering a promising solution for this task. However, traditional RBP algorithms require a fixed number of

James MacDonald

Although most white dwarf stars have hydrogen-dominated atmospheres, a significant fraction have atmospheres in which hydrogen is spectroscopically absent, with the fraction of hydrogen-free atmospheres varying with effective temperature. Estimates of the total mass of hydrogen, MH, in the stellar envelope from either asteroseismology or spectral evolution a

Zhijie Deng, Chris Yuhao Liu, Zirui Pang, Xinlei He

Large Language Models (LLMs) have demonstrated strong capabilities in memorizing vast amounts of knowledge across diverse domains. However, the ability to selectively forget specific knowledge is critical for ensuring the safety and compliance of deployed models. Existing unlearning efforts typically fine-tune the model with resources such as forget data, re

Saleh Soudijani, Rayna Dimitrova

We study stochastic multi-agent systems in which agents must cooperate to maximize the probability of achieving a common reach-avoid objective. In many applications, during the execution of the system, the communication between the agents can be constrained by restrictions on the bandwidth currently available for exchanging local-state information between th

Mohammad Shahmoradi, Korkut Kaan Tokgöz, Eduard Alarcón, Sergi Abadal

Wireless Network-on-Chip (WNoC) systems, which wirelessly interconnect the chips of a computing system, have been proposed as a complement to existing chip-to-chip wired links. However, their feasibility depends on the availability of custom-designed high-speed, tiny, ultra-efficient transceivers. This represents a challenge due to the tradeoffs between band

Jad Mansour, Sebastian Realpe, Hayat Rajani, Michele Grimaldi

Event-based cameras (EBCs) are poised to transform underwater robotics, yet the absence of labelled event-based datasets for underwater environments severely limits progress in tasks such as visual odometry and obstacle avoidance. Real-world event-based optical flow datasets are scarce, resource-intensive to collect, and lack diversity, while no prior benchm

Hengli Li, Chenxi Li, Tong Wu, Xuekai Zhu

Reasoning ability, a core component of human intelligence, continues to pose a significant challenge for Large Language Models (LLMs) in the pursuit of AGI. Although model performance has improved under the training scaling law, significant challenges remain, particularly with respect to training algorithms, such as catastrophic forgetting, and the limited a

Qiguang Chen, Libo Qin, Jinhao Liu, Yue Liao

Chain-of-Thought (CoT) reasoning has proven effective in enhancing large language models (LLMs) on complex tasks, spurring research into its underlying mechanisms. However, two primary challenges remain for real-world applications: (1) the lack of quantitative metrics and actionable guidelines for evaluating and optimizing measurable boundaries of CoT capabi

Chengsong Sun, Weiping Li, Xiang Li, Yuankun Liu

Few-shot cross-modal retrieval focuses on learning cross-modal representations with limited training samples, enabling the model to handle unseen classes during inference. Unlike traditional cross-modal retrieval tasks, which assume that both training and testing data share the same class distribution, few-shot retrieval involves data with sparse representat

Nikyta Shchutskyi, Matthieu Schaller, Orestis A. Karapiperis, Federico A. Stasyszyn

Understanding the origin and evolution of magnetic fields on cosmological scales opens up a window into the physics of the early Universe. Numerical simulations of such fields require a careful treatment to faithfully solve the equations of magnetohydrodynamics (MHD) without introducing numerical artefacts. In this paper, we study the growth of the magnetic

Hector Roche Carrasco, Justin Schirmann, Aurelien Mordret, Adolfo G. Grushin

The strict geometric rules that define aperiodic tilings lead to the unique spectral and transport properties of quasicrystals, but also limit our ability to design them. In this Letter, we explore a novel example of a continuously tunable family of two-dimensional aperiodic tilings in which the underlying real-space geometry becomes a control knob of the wa

Nguyen Thi Thai Ha, Tran Nam Son, Pham Duy Vinh

We study whether a unital associative algebra $ A $ over a field admits a decomposition of the form $A = Z(A) + [A,A]$ where $ Z(A) $ is the center of $ A $ and $ [A,A] $ denotes the additive subgroup of $A$ generated by all additive commutators of $A$. Among our main considerations are the cases in which $A$ is the matrix ring over a division ring, a genera

Alice Plebe, Timothy Douglas, Diana Riazi, R. Maria del Rio-Chanona

As language and vision-language models (VLMs) become central to information access and online interaction, concerns grow about their potential to amplify misinformation. Human studies show that images boost the perceived credibility and shareability of information, raising the question of whether VLMs exhibit the same vulnerability. We present the first stud

Antonios Katsianis, Qingshan Wang, Xiaohu Yang, Xian Zhong Zheng

The $\Gamma$ growth model is an effective parameterization employed across various scientific disciplines and scales to depict growth. It has been demonstrated that the cosmic star formation rate density (CSFRD) can also be described broadly by this pattern, i.e. $\frac{dM(T)}{dT} = M_{z,0}\, \times \frac{\beta^{\alpha}}{\Gamma(\alpha)} \, T^{\alpha-1} e^{-\

Zekai Li, Xinhao Zhong, Samir Khaki, Zhiyuan Liang

In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance comparable to those trained on the original datasets. To further improve the performance of synthetic datasets, various training pipelines and optimization objectives have been propo

Likai Chen, Georg Keilbar, Wei Biao Wu

This paper considers the estimation of quantiles via a smoothed version of the stochastic gradient descent (SGD) algorithm. By smoothing the score function with a bandwidth tied to the learning rate, we obtain estimates that are monotone in the quantile level at every iteration, while retaining the memory and computational efficiency required for streaming d

Yu-Qi Dong, Xiao-Bin Lai, Yu-Zhi Fan, Yu-Xiao Liu

In this paper, we investigate the polarization modes of gravitational waves within the most general symmetric teleparallel gravity theory that allows for second-order field equations We consider both scenarios where test particles either carry or do not carry a hypermomentum charge. Our findings reveal the existence of tensor, vector, and scalar modes of gra

Kaitlyn Shin, Alice Curtin, Maxwell Fine, Ayush Pandhi

Among the thousands of observed fast radio bursts (FRBs), a few sources exhibit exceptionally high burst activity observable by many telescopes across a broad range of radio frequencies. Almost all of these highly active repeaters have been discovered by CHIME/FRB, due to its daily observations of the entire Northern sky as a transit radio telescope. FRB 202

R. Krivonos, R. Burenin, E. Filippova, I. Lapshov

The knowledge of the point spread function (PSF) of the Mikhail Pavlinsky Astronomical Roentgen Telescope - X-ray Concentrator (ART-XC) telescope aboard the Spectrum-Roentgen-Gamma (SRG) observatory plays an especially crucial role in the detection of point X-ray sources in the all-sky survey and the studies of extended X-ray objects with low surface brightn

Pengyu Liu, Matthew A. Kenworthy, Beth A. Biller, Alex Wallace

Directly imaged exoplanets in wide orbits challenge current gas giant formation theories. They need to form quickly and acquire enough material before the disk dissipates, which cannot be accommodated by in-situ formation by core accretion. We search for wide separation ($>$ 100 au) planetary-mass companions with the Young Suns Exoplanet Survey (YSES). Here,

Gabriel de Albuquerque Gleizer

We address the problem of identifying a system subject to additive faults, while simultaneously reconstructing the fault signal via subspace methods. We do not require nominal data for the identification, neither do we impose any assumption on the class of faults, e.g., sensor or actuator faults. We show that, under mild assumptions on the fault signal, stan

Anežka Dostálová, Dominik Vašinka, Robert Stárek, Miroslav Ježek

Molecular fluorescence microscopy is a leading approach to super-resolution and nanoscale imaging in life and material sciences. However, super-resolution fluorescence microscopy is often bottlenecked by system-specific calibrations and long acquisitions of sparsely blinking molecules. We present a deep-learning approach that reconstructs super-resolved imag

Huaiying Luo, Cheng Ji

In the age of cloud computing, data privacy protection has become a major challenge, especially when sharing sensitive data across cloud environments. However, how to optimize collaboration across cloud environments remains an unresolved problem. In this paper, we combine federated learning with large-scale language models to optimize the collaborative mecha

Yifu Cai, Xinyu Li, Mononito Goswami, Michał Wiliński

We introduce TimeSeriesGym, a scalable benchmarking framework for evaluating Artificial Intelligence (AI) agents on time series machine learning engineering challenges. Existing benchmarks lack scalability, focus narrowly on model building in well-defined settings, and evaluate only a limited set of research artifacts (e.g., CSV submission files). To make AI

Bastien Carreres, Rebecca C. Chen, Erik R. Peterson, Dan Scolnic

Measurement of the growth rate of structures ($\fsig$) with Type Ia supernovae (\sns) will improve our understanding of the nature of dark energy and enable tests of general relativity. In this paper, we generate simulations of the 10 year \sn\ dataset of the Rubin-LSST survey, including a correlated velocity field from a N-body simulation and realistic mode

Alonso Urbano, David W. Romero, Max Zimmer, Sebastian Pokutta

Real world data often exhibits unknown, instance-specific symmetries that rarely exactly match a transformation group $G$ fixed a priori. Class-pose decompositions aim to create disentangled representations by factoring inputs into invariant features and a pose $g\in G$ defined relative to a training-dependent, arbitrary canonical representation. We introduc

Hee Oh

For a vector $v=(v_1,\dots ,v_n)$ with $v_1>\cdots>v_n$ and $\sum v_i=0$, we study the "directional entropy" of two arithmetic objects: (1) the logarithmic embeddings of degree-$n$ totally real units, and (2) the logarithmic eigenvalue data of $\operatorname{SL}_n(\mathbb Z)$. In each case, the entropy in the direction of $v$ is $\mathsf E_n(v)= \rho_{\opera