arXiv papers from May 2025

Harvineet Singh, Fan Xia, Alexej Gossmann, Andrew Chuang

Machine learning (ML) models frequently experience performance degradation when deployed in new contexts. Such degradation is rarely uniform: some subgroups may suffer large performance decay while others may not. Understanding where and how large differences in performance arise is critical for designing targeted corrective actions that mitigate decay for t

Benjamin Civjan, Bo Chen, Ruixiao Zhang, Klara Nahrstedt

Video processing for real-time analytics in resource-constrained environments presents a significant challenge in balancing energy consumption and video semantics. This paper addresses the problem of energy-efficient video processing by proposing a system that dynamically optimizes processing configurations to minimize energy usage on the edge, while preserv

Ivan Novak

We determine all groups which occur as torsion subgroups of $\mathbb Q$-curves defined over number fields of degrees $3$, $5$ and $7$. In particular, we prove that every torsion subgroup of a $\mathbb Q$-curve defined over a number field of degree $3,5$ or $7$ already occurs as a torsion subgroup of an elliptic curve with rational $j$-invariant. As the quadr

Dongzhe Zheng

This paper establishes a metric framework for Spencer complexes based on the geometric theory of compatible pairs $(D,\lambda)$ in principal bundle constraint systems, solving fundamental technical problems in computing Spencer cohomology of constraint systems. We develop two complementary and geometrically natural metric schemes: a tensor metric based on co

Zhuojun Gu, Quan Wang, Shuchu Han

Recent advances in Large Language Models (LLMs) highlight the need to align their behaviors with human values. A critical, yet understudied, issue is the potential divergence between an LLM's stated preferences (its reported alignment with general principles) and its revealed preferences (inferred from decisions in contextualized scenarios). Such deviations

Monoshi Kumar Roy, Simin Chen, Benjamin Steenhoek, Jinjun Peng

Understanding and reasoning about code semantics is essential for enhancing code LLMs' abilities to solve real-world software engineering (SE) tasks. Although several code reasoning benchmarks exist, most rely on synthetic datasets or educational coding problems and focus on coarse-grained reasoning tasks such as input/output prediction, limiting their effec

Sima Farokhnejad, Angélica S. da Mata, Mariana Macedo, Ronaldo Menezes

Commodities, including livestock, flow through trade networks globally, with trajectories that can be effectively captured using mobility pattern modelling approaches similar to those used in human mobility studies. However, documenting these movements comprehensively presents significant challenges; it can be unrealistic, costly, and may conflict with data

Mania Abdi, Peter Desnoyers, Mark Crovella, Raja R. Sambasivan

Diagnosing problems in deployed distributed applications continues to grow more challenging. A significant reason is the extreme mismatch between the powerful abstractions developers have available to build increasingly complex distributed applications versus the simple ones engineers have available to diagnose problems in them. To help, we present a novel a

Pardis Sadat Zahraei, Ali Emami

Addressing gender bias and maintaining logical coherence in machine translation remains challenging, particularly when translating between natural gender languages, like English, and genderless languages, such as Persian, Indonesian, and Finnish. We introduce the Translate-with-Care (TWC) dataset, comprising 3,950 challenging scenarios across six low- to mid

Kang Zhou

In this paper, we extend the method proposed in \cite{Arkani-Hamed:2024fyd} for deriving soft theorems of amplitudes, which relies exclusively on factorization properties including conventional factorizations on physical poles, as well as newly discovered $2$-splits on special loci in kinematic space. Using the extended approach, we fully reproduce the leadi

Julian Andrej, Tzanio Kolev, Boyan Lazarov

This article aims to demonstrate and discuss the applications of automatic differentiation (AD) for finding derivatives in PDE-constrained optimization problems and Jacobians in non-linear finite element analysis. The main idea is to localize the application of AD at the integration point level by combining it with the so-called Finite Element Operator Decom

Dung Nguyen, Aravind Srinivasan, Renata Valieva, Anil Vullikanti

Designing effective strategies for controlling epidemic spread by vaccination is an important question in epidemiology, especially in the early stages when vaccines are limited. This is a challenging question when the contact network is very heterogeneous, and strategies based on controlling network properties, such as the degree and spectral radius, have be

Kazuki Irie, Morris Yau, Samuel J. Gershman

We develop hybrid memory architectures for general-purpose sequence processing neural networks, that combine key-value memory using softmax attention (KV-memory) with fast weight memory through dynamic synaptic modulation (FW-memory) -- the core principles of quadratic and linear transformers, respectively. These two memory systems have complementary but ind

Yeshwanth Venkatesha, Souvik Kundu, Priyadarshini Panda

Parameter Efficient Fine-Tuning (PEFT) has become the de-facto approach in adapting Large Language Models (LLMs) for downstream tasks in Natural Language Processing. However, its adoption in privacy-preserving distributed learning frameworks, such as Federated Learning (FL), remains relatively limited. This is mainly due to challenges specific to FL, such as

Zeqi Gu, Yin Cui, Zhaoshuo Li, Fangyin Wei

Designing 3D scenes is traditionally a challenging task that demands both artistic expertise and proficiency with complex software. Recent advances in text-to-3D generation have greatly simplified this process by letting users create scenes based on simple text descriptions. However, as these methods generally require extra training or in-context learning, t

Shangbin Feng, Yike Wang, Weijia Shi, Yulia Tsvetkov

We propose Data Swarms, an algorithm to optimize the generation of synthetic evaluation data and advance quantitative desiderata of LLM evaluation. We first train a swarm of initial data generators using existing data, and define various evaluation objectives to reflect the desired properties of evaluation (e.g., generate more difficult problems for the eval

Harveen Singh Chadha, Aswin Shanmugam Subramanian, Vikas Joshi, Shubham Bansal

In video dubbing, aligning translated audio with the source audio is a significant challenge. Our focus is on achieving this efficiently, tailored for real-time, on-device video dubbing scenarios. We developed a phoneme-based end-to-end length-sensitive speech translation (LSST) model, which generates translations of varying lengths short, normal, and long u

Chiyu Zhang, Marc-Alexandre Cote, Michael Albada, Anush Sankaran

Large language model (LLM) agents have shown impressive capabilities in human language comprehension and reasoning, yet their potential in cybersecurity remains underexplored. We introduce DefenderBench, a practical, open-source toolkit for evaluating language agents across offense, defense, and cybersecurity knowledge-based tasks. DefenderBench includes env

Rynaa Grover, Jayant Sravan Tamarapalli, Sahiti Yerramilli, Nilay Pande

Recent Multimodal Large Language Models (MLLMs) demonstrate strong high-level visual reasoning on tasks such as visual question answering and image captioning. Yet existing benchmarks largely overlook their ability to capture fine-grained perceptual details. As MLLMs are increasingly deployed in safety and reliability critical settings, perceptual acuity bec

Darij Grinberg

This is an introduction to algebraic combinatorics, written for a quarter-long graduate course. It starts with a rigorous introduction to formal power series with some combinatorial applications, then discusses integer partitions (proving Jacobi's triple product identity), permutations (Lehmer codes, cycles) and subtractive methods (alternating sums, cancell

Yulia Otmakhova, Lea Frermann

Narrative frames are a powerful way of conceptualizing and communicating complex, controversial ideas, however automated frame analysis to date has mostly overlooked this framing device. In this paper, we connect elements of narrativity with fundamental aspects of framing, and present a framework which formalizes and operationalizes such aspects. We annotate

Kuan-Po Huang, Shu-wen Yang, Huy Phan, Bo-Ru Lu

Text-to-audio generation synthesizes realistic sounds or music given a natural language prompt. Diffusion-based frameworks, including the Tango and the AudioLDM series, represent the state-of-the-art in text-to-audio generation. Despite achieving high audio fidelity, they incur significant inference latency due to the slow diffusion sampling process. MAGNET,

T. Ahmed, S. Jannat, Md. F. Islam, J. Noor

Agriculture is vital for global food security, but crops are vulnerable to diseases that impact yield and quality. While Convolutional Neural Networks (CNNs) accurately classify plant diseases using leaf images, their high computational demands hinder their deployment in resource-constrained settings such as smartphones, edge devices, and real-time monitorin

Kenji Nakagawa, Yoshinori Takei

In this paper, we consider the problem of finding the center $Q^\ast$ of the SEB (smallest enclosing ball) for $n$ points in $d$-dimensional Euclidean space. One application of the SEB is SVDD (support vector data description) in support vector machines. Our objective is to develop a sequential computation algorithm for determining the barycentric coordinate

Miao Zhang, Aref Farhadipour, Annie Baker, Jiachen Ma

With its crosslinguistic and cross-speaker diversity, the Mozilla Common Voice Corpus (CV) has been a valuable resource for multilingual speech technology and holds tremendous potential for research in crosslinguistic phonetics and speech sciences. Properly accounting for speaker variation is, however, key to the theoretical and statistical bases of speech r

Caio Corro, Mathieu Lacroix, Joseph Le Roux

We propose a novel discriminative model for sequence labeling called Bregman conditional random fields (BCRF). Contrary to standard linear-chain conditional random fields, BCRF allows fast parallelizable inference algorithms based on iterative Bregman projections. We show how such models can be learned using Fenchel-Young losses, including extension for lear

Mohammad Saleh Torkestani, Taha Mansouri

This paper presents a theoretical framework for addressing the challenges posed by generative artificial intelligence (AI) in higher education assessment through a machine-versus-machine approach. Large language models like GPT-4, Claude, and Llama increasingly demonstrate the ability to produce sophisticated academic content, traditional assessment methods

Binghang Lu, Changhong Mou, Guang Lin

Physics-informed neural networks (PINNs) have emerged as a powerful tool for solving forward and inverse problems involving partial differential equations (PDEs) by incorporating physical laws into the training process. However, the performance of PINNs is often hindered in real-world scenarios involving noisy observational data and missing physics, particul

Mehmet Aziz Yirik, Jakob Lykke Andersen, Rolf Fagerberg, Daniel Merkle

Reservoir computing is a type of a recurrent neural network, mapping the inputs into higher dimensional space using fixed and nonlinear dynamical systems, called reservoirs. In the literature, there are various types of reservoirs ranging from in-silico to in-vitro. In cheminformatics, previous studies contributed to the field by developing simulation-based

William G Coon, Diego Luna, Akshita Panagrahi, Matthew Reid

Transfer learning, a technique commonly used in generative artificial intelligence, allows neural network models to bring prior knowledge to bear when learning a new task. This study demonstrates that transfer learning significantly enhances the accuracy of sleep-stage decoding from peripheral wearable devices by leveraging neural network models pretrained o

Patrick Nyadjo Fonga

Let $E$ be a subset of the projective line over a commutative field $\mathbb{K}$. When $\mathbb{K}$ has infinite cardinality, it is well known that if $E$ contains at most three elements, then the group of linear fractional transformations preserving $E$ is either infinite or isomorphic to the symmetric group on three elements. In this work, we investigate t

Dongzhe Zheng

This paper develops a mirror symmetry theory of Spencer cohomology within the geometric framework of constrained systems on principal bundles, revealing deep symmetric structures in constraint geometry. Based on compatible pairs $(D,\lambda)$ under strong transversality conditions, we construct a systematic family of mirror transformations: from basic sign m

Javier Bisbal, Julio Sotelo, Maria I Valdés, Pablo Irarrazaval

Background and Objective: Plane reformatting for four-dimensional phase contrast MRI (4D flow MRI) is time-consuming and prone to inter-observer variability, which limits fast cardiovascular flow assessment. Deep reinforcement learning (DRL) trains agents to iteratively adjust plane position and orientation, enabling accurate plane reformatting without the n

Hongye Zheng, Yichen Wang, Ray Pan, Guiran Liu

This paper presents a gradient-informed fine-tuning method for large language models under few-shot conditions. The goal is to enhance task adaptability and training stability when data is limited. The method builds on a base loss function and introduces two gradient-related regularization terms. The first enforces gradient direction consistency to guide par

Mouyang Cheng, Chu-Liang Fu, Bowen Yu, Eunbi Rha

Defects are ubiquitous in solids and strongly influence materials' mechanical and functional properties. However, non-destructive characterization and quantification of defects, especially when multiple types coexist, remain a long-standing challenge. Here we introduce DefectNet, a foundation machine learning model that predicts the chemical identity and con

Siddharth Prabhu, Srinivas Rangarajan, Mayuresh Kothare

Multiple-shooting is a parameter estimation approach for ordinary differential equations. In this approach, the trajectory is broken into small intervals, each of which can be integrated independently. Equality constraints are then applied to eliminate the shooting gap between the end of the previous trajectory and the start of the next trajectory. Unlike si

Daniel Paleka, Shashwat Goel, Jonas Geiping, Florian Tramèr

Large language models (LLMs) have recently been applied to forecasting tasks, with some works claiming these systems match or exceed human performance. In this paper, we argue that, as a community, we should be careful about such conclusions as evaluating LLM forecasters presents unique challenges. We identify two broad categories of issues: (1) difficulty i

Siddhant Arora, Jinchuan Tian, Hayato Futami, Jee-weon Jung

Unlike traditional cascaded pipelines, end-to-end (E2E) spoken dialogue systems preserve full differentiability and capture non-phonemic information, making them well-suited for modeling spoken interactions. However, existing E2E approaches often require large-scale training data and generates responses lacking semantic coherence. We propose a simple yet eff

Tianze Yang, Tyson Jordan, Ruitong Sun, Ninghao Liu

We present Common Inpainted Objects In-N-Out of Context (COinCO), a novel dataset addressing the scarcity of out-of-context examples in existing vision datasets. By systematically replacing objects in COCO images through diffusion-based inpainting, we create 97,722 unique images featuring both contextually coherent and inconsistent scenes, enabling effective

Siddharth Prabhu, Srinivas Rangarajan, Mayuresh Kothare

Inverse problem or parameter estimation of ordinary differential equations (ODEs), the iterative process of minimizing the mismatch between model-predicted and experimental states by tuning the parameter values within an optimization formulation, is commonplace in chemical engineering applications. A popular method for parameter estimation is sequential opti

Mordechai Guri, Dor Fibert

Web client fingerprinting has become a widely used technique for uniquely identifying users, browsers, operating systems, and devices with high accuracy. While it is beneficial for applications such as fraud detection and personalized experiences, it also raises privacy concerns by enabling persistent tracking and detailed user profiling. This paper introduc

Tianqin Li, Ziqi Wen, Leiran Song, Jun Liu

Human vision organizes local cues into coherent global forms using Gestalt principles like closure, proximity, and figure-ground assignment -- functions reliant on global spatial structure. We investigate whether modern vision models show similar behaviors, and under what training conditions these emerge. We find that Vision Transformers (ViTs) trained with

Mina Huh, Zihui Xue, Ujjaini Das, Kumar Ashutosh

People use videos to learn new recipes, exercises, and crafts. Such videos remain difficult for blind and low vision (BLV) people to follow as they rely on visual comparison. Our observations of visual rehabilitation therapists (VRTs) guiding BLV people to follow how-to videos revealed that VRTs provide both proactive and responsive support including detaile

Artur Novais, André L. B. Ribeiro

The phenomenon of augmented gravity on the scale of galaxies, conventionally attributed to dark matter halos, is shown to possibly result from the incremental growth of galactic masses and radii over time. This approach elucidates the cosmological origins of the acceleration scale $a_0\approx cH_0/2\pi\approx10^{-10}$ms$^{-2}$ at which galaxy rotation curves

Shi Mao, Yogeshwar Nath Mishra, Wolfgang Heidrich

The desire for cameras with smaller form factors has recently lead to a push for exploring computational imaging systems with reduced optical complexity such as a smaller number of lens elements. Unfortunately such simplified optical systems usually suffer from severe aberrations, especially in off-axis regions, which can be difficult to correct purely in so

Zhengning Hu, Rohan Joshi

We provide a systematic method to classify all smooth weak Fano toric varieties of Picard rank $3$ in any dimension using Macaulay2, and describe the classification explicitly in dimensions $3$ and $4$. There are $28$ and $114$ isomorphism classes of rank $3$ weak Fano toric threefolds and fourfolds, respectively.

Mingwei Zheng, Chengpeng Wang, Xuwei Liu, Jinyao Guo

Functional correctness is critical for ensuring the reliability and security of network protocol implementations. Functional bugs, instances where implementations diverge from behaviors specified in RFC documents, can lead to severe consequences, including faulty routing, authentication bypasses, and service disruptions. Detecting these bugs requires deep se

Debarati Bhattacharjee, Ashish Anand

This paper presents a framework to convert argumentative texts into argument knowledge graphs (AKG). The proposed argumentative knowledge representation framework (AKReF) extends the theoretical foundation and enables the AKG to provide a graphical view of the argumentative structure that is easier to understand. Starting with basic annotations of argumentat

Joan Hernández

We characterize the s-parabolic Lipschitz caloric capacity of corner-like $s$-parabolic Cantor sets in $\mathbb{R}^{n+1}$ for $1/2<s\leq 1$. Despite the spatial gradient of the s-heat kernel lacking temporal anti-symmetry, we obtain analogous results to those known for analytic and Riesz capacities.

Wei Dai, Peilin Chen, Chanakya Ekbote, Paul Pu Liang

Clinical decision-making routinely demands reasoning over heterogeneous data, yet existing multimodal language models (MLLMs) remain largely vision-centric and fail to generalize across clinical specialties. To bridge this gap, we introduce QoQ-Med-7B/32B, the first open generalist clinical foundation model that jointly reasons across medical images, time-se

Valter Hudovernik, Minkai Xu, Juntong Shi, Lovro Šubelj

Real-world databases are predominantly relational, comprising multiple interlinked tables that contain complex structural and statistical dependencies. Learning generative models on relational data has shown great promise in generating synthetic data and imputing missing values. However, existing methods often struggle to capture this complexity, typically r

Yarou M. Assimiou, Daniel S. Takou, Boukari Amidou, Guingarey Issoufou

In this paper, we present the solutions of the Dirac-Weyl equation for graphene under a constant magnetic field. The resulting spectrum is used to determine the partition function, a key quantity in the study of thermodynamic properties. From this function, we analyze the mean energy, specific heat, entropy, and free energy in two different frameworks: the c

Shahbaz Hussain

The economic dispatch of generators is a major concern in thermal power plants that governs the share of each generating unit with an objective of minimizing fuel cost by fulfilling load demand. This problem is not as simple as it looks because of system constraints that cannot be neglected practically. Moreover, increased awareness of clean technology impos

Yongkang Xiao, Sinian Zhang, Yi Dai, Huixue Zhou

Knowledge graph completion (KGC) aims to predict missing triples in knowledge graphs (KGs) by leveraging existing triples and textual information. Recently, generative large language models (LLMs) have been increasingly employed for graph tasks. However, current approaches typically encode graph context in textual form, which fails to fully exploit the poten

Marjana Mahdia, Artur Lozovoi, Jared Rovny, Zhiyang Yuan

Nitrogen vacancy (NV) centers in diamond exhibit long spin coherence times, optical initialization, and optical spin readout under ambient conditions, making them excellent quantum sensors. However, the conventional scheme for charge state initialization based on off-resonant green excitation results in significant state preparation errors, typically around

Jan-Christoph Kalo, Fina Polat, Shubha Guha, Paul Groth

Modern AI systems are complex workflows containing multiple components and data sources. Data provenance provides the ability to interrogate and potentially explain the outputs of these systems. However, provenance is often too detailed and not contextualized for the user trying to understand the AI system. In this work, we present our vision for an interact

Xinxu Wei, Kanhao Zhao, Yong Jiao, Lifang He

As large language models (LLMs) continue to revolutionize AI research, there is a growing interest in building large-scale brain foundation models to advance neuroscience. While most existing brain foundation models are pre-trained on time-series signals or connectome features, we propose a novel graph-based pre-training paradigm for constructing a brain gra

Silpa B S, Shovan Kanti Barik, Varna Shenoy, Soham Chandak

We demonstrate the effect of inter-atomic interaction in highly excited Rydberg atoms via Autler- Townes splitting in cold atoms. We measure the Autler-Townes (AT) splitting of the 5S1/2, F=2 to 5P3/2, F'=3 transition of 87Rb atoms arising due to the strong coupling of the transition via the cooling beams used for the magneto-optical trap (MOT). The AT split

Daozhe Lin, Qiang Du

This paper investigates solution strategies for nonlinear problems in Hilbert spaces, such as nonlinear partial differential equations (PDEs) in Sobolev spaces, when only finite measurements are available. We formulate this as a nonlinear optimal recovery problem, establishing its well-posedness and proving its convergence to the true solution as the number

Anahita Jain, Husni Idris, John-Paul Clarke, Daniel Delahaye

We present an adaptive control scheme to enable the emergence of order within distributed, autonomous multi-agent systems. Past studies showed that under high-density conditions, order generated from traffic-following behavior reduces travel times, while under low densities, choosing direct paths is more beneficial. In this paper, we leveraged those findings

Ibrahima Dione

Although it is relatively easy to apply, the gradient method often displays a disappointingly slow rate of convergence. Its convergence is specially based on the structure of the matrix of the algebraic linear system, and on the choice of the stepsize defining the new iteration. We propose here a simple and robust stabilization of the gradient method, which

Julio Cesar Jaramillo Quiceno

This work introduces a novel $q$-$\hbar$ deformation of the Heisenberg algebra, designed to unify and extend several existing $q$-deformed formulations. Starting from the canonical Heisenberg algebra defined by the commutation relation $[\hat{x}, \hat{p}] = i\hbar$ on a Hilbert space \cite{Zettili2009}, we survey a variety of $q$-deformed structures previous

Yongchao Huang

Determining whether a dataset was part of a machine learning model's training data pool can reveal privacy vulnerabilities, a challenge often addressed through membership inference attacks (MIAs). Traditional MIAs typically require access to model internals or rely on computationally intensive shadow models. This paper proposes an efficient, interpretable an

Nikola Milosevic, Johannes Müller, Nico Scherf

In constrained Markov decision processes, enforcing constraints during training is often thought of as decreasing the final return. Recently, it was shown that constraints can be incorporated directly into the policy geometry, yielding an optimization trajectory close to the central path of a barrier method, which does not compromise final return. Building o

Tianze Yang, Yucheng Shi, Mengnan Du, Xuansheng Wu

Vector-Quantized Generative Models (VQGMs) have emerged as powerful tools for image generation. However, the key component of VQGMs -- the codebook of discrete tokens -- is still not well understood, e.g., which tokens are critical to generate an image of a certain concept? This paper introduces Concept-Oriented Token Explanation (CORTEX), a novel approach f

João Marcos do Ó, Jaqueline de Lima, Márcio Santos

We establish a rigidity theorem for annular sector-like domains in the setting of overdetermined elliptic problems on model Riemannian manifolds. Specifically, if such a domain admits a solution to the inhomogeneous Helmholtz equation satisfying both constant Dirichlet and constant Neumann boundary conditions, then the domain must be a spherical sector, and

Megan S. Harris, Ehsanoddin Ghorbanichemazkati, Mohammad Mahdi Naderi, John C. Little

Worsening global challenges demand solutions grounded in a systems-level understanding of coupled social and environmental dynamics. Existing environmental models encode extensive knowledge of individual systems, yet much of this information remains isolated within domain-specific formulations and data structures. This paper introduces a unified modeling fra

Ben Zindorf, Sougato Bose

Multi-controlled single-target (MC) gates are some of the most crucial building blocks for varied quantum algorithms. How to implement them optimally is thus a pivotal question. To answer this question in an architecture-independent manner, and to get a worst-case estimate, we should look at a linear nearest-neighbor (LNN) architecture, as this can be embedd

Li Zhang, Morgan Gray, Jaromir Savelka, Kevin D. Ashley

Large Language Models (LLMs) demonstrate potential in complex legal tasks like argument generation, yet their reliability remains a concern. Building upon pilot work assessing LLM generation of 3-ply legal arguments using human evaluation, this paper introduces an automated pipeline to evaluate LLM performance on this task, specifically focusing on faithfuln

Emory Smith, Robert Wallace, Matthew Robison, Yu Huang

This paper presents a study of human visual attention during localization of memory bugs in C. Human visual attention refers to the mechanical processes by which we selectively process and prioritize information. Visual attention is important to study because it is central to what information people (who are sighted) use to solve a particular problem. Meanwh

David Peede, Trevor Cousins, Arun Durvasula, Anastasia Ignatieva

Genomes contain the mutational footprint of an organism's evolutionary history, shaped by diverse forces including ecological factors, selective pressures, and life history traits. The sequentially Markovian coalescent (SMC) is a versatile and tractable model for the genetic genealogy of a sample of genomes, which captures this shared history. Methods that u

Junaid Muzaffar, Khubaib Ahmed, Ingo Frommholz, Zeeshan Pervez

Training reinforcement learning (RL) agents often requires significant computational resources and prolonged training durations. To address this challenge, we build upon prior work that introduced a neural architecture with permutation-invariant sensory processing. We propose a modified attention mechanism that applies a non-linear transformation to the key

Abhiram B R, Ilia Ponomarev, Tomas Polcar

The frictional and wear performance of molybdenum disulfide (MoS2) is significantly influenced by its intrinsic arrangement of crystals or crystallinity. In this study, we investigate the effect of crystallinty on coefficient of friction (COF) and wear in MoS2 using a suite of reactive molecular dynamics (MD) simulations. A range of configurations, from amor

J. Leaños, M. Lomelí-Haro, Christophe Ndjatchi, L. M. Ríos-Castro

Let $G=(V(G),E(G))$ be a simple graph, and let $U\subseteq V(G)$. Two distinct vertices $x,y\in U$ are $U$-mutually visible if $G$ contains a shortest $x$-$y$ path that is internally disjoint from $U$. $U$ is called a mutual-visibility set of $G$ if any two vertices of $U$ are $U$-mutually visible. The mutual-visibility number $\mu(G)$ of $G$ is the size of

Zhili Feng, Yixuan Even Xu, Alexander Robey, Robert Kirk

Machine unlearning aims to remove sensitive or undesired data from large language models. However, recent studies suggest that unlearning is often shallow, claiming that removed knowledge can easily be recovered. In this work, we critically examine standard unlearning evaluation practices and uncover key limitations that shake our trust in those findings. Fi

Luis Felipe Leon-Pinzon, Elisabeth Restrepo-Parra, Andres Manuel Garay-Tapia

Thermoelectric materials require high electronic conductivity and low thermal conductivity. CrN has been shown to have low phononic thermal conductivity, making it a potential candidate for thermoelectric applications. In addition, similarities have been observed between YN and ScN suggesting that the CrYN alloy may have interesting thermoelectric properties

Frank Göhmann, Andreas Klümper, Karol K. Kozlowski

Owing to the fact that the particle current operator in non-relativistic gases is proportional to the total momentum operator, the particle transport in such systems is always ballistic and fully characterized by a Drude weight $\Delta$. The Drude weight can be calculated within linear response theory. It is given by the formula $\Delta = 2 \pi D$, where $D$

Max Chicky Fang

The Kruithof iterative scaling process, which adjusts matrices to meet target row and column sums, is a longstanding problem that lacks a general closed form for its limit. While Nathanson derived the closed form for the Sinkhorn limit of $2\times 2$ matrices when target row and column sums are 1, and recent work by Rowland and Wu has advanced understanding

Alessandro Ferreri, Vincenzo Macrì, Yoshihiko Hasegawa, David Edward Bruschi

Quantum optomechanics describes the interaction between a confined field and a fluctuating wall due to radiation pressure. The dynamics of this system is typically understood using perturbation theory up to second order in the small coupling. Improving beyond this regime can shed light onto new phenomena. In this work we study high-order resonant wall-field

Changhong Li

We exploit forecast sensitivities to high frequency gravitational waves (HFGWs) from superconducting LC circuits, traditional resonant cavity and superconducting radio frequency (SRF) cavities with electromagnetic and mechanical modes to derive the first projections of the bounce energy scale within the generic bouncing cosmology framework over the frequency

Kausthubh Chandramouli, Kelly Mae Allen, Christopher Mori, Dror Baron

In the noisy intermediate-scale quantum (NISQ) era, quantum error mitigation (QEM) is essential for producing reliable outputs from quantum circuits. We present a statistical signal processing approach to QEM that estimates the most likely noiseless outputs from noisy quantum measurements. Our model assumes that circuit depth is sufficient for depolarizing n

Ali Murad, Bo Hui, Wei-Shinn Ku

Federated Learning (FL) is a distributed framework for collaborative model training over large-scale distributed data, enabling higher performance while maintaining client data privacy. However, the nature of model aggregation at the centralized server can result in a performance drop in the presence of non-IID data across different clients. We remark that t

Vahid Garousi, Zafar Jafarov, Aytan Movsumova, Atif Namazov

Context: As generative AI (GenAI) tools such as ChatGPT and GitHub Copilot become pervasive in education, concerns are rising about students using them to complete rather than learn from coursework-risking overreliance, reduced critical thinking, and long-term skill deficits. Objective: This paper proposes and empirically applies a causal model to help educa

Batuhan Hangun, Emine Akpinar, Oguz Altun, Onder Eyecioglu

Quantum Machine Learning (QML) is an emerging field at the intersection of quantum computing and machine learning, aiming to enhance classical machine learning methods by leveraging quantum mechanics principles such as entanglement and superposition. However, skepticism persists regarding the practical advantages of QML, mainly due to the current limitations

Qingyi Lu, Haotian Lyu, Jiayun Zheng, Yang Wang

As e-commerce platforms expand their product catalogs, accurately recommending long-tail items becomes increasingly important for enhancing both user experience and platform revenue. A key challenge is the long-tail problem, where extreme data sparsity and cold-start issues limit the performance of traditional recommendation methods. To address this, we prop

Anton Tausendfreund, Sarah Schreyer, Florian Immig, Ulrich Oberhofer

Natural gas supplies in Europe were disrupted and energy prices soared in the context of Russia's invasion of Ukraine. Electricity prices in France experienced the largest relative increase among European countries, even though natural gas plays a negligible role in the French electricity system. In this article, we demonstrate the importance of causal stati

Yunguan Fu, Wenjia Bai, Weixi Yi, Charlotte Manisty

Here we present a versatile foundation model that can perform a range of clinically-relevant image analysis tasks, including segmentation, landmark localisation, diagnosis, and prognostication. A multi-view convolution-transformer masked autoencoder, named as CineMA, was trained on 15 million cine images from 74,916 subjects. The model was validated on multi

Shalin Patil, Catalin Gainaru, Roland Böhmer, Shiwang Cheng

Non-hydrogen bonding van der Waals liquids with dipole-dipole interactions are typically viewed as non-associative and not considered able to sustain large supramolecular structures. Combining broadband dielectric spectroscopy (BDS) and rheology, we demonstrate the supramolecular formation in a group of non-hydrogen-bonding van der Waals liquids, i.e. 1-brom

Uddalok Sarkar, Sourav Chakraborty, Kuldeep S. Meel

Randomized algorithms depend on accurate sampling from probability distributions, as their correctness and performance hinge on the quality of the generated samples. However, even for common distributions like Binomial, exact sampling is computationally challenging, leading standard library implementations to rely on heuristics. These heuristics, while effic

Saad Hossain, Samanvay Vajpayee, Sirisha Rambhatla

As large language models (LLMs) become ubiquitous, parameter-efficient fine-tuning methods and safety-first defenses have proliferated rapidly. However, the number of approaches and their recent increase have resulted in diverse evaluations-varied datasets, metrics, and inconsistent threat settings-making it difficult to fairly compare safety, utility, and r

Zhiwei Zhang, Samy Wu Fung, Anastasios Kyrillidis, Stanley Osher

The Boolean satisfiability (SAT) problem lies at the core of many applications in combinatorial optimization, software verification, cryptography, and machine learning. While state-of-the-art solvers have demonstrated high efficiency in handling conjunctive normal form (CNF) formulas, numerous applications require non-CNF (hybrid) constraints, such as XOR, c

Eun Hye Lee, Taeho Jo

Alzheimer's disease (AD) dementia is the most common form of dementia. With the emergence of disease-modifying therapies, predicting disease risk before symptom onset has become critical. We introduce DuAL-Net, a hybrid deep learning framework for AD dementia prediction using whole genome sequencing (WGS) data. DuAL-Net integrates two components: local proba

Aminu Ma'aruf Nass, Kassimu Mpungu, Rahmatullah Ibrahim Nuruddeen

This paper uses Lie symmetry analysis to investigate the biharmonic heat equation on a generalized surface of revolution. We classify the Lie point symmetries associated with this equation, allowing for the identification of surfaces and the corresponding infinitesimal generators. In a significant move, we demonstrate that the biharmonic heat equation on a s

Yuelyu Ji, Hang Zhang, Shiven Verma, Hui Ji

We propose DeepRAG, a novel framework that integrates DeepSeek hierarchical question decomposition capabilities with RAG Gym unified retrieval-augmented generation optimization using process level supervision. Targeting the challenging MedHopQA biomedical question answering task, DeepRAG systematically decomposes complex queries into precise sub-queries and

Vladislav V. Kravchenko

A variety of inverse Sturm-Liouville problems is considered, including the two-spectrum inverse problem, the problem of recovering the potential from the Weyl function, as well as the recovery from the spectral function. In all cases the potential in the Sturm-Liouville equation is assumed to be complex valued. A unified approach for the approximate solution

Adina D. Feinstein, Richard A. Booth, Jennifer B. Bergner, Joshua D. Lothringer

Measuring a single elemental ratio (e.g., carbon-to-oxygen) provides insufficient information for understanding the formation mechanisms and evolution that affect our observations of gas giant planet atmospheres. Although the fields of planet formation, protoplanetary disks, and exoplanets are well established and interconnected, our understanding of how to

Martin Kuo, Jianyi Zhang, Aolin Ding, Louis DiValentin

Malicious attackers can exploit large language models (LLMs) by engaging them in multi-turn dialogues to achieve harmful objectives, posing significant safety risks to society. To address this challenge, we propose a novel defense mechanism: SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues (STREAM). STREAM defends LLMs against multi-turn attac

Vasilii Korolkov

Robust scene segmentation and keyframe extraction are essential preprocessing steps in video understanding pipelines, supporting tasks such as indexing, summarization, and semantic retrieval. However, existing methods often lack generalizability across diverse video types and durations. We present a unified, adaptive framework for automatic scene detection a

B. Barsbay

The full widths of the vector charmonium and bottomonium hybrid mesons $H_{ \mathrm{c}}$ and $H_{\mathrm{b}}$, characterized by the quantum numbers $1^{ \mathrm{--}}$, are determined by analyzing their dominant strong decay modes: $H_{\mathrm{c}} \to D^{+}D^{-}$, $D_{0}\overline{D}_{0}$, $ D_{s}^{+}D_{s}^{-} $ and $H_{\mathrm{b}} \to B^{+}B^{-}$, $B_{0}\over

Yash Tiwari, Owais Ahmad Lone, Mayukha Pal

Ontologies are pivotal for structuring knowledge bases to enhance question answering (QA) systems powered by Large Language Models (LLMs). However, traditional ontology creation relies on manual efforts by domain experts, a process that is time intensive, error prone, and impractical for large, dynamic knowledge domains. This paper introduces OntoRAG, an aut