arXiv papers from June 2025

Keun Soo Yim

Time series forecasting models have diverse real world applications (e.g., from electricity metrics to software workload). Latest foundational models trained for time series forecasting show strengths (e.g., for long sequences and in zero-shot settings). However, foundational model was not yet used for forecasting rare, spiky events, i.e., a challenging targ

Benedikt Jahnel, Lukas Lüchtrath, Anh Duc Vu

We investigate oriented bond-site percolation on the planar lattice in which entire columns are stretched. Generalising recent results by Hil\'ario et al., we establish non-trivial percolation under a $(1+\varepsilon)$-th moment condition on the stretches and use this to prove survival of contact processes with periodic recoveries as well as in random enviro

Xuan Liu, Yinhao Ren, Marc D. Ryser, Lars J. Grimm

Accurate lesion tracking in temporal mammograms is essential for monitoring breast cancer progression and facilitating early diagnosis. However, automated lesion correspondence across exams remains a challenges in computer-aided diagnosis (CAD) systems, limiting their effectiveness. We propose MammoTracker, a mask-guided lesion tracking framework that automa

Chuyan Zhang, Kefan Wang, Yun Gu

Low-Rank Adaptation (LoRA) has proven effective in reducing computational costs while maintaining performance comparable to fully fine-tuned foundation models across various tasks. However, its fixed low-rank structure restricts its adaptability in scenarios with substantial domain gaps, where higher ranks are often required to capture domain-specific comple

Matías Bruna, Alex Capuñay, Eduardo Friedman

We associate to a semisimple complex Lie algebra $\mathfrak{g}$ a sequence of polynomials $P_{\ell,\mathfrak{g}}(x)\in\mathbb{Q}[x]$ in $r$ variables, where $r$ is the rank of $\mathfrak{g}$ and $\ell=0,1,2,\ldots $. The polynomials $P_{\ell,\mathfrak{g}}(x)$ are uniquely associated to the isomorphism class of $\mathfrak{g}$, up to re-numbering the variables

Subham Bhakta, Igor E. Shparlinski

A discrete model of quantum ergodicity of linear maps generated by symplectic matrices $A \in \mathrm{Sp}(2d,\mathbb{Z})$ modulo an integer $N\ge 1$, has been studied for $d=1$ and almost all $N$ by P. Kurlberg and Z. Rudnick (2001). Their result has been strengthened by J. Bourgain (2005) and subsequently by A. Ostafe, I. E. Shparlinski, and J. F. Voloch (2

Hashim Ali, Surya Subramani, Raksha Varahamurthy, Nithin Adupa

Recent advances in speech synthesis have introduced unprecedented challenges in maintaining voice authenticity, particularly concerning public figures who are frequent targets of impersonation attacks. This paper presents a comprehensive methodology for collecting, curating, and generating synthetic speech data for political figures and a detailed analysis o

Rysa Greenwood, Bradley G. Guislain, MengXing Na, Alexandra B. Tully

Exciton lifetimes play a critical role in the performance of organic optoelectronic devices. In this work, we investigate how the presence of multiple rotational domains, and therefore grain boundaries, impacts exciton dynamics in thin films of C60/Au(111) using time and angle-resolved photoemission spectroscopy (TR-ARPES). We find that films with multiple r

Dharamveer Kumar, Amuthan A. Ramabathiran

The non-interacting kinetic energy functional, $T_{KS}(\rho)$, plays a fundamental role in Density Functional Theory (DFT), but its explicit form remains unknown for arbitrary $N$-representable densities. Although it can, in principle, be evaluated by solving a constrained optimization problem, the associated adjoint problem is not always well-posed; moreove

Christiana Westlin, Ashutosh Singh, Deniz Erdogmus, Georgios Stratis

In the science of emotion, it is widely assumed that folk emotion categories form a biological and psychological typology, and studies are routinely designed and analyzed to identify emotion-specific patterns. This approach shapes the observations that studies report, ultimately reinforcing the assumption that guided the investigation. Here, we reanalyzed da

Tanmay Vilas Samak, Chinmay Vilas Samak, Bing Li, Venkat Krovi

Simulation frameworks have been key enablers for the development and validation of autonomous driving systems. However, existing methods struggle to comprehensively address the autonomy-oriented requirements of balancing: (i) dynamical fidelity, (ii) photorealistic rendering, (iii) context-relevant scenario orchestration, and (iv) real-time performance. To a

Marko Mladenović, Manasa Kaniselvan, Christoph Weilenmann, Alexandros Emboras

Valence change memory (VCM) cells based on SrTiO$_3$ (STO), a perovskite oxide, are a promising type of emerging memory device. While the operational principle of most VCM cells relies on the growth and dissolution of one or multiple conductive filaments, those based on STO are known to exhibit a distinctive, 'interface-type' switching, which is associated w

Yunier Bello-Cruz, Roy Quintero-Contreras

In this paper, we investigate properties of the fixed point sequence of the Josephus function $J_3$. First, we establish a connection between this sequence and the Chinese Remainder Theorem. Next, we identify a clear numerical pattern for the digits of two consecutive fixed points when they are written in a non-standard fractional number system in base $3/2$

Siyou Li, Pengyao Qin, Huanan Wu, Dong Nie

Automated radiology report generation (RRG) aims to produce detailed textual reports from clinical imaging, such as computed tomography (CT) scans, to improve the accuracy and efficiency of diagnosis and provision of management advice. RRG is complicated by two key challenges: (1) inherent complexity in extracting relevant information from imaging data under

Travis Seth Rippentrop, Avijit Bera, Mustapha Ishak

The stability of thin shell wormholes and black holes to linearized spherically symmetric perturbations about a static equilibrium is analyzed. Thin shell formalism is explored and junctions formed from combinations of Schwarzschild, Schwarzschild - de Sitter, and Schwarzschild - anti-de Sitter, as well as Friedmann-Lemaitre-Robertson-Walker (FLRW) spacetime

Muhammed A. Dada, Sarah Pak, Matthew N. Ward, Megan Simons

The Tamm-Dancoff Approximation (TDA) offers a computationally efficient alternative to full linear-response Time-Dependent Density Functional Theory (TDDFT) for calculating electronic excited states, particularly in large molecular systems. By neglecting the coupling between excitation and de-excitation channels, TDA simplifies the TDDFT response equations i

Martin Balko, Anna Brötzner, Fabian Klute, Josef Tkadlec

We initiate the study of extremal problems about faces in convex rectilinear drawings of~$K_n$, that is, drawings where vertices are represented by points in the plane in convex position and edges by line segments between the points representing the end-vertices. We show that if a convex rectilinear drawing of $K_n$ does not contain a common interior point o

T. M. Crispim, Marcos V. de S. Silva, G. Alencar, Diego Sáez-Chillón Gómez

Black bounces are compact objects that combine the structures of regular black holes with those of wormholes. These spacetimes exhibit a rich causal structure and can differ fundamentally from usual black holes. In this work, we study the behavior of the tidal forces by considering different black bounce models. To this end, we start with the geodesic deviat

Dhruv Agarwal, Bodhisattwa Prasad Majumder, Reece Adamson, Megha Chakravorty

The promise of autonomous scientific discovery (ASD) hinges not only on answering questions, but also on knowing which questions to ask. Most recent works in ASD explore the use of large language models (LLMs) in goal-driven settings, relying on human-specified research questions to guide hypothesis generation. However, scientific discovery may be accelerate

Rin Ray

Suppose we are given a profinite group $G$ acting on a formal moduli stack $\mathcal{M}$, and we want to understand the group action, and compute cohomology related to this group action. How can we do it? This prolegomenon surveys two methods of pinning down such an action: geometric modeling and the two tower method. We highlight their use on a specific act

Lucas Barrault, Lisa Bugnet, Stéphane Mathis, Joey S. G. Mombarg

Gamma-Dor stars are ideal targets for studies of stellar innermost dynamical properties due to their rich asteroseismic spectrum of gravity modes. Integrating internal magnetism to the picture appears as the next milestone of detailed asteroseismic studies, for its prime importance on stellar evolution. The inertial dip in prograde dipole modes period-spacin

Chao Zhang, Neha Arora, Christopher Bian, Yechen Li

Estimating Origin-Destination (OD) travel demand is vital for effective urban planning and traffic management. Developing universally applicable OD estimation methodologies is significantly challenged by the pervasive scarcity of high-fidelity traffic data and the difficulty in obtaining city-specific prior OD estimates (or seed ODs), which are often prerequ

Deland Liu, Frigyes Samuel Racz, Zoe Lalji, Jose del R. Millan

Patients with amyotrophic lateral sclerosis (ALS) in the completely locked-in state (CLIS) can lose all reliable motor control and are left without any means of communication. It remains unknown whether non-invasive electroencephalogram (EEG) based brain-computer interfaces (BCIs) can support volitional communication in CLIS. Here, we show that a CLIS patien

Yujun Zhang, Runlong Li, Xiaoxiang Liang, Xinhao Yang

The abnormal fluctuations in network traffic may indicate potential security threats or system failures. Therefore, efficient network traffic prediction and anomaly detection methods are crucial for network security and traffic management. This paper proposes a novel network traffic prediction and anomaly detection model, MamNet, which integrates time-domain

Brayan Murgas, Avanish Mishra, Nithin Mathew, Abigail Hunter

A new phase field dislocation dynamics formulation is presented, which couples micromechanical solvers and the time-dependent Ginzburg-Landau equation. Grain boundary (GB)-dislocation interactions are studied by describing GBs as inclusions. Grain boundary properties are computed from Molecular Statics simulations and an additional contribution to the total

Andrea Gnarini, Francesco Ursini, Giorgio Matt, Stefano Bianchi

Z-sources are a particular class of neutron star low-mass X-ray binaries characterized by a wide Z-like track in their hard colorsoft color (or hardness-intensity) diagrams, with three branches: the horizontal (HB), the normal (NB), and the flaring branch (FB). Spectropolarimetric observations with the Imaging X-ray Polarimetry Explorer (IXPE) show that the

Harsh Sharma, Juan Diego Draxl Giannoni, Boris Kramer

This work presents structure-preserving Lift & Learn, a scientific machine learning method that employs lifting variable transformations to learn structure-preserving reduced-order models for nonlinear partial differential equations (PDEs) with conservation laws. We propose a hybrid learning approach based on a recently developed energy-quadratization strate

Mubashir Mansoor, Kamil Czelej, Sally Eaton-Magaña, Mehya Mansoor

Achieving high NV center conversion efficiency remains a key challenge in advancing diamond-based quantum technologies. The generally accepted mechanism for NV formation is that irradiation-induced vacancies become mobile during annealing and are trapped by substitutional nitrogen. However, the suggested mechanism does not consider the presence and role of h

Olivia Figueira, Pranathi Chamarthi, Tu Le, Athina Markopoulou

TikTok, the social media platform that is popular among children and adolescents, offers a more restrictive "Under 13 Experience" exclusively for young users in the US, also known as TikTok's "Kids Mode". While prior research has studied various aspects of TikTok's regular mode, including privacy and personalization, TikTok's Kids Mode remains understudied,

Arkaprabha Ganguli, Nesar Ramachandra, Julie Bessac, Emil Constantinescu

This study addresses the challenge of statistically extracting generative factors from complex, high-dimensional datasets in unsupervised or semi-supervised settings. We investigate encoder-decoder-based generative models for nonlinear dimensionality reduction, focusing on disentangling low-dimensional latent variables corresponding to independent physical f

David Ifeoluwa Adelani

Recent advances in word embeddings and language models use large-scale, unlabelled data and self-supervised learning to boost NLP performance. Multilingual models, often trained on web-sourced data like Wikipedia, face challenges: few low-resource languages are included, their data is often noisy, and lack of labeled datasets makes it hard to evaluate perfor

ATLAS Collaboration

A search for R-parity-conserving supersymmetry in events with large missing transverse momentum, jets and at least one hadronically decaying $\tau$-lepton is presented. Both gluino and squark pair production are considered, with the cascade decay of each gluino or squark producing either a $\tau$-slepton or a $\tau$-sneutrino. Three channels are examined, re

Shibo Liu

We obtain multiple solutions for the zero mass Schr{\"o}dinger-Poisson-Slater equation \[ - \Delta u + \left( \frac{1}{4 \pi | x |} \ast u^2 \right) u = \lambda g (x) | u |^{p - 2} u + | u |^{6 - 2} u \text{, \ \ \ \ } u \in \mathcal{D}^{1, 2} (\mathbb{R}^3) \text{} \] for $\lambda \gg 1$, where $p \in (4, 6)$ and $g \in L^{6 / (6 - p)} (\mathbb{R}^3)$. The

Amir Javadi Rad, Amirafshar Moshtaghpour, Dongdong Chen, Angus I. Kirkland

Scanning Transmission Electron Microscopy (STEM) is a critical tool for imaging the properties of materials and biological specimens at atomic scale, yet our understanding of relevant electron beam damage mechanisms is incomplete. Recent studies suggest that certain types of damage can be modelled as a diffusion process. However, numerical simulation of such

Milos Indjin, Nick Keepfer, I-Kang Liu, Nick P. Proukakis

We examine the impact of moderate repulsive self-interactions on fuzzy dark matter halos generated by merging smaller Gaussian density concentrations. We study the size of the core and the granules, the spatial dependence of the field's coherence, the turbulent vortex tangle and the oscillation frequency of the central soliton, covering the range from quantu

Ali Mammadov, Loïc Le Folgoc, Guillaume Hocquet, Pietro Gori

Digital pathology has revolutionized the field by enabling the digitization of tissue samples into whole slide images (WSIs). However, the high resolution and large size of WSIs present significant challenges when it comes to applying Deep Learning models. As a solution, WSIs are often divided into smaller patches with a global label (\textit{i.e., diagnosti

Lei Chen, Bei-Bei Wang, Jianmin Yuan, Long Zhang

A common wisdom about quantum many-body systems is that emergent phases typically fall into either the Landau-Ginzburg paradigm or topological classifications. Experimentally realizing the intertwined emergence of spontaneous symmetry breaking and topological order remains challenging. Here, we present an experimentally accessible platform for studying magne

Pavel Pokhilko, Dominika Zgid

Solution of the Dyson equation for the small-gap systems can be plagued by large non-converging iterations. In addition to the convergence issues, due to a high non-linearity, the Dyson equation may have multiple solutions. We apply the homotopy continuation approach to control the behavior of iterations. We used the homotopy continuation to locate multiple

Ben Deaner, Soonwoo Kwon

We present a novel approach for extrapolating causal effects away from the margin between treatment and non-treatment in sharp regression discontinuity designs with multiple covariates. Our methods apply both to settings in which treatment is a function of multiple observables and settings in which treatment is determined based on a single running variable.

Claire Li, David Freeborn

This study explores how AI-powered digital innovations are reshaping organisational accountability in a transnational governance context. As AI systems increasingly mediate decision-making in domains such as auditing and financial reporting, traditional mechanisms of accountability, based on control, transparency, and auditability, are being destabilised. We

Mohamad Dabboussi, Malo Huard, Yann Gousseau, Pietro Gori

Accurate interpretation of multi-view radiographs is crucial for diagnosing fractures, muscular injuries, and other anomalies. While significant advances have been made in AI-based analysis of single images, current methods often struggle to establish robust correspondences between different X-ray views, an essential capability for precise clinical evaluatio

Tanusree Sharma, Yu-Yun Tseng, Lotus Zhang, Ayae Ide

Blind and low vision (BLV) individuals use Generative AI (GenAI) tools to interpret and manage visual content in their daily lives. While such tools can enhance the accessibility of visual content and so enable greater user independence, they also introduce complex challenges around visual privacy. In this paper, we investigate the current practices and futu

Charles B. Walker, Matthew Stern, Judit Romhányi

We investigate the symmetry-enforced line nodes of the triplet excitations of XCuCl$_{3}$ (X= K, Tl), showing that they are protected by the nonsymmorphic symmetries and are unaffected by the microscopic details, such as interaction and anisotropy strength, as long as the ground state and the symmetry group remain unaltered. Extending the conventionally used

Michael Dougherty, Jon McCammond

This article examines noncrossing partitions of the unit circle in the complex plane; we call these continuous noncrossing partitions. More precisely, we focus on the degree-$d$ continuous noncrossing partitions where unit complex numbers in the same block have identical $d$-th powers. We prove that the degree-$d$ continuous noncrossing partitions form a top

Gregg Wade, Mary Oksala, Coralie Neiner, Etienne Boucher

We report magnetic field measurements spanning about 15 years of four massive ($7.5-15 M_\odot$) supergiant stars: $\alpha$ Per (HD\,20902, F5Iab), $\alpha$ Lep (HD\,36673A, F0Ib), $\eta$ Leo (HD\,87737, A0Ib) and 13 Mon (HD\,46300, A1Ib). For each star, spectropolarimetric observations were collected using ESPaDOnS at the Canada-France-Hawaii Telescope. The

Mohamed Khalafalla, Tejal Mulay, Shonda L Bernadin

Change orders (COs) are a common occurrence in construction projects, leading to increased costs and extended durations. Design-Bid-Build (DBB) projects, favored by state transportation agencies (STAs), often experience a higher frequency of COs compared to other project delivery methods. This study aims to identify areas of improvement to reduce CO frequenc

Meng-Zhi Wu, Marko Toroš, Sougato Bose, Anupam Mazumdar

Matter-wave interferometry is highly susceptible to inertial acceleration noises arising from the vibration of the experimental apparatus. There are various methods for noise suppression. In this paper, we propose leveraging the cross-correlation of multi-directional vibration noises to mitigate their dephasing effect in matter-wave interferometers. Specific

Xiao Hui Tai, Suraj R. Nair, Shikhar Mehra, Joshua E. Blumenstock

Seasonal migration plays a critical role in stabilizing rural economies and sustaining the livelihoods of agricultural households. Violence and civil conflict have long been thought to disrupt these labor flows, but this hypothesis has historically been hard to test given the lack of reliable data on migration in conflict zones. Focusing on Afghanistan in th

Igor Ivanov

In this paper, LLMs are tasked with completing an impossible quiz, while they are in a sandbox, monitored, told about these measures and instructed not to cheat. Some frontier LLMs cheat consistently and attempt to circumvent restrictions despite everything. The results reveal a fundamental tension between goal-directed behavior and alignment in current LLMs

Youngkyu Lee, Shanqing Liu, Jerome Darbon, George Em Karniadakis

We present a general and scalable framework for the automated discovery of optimal meta-solvers for the solution of time-dependent nonlinear partial differential equations after appropriate discretization. By integrating classical numerical methods (e.g., Krylov-based methods) with modern deep learning components, such as neural operators, our approach enabl

Casper Moldrup Rysgaard, Sebastian Wild

Lazy search trees (Sandlund & Wild FOCS 2020, Sandlund & Zhang SODA 2022) are sorted dictionaries whose update and query performance smoothly interpolates between that of efficient priority queues and binary search trees - automatically, depending on actual use; no adjustments are necessary to the data structure to realize the cost savings. In this paper, we

Hidetoshi Nishimori, Masayuki Ohzeki, Manaka Okuyama

Temperature chaos is a striking phenomenon in spin glasses, where even slight changes in temperature lead to a complete reconfiguration of the spin state. Another intriguing effect is the reentrant transition, in which lowering the temperature drives the system from a ferromagnetic phase into a less ordered spin-glass or paramagnetic phase. In the present pa

Torben C. Frost

When photons, gravitational waves, and massive particles such as neutrinos are gravitationally lensed the signals detected by telescopes or detectors on and around Earth are usually either magnified or demagnified. However, for stationary and axisymmetric spacetimes conventional methods for calculating the magnification factor usually only allow to calculate

Yusuke Tanaka, Alvin Zhu, Quanyou Wang, Yeting Liu

Reinforcement learning (RL) has enabled advances in humanoid robot locomotion, yet most learning frameworks do not account for mechanical intelligence embedded in parallel actuation mechanisms due to limitations in simulator support for closed kinematic chains. This omission can lead to inaccurate motion modeling and suboptimal policies, particularly for rob

Alan Yang, Stephen Boyd

The Kalman filter (KF) provides optimal recursive state estimates for linear-Gaussian systems and underpins applications in control, signal processing, and others. However, it is vulnerable to outliers in the measurements and process noise. We introduce the iteratively saturated Kalman filter (ISKF), which is derived as a scaled gradient method for solving a

Zhuochao Peng, Jiaxin Xu, Jun Hu, Haian Xue

While recent research highlights the potential of social robots to support mood regulation, little is known about how prospective users view their integration into everyday life. To explore this, we conducted an exploratory case study that used a speculative robot concept "Mora" to provoke reflection and facilitate meaningful discussion about using social ro

Subed Lamichhane, Haotian Lu, Sheldon X. -D. Tan

Electromigration (EM) remains a critical reliability concern in current and future copper-based VLSI circuits. As technology scales down, EM-induced IR drop becomes increasingly severe. While several EM-aware IR drop analysis tools have been proposed, few incorporate the real impact of temperature distribution on both EM and IR drop effects. In this work, we

Ha Na Cho, Kyuha Jung, Daniel Eisenberg, Cheryl A. King

This qualitative study explores barriers to utilization of digital mental health Intervention (DMHI) among college students. Data are from a large randomized clinical trial of an intervention, eBridge, that used motivational interviewing for online counseling to connect students with mental health issues to professional services. We applied thematic analysis

Omar Claflin

Current sparse autoencoder (SAE) approaches to neural network interpretability assume that activations can be decomposed through linear superposition into sparse, interpretable features. Despite high reconstruction fidelity, SAEs consistently fail to eliminate polysemanticity and exhibit pathological behavioral errors. We propose that neural networks encode

Oren Fivel, Matan Rudman, Kobi Cohen

Deep reinforcement learning (DRL) has become a powerful tool for complex decision-making in machine learning and AI. However, traditional methods often assume perfect action execution, overlooking the uncertainties and deviations between an agent's selected actions and the actual system response. In real-world applications, such as robotics, mechatronics, an

Hussam Al Daas, Davide Palitta

The solution of sequences of shifted linear systems is a classic problem in numerical linear algebra, and a variety of efficient methods have been proposed over the years. Nevertheless, there still exist challenging scenarios witnessing a lack of performing solvers. For instance, state-of-the-art procedures struggle to handle nonsymmetric problems where the

Farhad Kamarei, Bo Zheng, John E. Dolbow, Oscar Lopez-Pamies

Since the turn of the millennium, capitalizing on modern advances in mathematics and computation, a slew of computational models have been proposed in the literature with the objective of describing the nucleation and propagation of fracture in materials subjected to mechanical, thermal, and/or other types of loads. By and large, each new proposal focuses on

Alexis Carrillo, Asieh Abolpour Mofrad, Anis Yazidi, Moises Betancort

Simulations offer a valuable tool for exploring stimulus equivalence (SE), yet the potential of reject relations to disrupt the assessment of equivalence class formation is contentious. This study investigates the role of reject relations in the acquisition of stimulus equivalence using computational models. We examined feedforward neural networks (FFNs), bi

Isabella Basso do Amaral, Renato Cordeiro Ferreira, Alfredo Goldman

The Python programming language is best known for its syntax and scientific libraries, but it is also notorious for its slow interpreter. Optimizing critical sections in Python entails special knowledge of the binary interactions between programming languages, and can be cumbersome to interface manually, with implementers often resorting to convoluted third-

Amirali Sajadi, Kostadin Damevski, Preetha Chatterjee

Large language models (LLMs) and their agentic frameworks are increasingly adopted to perform development tasks such as automated program repair (APR). While prior work has identified security risks in LLM-generated code, most have focused on synthetic, simplified, or isolated tasks that lack the complexity of real-world program repair. In this study, we pre

Felix Kling, Toni Mäkelä, Josh McFayden

Most existing and proposed high energy neutrino experiments have excellent muon charge identification capabilities, enabling the distinction of $\nu_\mu$ and $\bar \nu_\mu$ charged current interactions. In contrast, distinguishing electrons and positrons from $\nu_e$ and $\bar \nu_e$ interactions is typically impossible, as they quickly interact within the c

Zhiyin Lin, Purvi Goel, Joy Yun, C. Karen Liu

Fencing is a sport where athletes engage in diverse yet strategically logical motions. While most motions fall into a few high-level actions (e.g. step, lunge, parry), the execution can vary widely-fast vs. slow, large vs. small, offensive vs. defensive. Moreover, a fencer's actions are informed by a strategy that often comes in response to the opponent's be

Amr Abourayya, Jens Kleesiek, Bharat Rao, Michael Kamp

Data heterogeneity poses a fundamental challenge in federated learning (FL), especially when clients differ not only in distribution but also in the reliability of their predictions across individual examples. While personalized FL (PFL) aims to address this, we observe that many PFL methods fail to outperform two necessary baselines, local training and cent

Jie Hou, Chuxiong Wu, Lannan Luo, Qiang Zeng

As the capabilities of pre-trained large language models (LLMs) continue to advance, the "pre-train and fine-tune" paradigm has become increasingly mainstream, leading to the development of various fine-tuning methods. However, the privacy risks arising from memorization during fine-tuning have received relatively little attention. To address this gap, we ca

Davide Salaorni, Vincenzo De Paola, Samuele Delpero, Giovanni Dispoto

In recent years, \emph{Reinforcement Learning} (RL) has made remarkable progress, achieving superhuman performance in a wide range of simulated environments. As research moves toward deploying RL in real-world applications, the field faces a new set of challenges inherent to real-world settings, such as large state-action spaces, non-stationarity, and partia

Alexey Dubinsky

We analyze the quasinormal modes (QNMs) of scalar, electromagnetic, and Dirac test fields in the background of a black hole immersed in a galactic dark matter halo. The analytic black hole solution considered here is sourced by a physically motivated halo density profile that leads to a flat galactic rotation curve. Using the sixth-order WKB method with Pad\

Sean Patrick O'Neil, Edmond Jonckheere, Sophie Schirmer

Differential sensitivity techniques originally developed to study the robustness of energy landscape controllers are generalized to the important case of closed quantum systems subject to continuously varying controls. Vanishing sensitivity to parameter variation is shown to coincide with perfect fidelity, as was the case for time-invariant controls. Upper b

Ming Wang, Ang Li, Frank Mueller

This work presents a hardware-efficient and fully parallelizable decoder for quantum LDPC codes that leverages belief propagation (BP) with a speculative post-processing strategy inspired by classical Chase decoding algorithm. By monitoring bit-level oscillation patterns during BP, our method identifies unreliable bits and generates multiple candidate vector

Zhuangzhuang Dai, Vincent Gbouna Zakka, Luis J. Manso, Chen Li

Enabling robots to understand human gaze target is a crucial step to allow capabilities in downstream tasks, for example, attention estimation and movement anticipation in real-world human-robot interactions. Prior works have addressed the in-frame target localization problem with data-driven approaches by carefully removing out-of-frame samples. Vision-base

Jean Cardinal, Yelena Yuditsky

We consider the existence and construction of \textit{biclique covers} of graphs, consisting of coverings of their edge sets by complete bipartite graphs. The \textit{size} of such a cover is the sum of the sizes of the bicliques. Small-size biclique covers of graphs are ubiquitous in computational geometry, and have been shown to be useful compact represent

Ben Szczesny

In this paper, we present an explicit method to identify equivariant suboperads of coinduced operads that contain only fixed points associated to any desired transfer system. Our method works for a class of operads that we call intersection operads, which includes many familiar operads of interest, including the little $k$-cube operads, the Steiner operad, a

E. A. Ladeishchikov, L. V. Lokutsievskiy, N. V. Prilepin

In this paper, we consider the problem of finding geodesics in a series of left-invariant problems endowed with sub-Lorentzian and Finsler structures. Explicit formulas for extremals are obtained in terms of convex trigonometric functions. In the sub-Lorentzian setting, the new trigonometric functions $\cosh_\Omega$ and $\sinh_\Omega$, developed here, prove

Nikhil Kumar

This paper presents a social learning model where the network structure is endogenously determined by signal precision and dimension choices. Agents not only choose the precision of their signals and what dimension of the state to learn about, but these decisions directly determine the underlying network structure on which social learning occurs. We show tha

Nikita Nikitin, Eugene Fomin

We present a novel framework for real-time sign language recognition using lightweight DNNs trained on limited data. Our system addresses key challenges in sign language recognition, including data scarcity, high computational costs, and discrepancies in frame rates between training and inference environments. By encoding sign language specific parameters, s

Daniel M. Harris, Jack-William Barotta

When a floating body is internally or externally vibrated, its self-generated wavefield can lead to steady propulsion along the interface. In this article, we review several related and recently discovered systems that leverage this propulsion mechanism and interact hydrodynamically with one another via these surface waves. Particles with an onboard oscillat

Sanchit Ahuja, Praneetha Vaddamanu, Barun Patra

Despite recent advances in Language Reasoning Models (LRMs), most research focuses solely on English, even though many models are pretrained on multilingual data. In this work, we investigate: Is English the most token-efficient language for reasoning? We evaluate three open-source RLMs: DeepSeek R1, Qwen 2.5 and Qwen 3, across four math datasets and seven t

Madhuvanthi Guruprasad Athani, Nathan Prouse, Niranjan Sarpangala, Patrick Noerr

How chirality propagates across scales remains an open question in many biological and synthetic systems. An especially clear manifestation of this propagation is found in in vitro gliding assays of cytoskeletal filaments on surfaces, driven by molecular motors. These assays have become model systems of active matter dynamics, as they spontaneously organize

Isabella Senturia, Matilde Marcolli

Within the context of the mathematical formulation of Merge and the Strong Minimalist Thesis, we present a mathematical model of the morphology-syntax interface. In this setting, morphology has compositional properties responsible for word formation, organized into a magma of morphological trees. However, unlike syntax, we do not have movement within morphol

Chi-Yao Huang, Zeel Bhatt, Yezhou Yang

Breakthroughs in visual odometry (VO) have fundamentally reshaped the landscape of robotics, enabling ultra-precise camera state estimation that is crucial for modern autonomous systems. Despite these advances, many learning-based VO techniques rely on rigid geometric assumptions, which often fall short in interpretability and lack a solid theoretical basis

Saurya Das, Peter Dunsby, S. Shajidul Haque, Seturumane Tema

We investigate the dynamics of the Friedmann-Lema\^itre-Robertson-Walker spacetime within the framework of $f(R)$ gravity using a compact, model-independent dynamical systems approach. By assuming a power-law scale factor, we explore ekpyrotic and accelerating solutions to address the big bang singularity. Our analysis demonstrates that a cosmological bounce

Shahab Shahidi, Shiva Kayedi

The hybrid metric-Palatini gravity with Lagrangian density $L =R+f(\mathcal{R},\mathcal{R}_{\mu\nu}\mathcal{R}^{\mu\nu})$ is considered, where $R$ is the metric Ricci scalar and $\mathcal{R}_{\mu\nu}$ is the Palatini Ricci tensor. Contrary to the standard hybrid metric-Palatini theory, because of the term $\mathcal{R}_{\mu\nu}\mathcal{R}^{\mu\nu}$ in the act

Mahmoud Abdelgalil, Vishal Shenoy, Guido Cavraro, Emiliano Dall'Anese

This paper studies a power transmission system with both conventional generators (CGs) and distributed energy assets (DEAs) providing frequency control. We consider an operating condition with demand aggregating two dynamic components: one that switches between different values on a finite set, and one that varies smoothly over time. Such dynamic operating c

Aryan Shrivastava, Ari Holtzman

Most commonly used language models (LMs) are instruction-tuned and aligned using a combination of fine-tuning and reinforcement learning, causing them to refuse users requests deemed harmful by the model. However, jailbreak prompts can often bypass these refusal mechanisms and elicit harmful responses. In this work, we study the extent to which information a

Anahid Kiani, S. Mahdi Fazeli, G. Reza Jafari

Classical Heider balance theory models the evolution of social networks towards balanced states with stress minimization. Triad relationships are classically either balanced or imbalanced. However, real-world relationships often exhibit uncertainty, complexity, and interconnected dynamics that transcend this classical framework, and we will sometimes see the

Oleg Kolosov, David Breitgand, Dean H. Lorenz, Gala Yadgar

Network virtualization allows hosting applications with diverse computation and communication requirements on shared edge infrastructure. Given a set of requests for deploying virtualized applications, the edge provider has to deploy a maximum number of them to the underlying physical network, subject to capacity constraints. This challenge is known as the v

Chinmay Vilas Samak, Tanmay Vilas Samak, Bing Li, Venkat Krovi

Simulation-based design, optimization, and validation of autonomous vehicles have proven to be crucial for their improvement over the years. Nevertheless, the ultimate measure of effectiveness is their successful transition from simulation to reality (sim2real). However, existing sim2real transfer methods struggle to address the autonomy-oriented requirement

Bubai Manna

In a connected simple graph G = (V(G),E(G)), each vertex is assigned a color from the set of colors C={1, 2,..., c}. The set of vertices V(G) is partitioned as V_1, V_2, ... ,V_c, where all vertices in V_j share the same color j. A subset S of V(G) is called Selective Subset if, for every vertex v in V(G), and if v is in V_j, at least one of its nearest neig

Jiztom Kavalakkatt Francis, Matthew J Darr

In this paper, we present a novel framework for enhancing model interpretability by integrating heatmaps produced separately by ResNet and a restructured 2D Transformer with globally weighted input saliency. We address the critical problem of spatial-temporal misalignment in existing interpretability methods, where convolutional networks fail to capture glob

Kevin F. Lee, Jacob Lampen, Peng Li, Jie Jiang

Modelocked frequency comb lasers have always operated with a single pulse circulating in the laser cavity. This meant that each laser technology had an associated limit on pulse repetition rate. Achieving higher rates required different technology, for example exchanging optical fiber for solid state gain media. However, this is a perceived, rather than a fu

Jean-Michel Scherer

Deformation band patterning in single crystals is investigated using a finite strain crystal viscoplasticity model based on the evolution of dislocation densities. In the presence of strong latent hardening and weak rate dependence, the deformation organizes into laminate microstructures consisting of single-slip regions separated by dislocation walls. The i

Gennadi Saiko, Timothy Burton, Faraz Sadrzadeh-Afsharazar, Shota Yamashita

Large neck vessels (carotid artery and internal jugular vein, IJV) offer a unique opportunity to monitor hemodynamics non-invasively by optical means. The primary shortcoming of past work has been the focus on healthy volunteers in normal physiological conditions and well-controlled environments. To drive the technology closer to the bedside, testing is requ

Peilin He, James Joshi

Reconstructing high-quality images from low-resolution inputs using Residual Dense Spatial Networks (RDSNs) is crucial yet challenging. It is even more challenging in centralized training where multiple collaborating parties are involved, as it poses significant privacy risks, including data leakage and inference attacks, as well as high computational and co

Tarikul Islam Tamiti, Biraj Joshi, Rida Hasan, Rashedul Hasan

Speech super-resolution (SSR) enhances low-resolution speech by increasing the sampling rate. While most SSR methods focus on magnitude reconstruction, recent research highlights the importance of phase reconstruction for improved perceptual quality. Therefore, we introduce CTFT-Net, a Complex Time-Frequency Transformation Network that reconstructs both magn

Bittu Chahal, Tapas Chatterjee, Sneha Chaubey

We investigate the distributional properties of the sequence of Farey fractions with $k$-free denominators in residue classes, defined as \[\mathscr{F}_{Q,k}^{(m)}:=\left\{\frac{a}{q}\ |\ 1\leq a\leq q\leq Q,\ \gcd(a,q)=1,\ q\ \text{is}\ k\text{-free}\ \&\ q\equiv b\pmod{m} \right\}.\] We show that $\left(\mathscr{F}_{Q,k}^{(m)}\right)_{Q\ge 1}$ is equidistr

Paul Mayer, Florian Lux, Alejandro Pérez-González-de-Martos, Angelina Elizarova

While generative methods have progressed rapidly in recent years, generating expressive prosody for an utterance remains a challenging task in text-to-speech synthesis. This is particularly true for systems that model prosody explicitly through parameters such as pitch, energy, and duration, which is commonly done for the sake of interpretability and control