May 2025 arXiv papers — page 39

Nitin Jha, Abhishek Parakh, Mahadevan Subramaniam

Quantum key distribution (QKD) will most likely be an integral part of any practical quantum network in the future. However, not all QKD protocols can be used in today's networks because of the lack of single-photon emitters and noisy intermediate quantum hardware. Attenuated-photon transmission, typically used to simulate single-photon emitters, severely li

Alessio Di Santo

This report presents a comprehensive analysis of a malicious software sample, detailing its architecture, behavioral characteristics, and underlying intent. Through static and dynamic examination, the malware core functionalities, including persistence mechanisms, command-and-control communication, and data exfiltration routines, are identified and its suppo

Cheng Luo, Jianghui Wang, Bing Li, Siyang Song

In this paper, we introduce Online Multimodal Conversational Response Generation (OMCRG), a novel task designed to produce synchronized verbal and non-verbal listener feedback online, based on the speaker's multimodal inputs. OMCRG captures natural dyadic interactions and introduces new challenges in aligning generated audio with listeners' facial responses.

Skyler Wu, Shihao Yang, S. C. Kou

In the era of AI, neural networks have become increasingly popular for modeling, inference, and prediction, largely due to their potential for universal approximation. With the proliferation of such deep learning models, a question arises: are leaner statistical methods still relevant? To shed insight on this question, we employ the mechanistic nonlinear ord

Ioannis Bantzis, James B. Simon, Arthur Jacot

When a deep ReLU network is initialized with small weights, gradient descent (GD) is at first dominated by the saddle at the origin in parameter space. We study the so-called escape directions along which GD leaves the origin, which play a similar role as the eigenvectors of the Hessian for strict saddles. We show that the optimal escape direction features a

Andrei Khrennikov, Atsushi Iriki, Irina Basieva

Quantum-like (QL) modeling, one of the outcomes of the quantum information revolution, extends quantum theory methods beyond physics to decision theory and cognitive psychology. While effective in explaining paradoxes in decision making and effects in cognitive psychology, such as conjunction, disjunction, order, and response replicability, it lacks a direct

Kyurae Kim, Yi-An Ma, Trevor Campbell, Jacob R. Gardner

We prove that, given a mean-field location-scale variational family, black-box variational inference (BBVI) with the reparametrization gradient converges at a rate that is nearly independent of explicit dimension dependence. Specifically, for a $d$-dimensional strongly log-concave and log-smooth target, the number of iterations for BBVI with a sub-Gaussian f

Martino Andrea Scarpolini, Giovanni Vagnoli, Fabio Guglietta, Roberto Verzicco

Aortic valve replacement is a key surgical procedure for treating aortic valve pathologies, such as stenosis and regurgitation. The precise placement of the prosthetic valve relative to the native aortic annulus plays a critical role in the post-operative hemodynamics. This study investigates how the positioning of a biological prosthetic valve -- either int

Vanessa Utz

As generative AI systems become widely adopted, they enable unprecedented creation levels of synthetic data across text, images, audio, and video modalities. While research has addressed the energy consumption of model training and inference, a critical sustainability challenge remains understudied: digital waste. This term refers to stored data that consume

J Morava

The principal result of this note is the existence of a complex topological orientation for Atiyah-Segal $\mathbb{T}$-equivariant K-theory which indexes the projective space of lines in complex (n+1)-space by the Fourier expansion $1 + q + \dots + q^n$.

Fabricio Dos Santos

In 2022, Osajda and Przytycki showed that any Coxeter group $W$ is biautomatic. Key to their proof is the notion of voracious projection of an element $g \in W$, which is used iteratively to construct a biautomatic structure for $W$: the voracious language. In this article, we generalize these two notions by defining them for any Garside shadow $B$ in a Coxe

Mónika Farsang, Ramin Hasani, Daniela Rus, Radu Grosu

We present LrcSSM, a $\textit{non-linear}$ recurrent model that processes long sequences as fast as today's linear state-space layers. By forcing its Jacobian matrix to be diagonal, the full sequence can be solved in parallel, giving $\mathcal{O}(TD)$ computational work and memory and only $\mathcal{O}(\log T)$ sequential depth, for input-sequence length $T$

Soumen Roy, Bruno Bertrand, Justin Janquart

The detection of gravitational waves opened up a new window to look into the Universe by probing phenomena invisible through electromagnetic observations. As gravitational waves interact very weakly with matter, their detection is challenging and expensive. So far, they have been observed in the nHz frequency and audible ranges. Future detectors are expected

Md. Zahid Hossain, Mustofa Ahmed, Most. Sharmin Sultana Samu, Md. Rakibul Islam

The automated generation of radiology reports from chest X-ray images holds significant promise in enhancing diagnostic workflows while preserving patient privacy. Traditional centralized approaches often require sensitive data transfer, posing privacy concerns. To address this, the study proposes a Multimodal Federated Learning framework for chest X-ray rep

Reed A. Foster, Stephen Kandeh, Owen Medeiros, Alejandro Simon

Novel electronic devices can often be operated in a plethora of ways, which makes testing circuits comprised of them difficult. Often, no single tool can simultaneously analyze the operating margins, maximum speed, and failure modes of a circuit, particularly when the intended behavior of subcomponents of the circuit is not standardized. This work demonstrat

Matias Blaña, Thomas H. Puzia, Yasna Ordenes-Briceño, Patricia B. Tissera

Satellite galaxies endure powerful environmental tidal forces that drive mass stripping of their outer regions. Consequently, satellites located in central regions of galaxy clusters or groups, where the tidal field is strongest, are expected to retain their central dense regions while losing their outskirts. This process produces a spatial segregation in th

Liang-Hong Mo, Roderich Moessner, Hongzheng Zhao

Emission and absorption of energy are fundamental aspects of non-equilibrium dynamics. The heating induced by driving a many-body system is perhaps the most straightforward diagnostic of the process of equilibration, or the lack thereof. Gapless systems are particularly susceptible to drive-induced heating, and the capacity to control such heating is of expe

Aranyak Sarkar

We present a continuous-space realization of the Coupled Memory Graph Process (CMGP), a minimal non-Markovian framework in which coherence emerges through internal feedback. A single Brownian particle evolves on a viscoelastic substrate that records its trajectory as a scalar memory field and exerts local forces via the gradient $\nabla$ of accumulated impri

Barbarestani Baran, Maks Isa, Vossen Piek

This study evaluates the effectiveness of ChatGPT, an advanced AI model for natural language processing, in identifying targeting and inappropriate language in online comments. With the increasing challenge of moderating vast volumes of user-generated content on social network sites, the role of AI in content moderation has gained prominence. We compared Cha

Y. Chapovskyi, A. Petravchuk

Let $\mathbb K$ be an algebraically closed field of characteristic zero, $A = \mathbb K[x_1,\dots,x_n]$ the polynomial ring, and let $W_n(\mathbb K)$ denote the Lie algebra of all $\mathbb K$-derivations on $A$. The Lie algebra $W_n := W_n(\mathbb K)$ admits a natural grading $W_n = \bigoplus_{i \ge -1} W^{[i]}_n$, where $W^{[i]}_n$ consists of all homogeneo

A. de la C. Rangel-Pantoja, I. Díaz-Saldaña, Carlos A. Vaquera-Araujo

The canonical quantization of a field theory for spin-$1/2$ massive bosons that satisfy the Klein-Gordon equation is presented. The breakdown of the usual spin-statistics connection is due to the redefinition of the dual field, rendering the theory pseudo-Hermitian. The normal-ordered Hamiltonian is bounded from below with real eigenvalues, and the theory is

Ludovic Huguet, Victor Lherm, Renaud Deguen, Joris Heyman

Turbulent thermals emerge in a wide variety of geophysical and industrial flows, such as atmospheric cumulus convection and pollutant dispersal in oceans and lakes. When a buoyant fluid mass rises, or sinks, heat and mass transfers occur by the engulfment of the fresh surrounding fluid inside the thermal - a process that spans over multiple scales from macro

Gonzalo Travieso, Joao Merenda, Odemir M. Bruno

Network models have been widely used to study diverse systems and analyze their dynamic behaviors. Given the structural variability of networks, an intriguing question arises: Can we infer the type of system represented by a network based on its structure? This classification problem involves extracting relevant features from the network. Existing literature

Brian K. Tran, Ben S. Southworth, Hannah F. Blumhoefer, Samuel Olivier

Achieving robust control and optimization in high-fidelity physics simulations is extremely challenging, especially for evolutionary systems whose solutions span vast scales across space, time, and physical variables. In conjunction with gradient-based methods, adjoint systems are widely used in the optimization of systems subject to differential equation co

Sandra Schulz, Juliane Jacob

This article focuses on how data literacy education such as research data management skills can be integrated into teacher training programmes in order to adequately train the teachers of tomorrow. To this end, interviews were conducted with three lecturers from the Faculty of Education and analysed both qualitatively and quantitatively. The lecturers descri

Julia Boone, Tolunay Seyfi, Fatemeh Afghah

Internet of Vehicles (IoV) systems, while offering significant advancements in transportation efficiency and safety, introduce substantial security vulnerabilities due to their highly interconnected nature. These dynamic systems produce massive amounts of data between vehicles, infrastructure, and cloud services and present a highly distributed framework wit

Raoyuan Zhao, Abdullatif Köksal, Ali Modarressi, Michael A. Hedderich

The reliability of large language models (LLMs) is greatly compromised by their tendency to hallucinate, underscoring the need for precise identification of knowledge gaps within LLMs. Various methods for probing such gaps exist, ranging from calibration-based to prompting-based methods. To evaluate these probing methods, in this paper, we propose a new proc

Sinchana Ramakanth Bhat, Max Rudat, Jannis Spiekermann, Nicolas Flores-Herr

Chunking is a crucial preprocessing step in retrieval-augmented generation (RAG) systems, significantly impacting retrieval effectiveness across diverse datasets. In this study, we systematically evaluate fixed-size chunking strategies and their influence on retrieval performance using multiple embedding models. Our experiments, conducted on both short-form

Zhengbo Zhou, Dooman Arefan, Margarita Zuley, Jules Sumkin

Predicting the risk of developing breast cancer is an important clinical tool to guide early intervention and tailoring personalized screening strategies. Early risk models have limited performance and recently machine learning-based analysis of mammogram images showed encouraging risk prediction effects. These models however are limited to the use of a sing

Xiaoling Hu, Peirong Liu, Dina Zemlyanker, Jonathan Williams Ramirez

Obtaining high-resolution (HR) segmentations from coarse annotations is a pervasive challenge in computer vision. Applications include inferring pixel-level segmentations from token-level labels in vision transformers, upsampling coarse masks to full resolution, and transferring annotations from legacy low-resolution (LR) datasets to modern HR imagery. These

Kerstin Andree, Santiago Berrezueta-Guzman, Stephan Krusche, Luise Pufahl

Soft skills are critical for academic and professional success, but are often neglected in early-stage technical curricula. This paper presents a semi-isolated teaching intervention aimed at fostering study ability and key soft skills-communication, collaboration, and project management-among first-year computer science students. The elective seminar Soft Sk

Ganglou Xu

Federated learning faces critical challenges in balancing communication efficiency and model accuracy. One key issue lies in the approximation of update errors without incurring high computational costs. In this paper, we propose a lightweight yet effective method called Gradient Difference Approximation (GDA), which leverages first-order information to esti

Jianfeng Hou, Siwei Lin, Qinghou Zeng

For an integer $k\ge 2$, let $G$ be a graph with $m$ edges and without cycles of length $2k$. The pivotal Alon-Krivelevich-Sudakov Theorem on Max-Cuts states that $G$ has a bipartite subgraph with at least $m/2+\Omega(m^{(2k+1)/(2k+2)})$ edges. In this paper, we present a bisection variant of it by showing that if $G$ has minimum degree at least $k$, then $G

Raoyuan Zhao, Beiduo Chen, Barbara Plank, Michael A. Hedderich

Large language models (LLMs) are used globally across many languages, but their English-centric pretraining raises concerns about cross-lingual disparities for cultural awareness, often resulting in biased outputs. However, comprehensive multilingual evaluation remains challenging due to limited benchmarks and questionable translation quality. To better asse

Omar Bennouna, Amine Bennouna, Saurabh Amin, Asuman Ozdaglar

We study the fundamental question of how informative a dataset is for solving a given decision-making task. In our setting, the dataset provides partial information about unknown parameters that influence task outcomes. Focusing on linear programs, we characterize when a dataset is sufficient to recover an optimal decision, given an uncertainty set on the co

Sainath Bitragunta

In this article, I construct a new set of functions based on Ramanujan sequences (RSEs), Gaussian pulse (GP), and its delayed Gaussian pulse (DGP). The motivation for this construction is based on the special properties of RSEs, GP, and DGP. First, I present a procedure for constructing Gauss-Ramanujan (GauRam) functions using selected RSEs. I develop an ins

Yanling Chen, Shuping Huang, Qinghou Zeng

This paper studies the following question of Bollob\'as and Scott: Let $G$ be a graph with $n$ vertices and $p\binom{n}{2}$ edges. What is the smallest $c(p, n)$ such that there is an ordering $v_1, \ldots, v_n$ of the vertices in $G$ with $\left|e(\{v_1, \ldots, v_i\})-p\binom{i}{2}\right|\leq c(p, n)$ for all $i\in \{1,\ldots,n\}$ ? We obtain upper and low

Avijit Gayen, Somyajit Chakraborty, Mainak Sen, Soham Paul

The persistent accumulation of unresolved legal cases, especially within the Indian judiciary, significantly hampers the timely delivery of justice. Manual methods of prioritizing petitions are often prone to inefficiencies and subjective biases further exacerbating delays. To address this issue, we propose LLMPR (Large Language Model-based Petition Ranking)

Mustafa A Mohamad, Di Qi

We study the statistical properties of passive tracer transport in turbulent flows with a mean gradient, emphasizing tracer intermittency and extreme events. An analytically tractable model is developed, coupling zonal and shear velocity components with both linear and nonlinear stochastic dynamics. Formulating the model in Fourier space, a simple explicit s

Behtom Adeli, John Mclinden, Pankaj Pandey, Ming Shao

In recent years, deep learning (DL) approaches have demonstrated promising results in decoding hemodynamic responses captured by functional near-infrared spectroscopy (fNIRS), particularly in the context of brain-computer interface (BCI) applications. This work introduces AbsoluteNet, a novel deep learning architecture designed to classify auditory event-rel

Jiemin Li, Yanhong Gu, Takemi Yamada, Zebin Wu

The many body electron-electron interaction in cuprates causes the broadening of the electronic bands in \textit{\textbf{k}}-space, leading to a deviation from the standard Fermi liquid. While a \textit{\textbf{k}}-dependent anisotropic electronic scattering (\textit{\textbf{k}}-DAES) has been assessed by photoemission, its fingerprint in \textit{\textbf{Q}}

Michele Gallo

Tensors provide a robust framework for managing high-dimensional data. Consequently, tensor analysis has emerged as an active research area in various domains, including machine learning, signal processing, computer vision, graph analysis, and data mining. This study introduces an efficient image storage approach utilizing tensors, aiming to minimize memory

Yinfang Chen, Jiaqi Pan, Jackson Clark, Yiming Su

In cloud-scale systems, failures are the norm. A distributed computing cluster exhibits hundreds of machine failures and thousands of disk failures; software bugs and misconfigurations are reported to be more frequent. The demand for autonomous, AI-driven reliability engineering continues to grow, as existing humanin-the-loop practices can hardly keep up wit

Yogev Bar-On, Ilan Komargodski, Omri Weinstein

Proof-of-Work (PoW) consensus is traditionally analyzed under the assumption that all miners incur similar costs per unit of computational effort. In reality, costs vary due to factors such as regional electricity cost differences and access to specialized hardware. These variations in mining costs become even more pronounced in the emerging paradigm of \emp

Joel Lidin, Amir Sarfi, Evangelos Pappas, Samuel Dare

We describe an incentive system for distributed deep learning of foundational models where peers are rewarded for contributions. The incentive system, \textit{Gauntlet}, has been deployed on the bittensor blockchain and used to train a 1.2B LLM with completely permissionless contributions of pseudo-gradients: no control over the users that can register or th

Iliyas Ibrahim Iliyas, Souley Boukari, Abdulsalam Yau Gital

This study introduces a framework that integrates nonlinear feature extraction, classification, and efficient optimization. First, kernel principal component analysis with a radial basis function kernel reduces dimensionality while preserving 95% of the variance. Second, a multilayer perceptron (MLP) learns to predict disease status. Finally, a modified mult

Wolfgang Cassing

In this study we aim for quantifying the role of in-medium 3$\leftrightarrow$3 collisions for systems of fermions which initially are out-off equilibrium. The formulation of the 3-body dynamics is based on the equations of motion method for identical fermions -- also denoted as quantum correlation dynamics -- and presented in detail. The on-shell 2-body coll

Angie Zhang, Madison Liao, Elizaveta, Kravchenko

City governments in the United States are increasingly pressured to adopt emerging technologies. Yet, these systems often risk biased and disparate outcomes. Scholars studying public sector technology design have converged on the need to ground these systems in the goals and organizational contexts of employees using them. We expand our understanding of empl

Sravan Kumar Ankireddy, Heasung Kim, Joonyoung Cho, Hyeji Kim

Despite significant advancements in deep learning based CSI compression, some key limitations remain unaddressed. Current approaches predominantly treat CSI compression as a source-coding problem, thereby neglecting transmission errors. Conventional separate source and channel coding suffers from the cliff effect, leading to significant deterioration in reco

Andrew J. Loza, Jun Yup Kim, Shangzheng Song, Yihang Liu

Real-world processes often generate data that are a mix of categorical and numeric values that are recorded at irregular and informative intervals. Discrete token-based approaches are limited in numeric representation capacity while methods like neural ordinary differential equations are not well suited for categorical data or informative sampling and requir

Thomas Grandits, Stefano Coss, Gundolf Haase

District heating networks play a vital role in thermal energy supply in many countries. Thus, it comes to no surprise that these has been a central role in improving energy efficiency for private and public energy suppliers alike around the globe. Many studies have previously investigated the potential of energy saving by low temperature operation of the DHN

Chang Zhang, Zachary Louis-Seize, Maxime Brazeau, Timothy Hodges

Temperature-based radiation detectors are an essential tool for long optical wavelengths detection even if they often suffer from important bandwidth limitations. Their responsivity, and hence their noise equivalent power (NEP), typically degrade at frequencies exceeding the cutoff set by their characteristic thermal response time ($\tau_\text{th}$), i.e., a

Hung Anh Vu, Galen Reeves, Emily Wenger

The internet serves as a common source of training data for generative AI (genAI) models but is increasingly populated with AI-generated content. This duality raises the possibility that future genAI models may be trained on other models' generated outputs. Prior work has studied consequences of models training on their own generated outputs, but limited wor

Yufeng Yang, Minghao Ning, Keqi Shu, Aladdin Saleh

Autonomous mobility systems increasingly operate in dense and dynamic environments where perception occlusions, limited sensing coverage, and multi-agent interactions pose major challenges. While onboard sensors provide essential local perception, they often struggle to maintain reliable situational awareness in crowded urban or indoor settings. This article

Olfa Dani, Johannes C. Bayer, Timo Wagner, Gertrud Zwicknagl

The interaction between localized spins on a quantum dot and free electrons in the reservoirs forms a many-particle entangled system giving rise to the Kondo effect. Here, we investigate electron transport in the third shell of a gate-defined GaAs quantum dot. The addition energy shows a maximum at half-filling of the shell which can be described analyticall

Michael Katz, Harsha Kokel, Christian Muise, Shirin Sohrabi

In over sixty years since its inception, the field of planning has made significant contributions to both the theory and practice of building planning software that can solve a never-before-seen planning problem. This was done through established practices of rigorous design and evaluation of planning systems. It is our position that this rigor should be app

Doaa Hassan, Mohammad Al Hasan

Predicting the emergence of future research collaborations between authors in academic social networks (SNs) is a very effective example that demonstrates the link prediction problem. This problem refers to predicting the potential existence or absence of a link between a pair of nodes (authors) on the co-authorship network. Various similarity and aggregatio

Yehia Abd Alrahman, Nir Piterman

We consider the problem of distributing a centralised transition system to a set of asynchronous agents recognising the same language. Existing solutions are either manual or involve a huge explosion in the number of states from the centralised system. The difficulty arises from the need to keep a rigid communication scheme, specifying a fixed mapping from e

Davin Choo, Yuqi Pan, Tonghan Wang, Milind Tambe

We study a sequential decision-making problem on a $n$-node graph $\mathcal{G}$ where each node has an unknown label from a finite set $\mathbf{\Omega}$, drawn from a joint distribution $\mathcal{P}$ that is Markov with respect to $\mathcal{G}$. At each step, selecting a node reveals its label and yields a label-dependent reward. The goal is to adaptively ch

Rahul Raman, Khushi Sharma, Sai Qian Zhang

Investigating outliers in large language models (LLMs) is crucial due to their significant impact on various aspects of LLM performance, including quantization and compression. Outliers often cause considerable quantization errors, leading to degraded model performance. Identifying and addressing these outliers can enhance the accuracy and efficiency of the

Yongchao Chen, Yueying Liu, Junwei Zhou, Yilun Hao

Practical guidance on training Large Language Models (LLMs) to leverage Code Interpreter across diverse tasks remains lacking. We present R1-Code-Interpreter, an extension of a text-only LLM trained via multi-turn supervised fine-tuning (SFT) and reinforcement learning (RL) to autonomously generate multiple code queries during step-by-step reasoning. Unlike

Soham Mitra, Praveer Tiwari, Archana Pai

Post-merger gravitational-wave emission from a binary neutron star merger carries crucial information about the equation of state (EoS) of matter at high temperatures. Current gravitational wave detectors have limited sensitivities at post-merger frequencies in the range [1.5, 4] kHz. Therefore, valuable inferences can only be made after combining informatio

Owen Oertell, Shikun Sun, Yiding Chen, Jin Peng Zhou

The controllable generation of diffusion models aims to steer the model to generate samples that optimize some given objective functions. It is desirable for a variety of applications including image generation, molecule generation, and DNA/sequence generation. Reinforcement Learning (RL) based fine-tuning of the base model is a popular approach but it can o

Hyeonseong Jeon, Ainaz Eftekhar, Aaron Walsman, Kuo-Hao Zeng

We introduce LOKI, a compute-efficient framework for co-designing morphologies and control policies that generalize across unseen tasks. Inspired by biological adaptation -- where animals quickly adjust to morphological changes -- our method overcomes the inefficiencies of traditional evolutionary and quality-diversity algorithms. We propose learning converg

Joe O'Brien, Jeremy Dolan, Jay Kim, Jonah Dykhuizen

Our survey of 53 specialists across 105 AI reliability and security research areas identifies the most promising research prospects to guide strategic AI R&D investment. As companies are seeking to develop AI systems with broadly human-level capabilities, research on reliability and security is urgently needed to ensure AI's benefits can be safely and broadl

Houcine Meftahi, Chayma Nssibi

We study the inverse problem of recovering the spatial support of parameter variations in a system of partial differential equations (PDEs) from boundary measurements. A reconstruction method is developed based on the monotonicity properties of the Neumann-to-Dirichlet operator, which provides a theoretical foundation for stable support identification. To im

Mateusz Wilinski, Anubha Goel, Alexandros Iosifidis, Juho Kanniainen

The rapid development of sophisticated machine learning methods, together with the increased availability of financial data, has the potential to transform financial research, but also poses a challenge in terms of validation and interpretation. A good case study is the task of classifying financial investors based on their behavioral patterns. Not only do w

Yue Guan, Yuanwei Fang, Keren Zhou, Corbin Robeck

In this work, we propose KPerfIR, a novel multilevel compiler-centric infrastructure to enable the development of customizable, extendable, and portable profiling tools tailored for modern artificial intelligence (AI) workloads on modern GPUs. Our approach integrates profiling capabilities directly into the compiler workflow, allowing profiling functionaliti

Xiaojie Xu, Xinli Xu, Sirui Chen, Haoyu Chen

Visual presentations are vital for effective communication. Early attempts to automate their creation using deep learning often faced issues such as poorly organized layouts, inaccurate text summarization, and a lack of image understanding, leading to mismatched visuals and text. These limitations restrict their application in formal contexts like business a

Isabel Pederneiras, Alexis Finoguenov, Eduardo Cypriano, Johan Comparat

Scaling relations between galaxy cluster properties are crucial for understanding cosmology and baryonic physics. Rigorous calibration of the $M-L_X$ relation, employing weak lensing mass and consistent statistical methodology, is challenging due to heterogeneous cluster samples. The release of LEGACY imaging data introduced the possibility of unifying the c

Brandon R. Feng, David Keetae Park, Xihaier Luo, Arantxa Urdangarin

Fitting Gaussian Processes (GPs) provides interpretable aleatoric uncertainty quantification for estimation of spatio-temporal fields. Spatio-temporal deep learning models, while scalable, typically assume a simplistic independent covariance matrix for the response, failing to capture the underlying correlation structure. However, spatio-temporal GPs suffer

Zeinab Dehghani, Mohammed Naveed Akram, Koorosh Aslansefat, Adil Khan

Large Language Models (LLMs) such as GPT, LLaMA, and Claude achieve remarkable performance in text generation but remain opaque in their decision-making processes, limiting trust and accountability in high-stakes applications. We present gSMILE (generative SMILE), a model-agnostic, perturbation-based framework for token-level interpretability in LLMs. Extend

Stéphane Ballet, Robert Rolland

Let us consider a generalized Artin-Schreier algebraic function field extension $F$ of the rational function field $\F_{p^n}(x)$ defined over the finite field extension $K=\F_{p^n}$ of the prime field $\F_p$. We assume that $K$ is algebraically closed in $F$. We give general results on the descent over the fields $k= \F_{p^t}$ for $t$ dividing $n$. Then, we

Arkabrata Ghosh

This article considers the family of elliptic curves given by $E_{pq}: y^2=x^3-5pqx$ and certain conditions on odd primed $p$ and $q$. More specifically, we have proved that if $p \equiv 33 \pmod {40}$ and $ q \equiv 7 \pmod {40}$, then the rank of $E_{pq}$ is zero over both $ \mathbb{Q} $ and $ \mathbb{Q}(i) $. Furthermore, if the primes $ p $ and $q$ are o

Ke Zhang, Cihan Xiao, Jiacong Xu, Yiqun Mei

Recent video diffusion models have demonstrated their great capability in generating visually-pleasing results, while synthesizing the correct physical effects in generated videos remains challenging. The complexity of real-world motions, interactions, and dynamics introduce great difficulties when learning physics from data. In this work, we propose DiffPhy

Yifan Yin, Zhengtao Han, Shivam Aarya, Jianxin Wang

Fine-grained robot manipulation, such as lifting and rotating a bottle to display the label on the cap, requires robust reasoning about object parts and their relationships with intended tasks. Despite recent advances in training general-purpose robot manipulation policies guided by language instructions, there is a notable lack of large-scale datasets for f

Nikola Surjanovic, Alexandre Bouchard-Côté, Trevor Campbell

The learning rate is an important tuning parameter for stochastic gradient descent (SGD) and can greatly influence its performance. However, appropriate selection of a learning rate schedule across all iterations typically requires a non-trivial amount of user tuning effort. To address this, we introduce AutoSGD: an SGD method that automatically determines w

Symphony Chakraborty, Hsien Shang

The study of shear layer instability in compressible flows is key to understanding phenomena from aerodynamics to astrophysical jets. Blumen's seminal paper [``Shear layer instability of an inviscid compressible fluid," J. Fluid Mech. {\bf 40}, 769--781 (1970)] established a linear stability framework for inviscid compressible shear flows, emphasizing veloci

Alif Ahmed, Minhajur Rahman, Mohammad Jawad Chowdhury, Khandakar Abdulla Al Mamun

The growing adoption of electric vehicles (EVs) necessitates the development of efficient and reliable charging infrastructure, particularly fast charging stations (FCS) for addressing challenges such as range anxiety and long charging times. This paper presents the design and feasibility analysis of a grid-connected DC fast charging station for the Dhaka-Ch

Eric Xing, Abby Stylianou, Robert Pless, Nathan Jacobs

Massive-scale pretraining has made vision-language models increasingly popular for image-to-image and text-to-image retrieval across a broad collection of domains. However, these models do not perform well when used for challenging retrieval tasks, such as instance retrieval in very large-scale image collections. Recent work has shown that linear transformat

Lei Zhang, Markus Stricker

The discovery and optimization of materials for specific applications is hampered by the practically infinite number of possible elemental combinations and associated properties, also known as the `combinatorial explosion'. By nature of the problem, data are scarce and all possible data sources should be used. In addition to simulations and experimental resu

Kenneth Ball, Erin Taylor, Nirav Patel, Andrew Bartels

Advances in machine learning, especially the introduction of transformer architectures and vision transformers, have led to the development of highly capable computer vision foundation models. The segment anything model (known colloquially as SAM and more recently SAM 2), is a highly capable foundation model for segmentation of natural images and has been fu

B. Topçu, P. Schady, S. Wuyts, A. Inkenhaag

Long gamma-ray bursts (GRBs) serve as powerful probes of distant galaxies. Their luminous afterglow pinpoints galaxies independent of luminosity, in contrast to most flux-limited surveys. Nevertheless, GRB-selected galaxy samples are not free from bias, instead tracing the conditions favoured by the progenitor stars. Characterising the galaxy populations tra

Reid McIlroy-Young

In the near future a governmental body will be asked to allow companies to use AI for age verification. If they allow it the resulting system will both be easily circumvented and disproportionately misclassify minorities and low socioeconomic status users. This is predictable by showing that other very similar systems (facial recognition and remote proctorin

Joshua Drexel, Esther Hänggi, Iyán Méndez Veiga

Supply chain attacks have emerged as a prominent cybersecurity threat in recent years. Reproducible and bootstrappable builds have the potential to reduce such attacks significantly. In combination with independent, exhaustive and periodic source code audits, these measures can effectively eradicate compromises in the building process. In this paper we intro

Maresa Schröder, Justin Hartenstein, Stefan Feuerriegel

The average treatment effect (ATE) is widely used to evaluate the effectiveness of drugs and other medical interventions. In safety-critical applications like medicine, reliable inferences about the ATE typically require valid uncertainty quantification, such as through confidence intervals (CIs). However, estimating treatment effects in these settings often

Jiaxing Li, Hanjiang Hu, Yujie Yang, Changliu Liu

Recent learning-based safety filters have outperformed conventional methods, such as hand-crafted Control Barrier Functions (CBFs), by effectively adapting to complex constraints. However, these learning-based approaches lack formal safety guarantees. In this work, we introduce a verifiable model-free safety filter based on Hamilton-Jacobi reachability analy

Oren Mangoubi, Neil He, Nisheeth K. Vishnoi

We introduce a framework for designing efficient diffusion models for $d$-dimensional symmetric-space Riemannian manifolds, including the torus, sphere, special orthogonal group and unitary group. Existing manifold diffusion models often depend on heat kernels, which lack closed-form expressions and require either $d$ gradient evaluations or exponential-in-$

Mauricio Junca, Esteban Leiva

The relationship between inverse reinforcement learning (IRL) and inverse optimization (IO) for Markov decision processes (MDPs) has been relatively underexplored in the literature, despite addressing the same problem. In this work, we revisit the relationship between the IO framework for MDPs, IRL, and apprenticeship learning (AL). We incorporate prior beli

Jaehyun Nam, Jinsung Yoon, Jiefeng Chen, Jinwoo Shin

Agents based on large language models (LLMs) for machine learning engineering (MLE) can automatically implement ML models via code generation. However, existing approaches to build such agents often rely heavily on inherent LLM knowledge and employ coarse exploration strategies that modify the entire code structure at once. This limits their ability to selec

Harry Desmond

Modified Newtonian Dynamics (MOND) is an alternative to the dark matter hypothesis that attempts to explain the "missing gravity" problem in astrophysics and cosmology through a modification to objects' dynamics. Since its conception in 1983, MOND has had a chequered history. Some phenomena difficult to understand in standard cosmology MOND explains remarkab

Xiaole Tang, Xiaoyi He, Xiang Gu, Jian Sun

Despite remarkable advances made in all-in-one image restoration (AIR) for handling different types of degradations simultaneously, existing methods remain vulnerable to out-of-distribution degradations and images, limiting their real-world applicability. In this paper, we propose a multi-source representation learning framework BaryIR, which decomposes the

Alexander Nemecek, Yuzhou Jiang, Erman Ayday

Watermarking has emerged as a leading technical proposal for attributing generative AI content and is increasingly cited in global governance frameworks. This position paper argues that current implementations risk serving as symbolic compliance rather than delivering effective oversight. We identify a growing gap between regulatory expectations and the tech

Alexander Nemecek, Yuzhou Jiang, Erman Ayday

Large language models (LLMs) are increasingly integrated into academic workflows, with many conferences and journals permitting their use for tasks such as language refinement and literature summarization. However, their use in peer review remains prohibited due to concerns around confidentiality breaches, hallucinated content, and inconsistent evaluations.

Haoqian Liang, Xiaohui Wang, Zhichao Li, Ya Yang

Object concepts play a foundational role in human visual cognition, enabling perception, memory, and interaction in the physical world. Inspired by findings in developmental neuroscience - where infants are shown to acquire object understanding through observation of motion - we propose a biologically inspired framework for learning object-centric visual rep

Chengyu Yang, Chengjun Liu

Laparoscopic surgeries often suffer from reduced visual clarity due to the presence of surgical smoke originated by surgical instruments, which poses significant challenges for both surgeons and vision based computer-assisted technologies. In order to remove the surgical smoke, a novel U-Net deep learning with new loss function and integrated differentiable

Fridrik Freyr Gautason, Jesse van Muiden

We discuss that the string/M-theory partition function requires a choice of ensembles, depending on which background fields are held fixed. The background fields correspond to worldvolume couplings in the effective action approach to the superstring, which we extrapolate to the M2-brane. One natural ensemble in this context, which we call the M2-ensemble, co

The CTAO-LST Project, :, K. Abe, S. Abe

Geminga is the third gamma-ray pulsar firmly detected by imaging atmospheric Cherenkov telescopes (IACTs) after the Crab and the Vela pulsars. Most of its emission is expected at tens of GeV, and, out of the planned telescopes of the upcoming Cherenkov Telescope Array Observatory (CTAO), the Large-Sized Telescopes (LSTs) are the only ones with optimised sens

Jahmall Bersini, Simeon Hellerman, Domenico Orlando, Susanne Reffert

We study the fixed point of the three-dimensional NJL model in a double-scaling limit where both the charge $Q$ and the number of fermion flavors $N$ become large with a fixed ratio $q=Q/(2N)$. While a similar analysis has been performed for the bosonic O(N) model, fermionic models pose new challenges. In this work, we systematically explore the CFT spectrum

Maryam Dehghanian, Nahum Arav, Mayank Sharma, Gwen Walker

Absorption outflows in quasars play an important role in understanding active galactic nuclei (AGN) feedback and their influence on galaxy evolution. The unprecedented spectral data provided by the Dark Energy Spectroscopic Instrument (DESI) opens new avenues to explore these outflows. We analyze five low-ionization absorption outflow systems in four interme