May 2025 arXiv papers — page 61

Ruihang Wang, Qingang Zhang, Yonggang Wen, Stuart Kennedy

The revolution in artificial intelligence (AI) has brought sustainable challenges in data center management due to the high carbon emissions and short cooling response time associated with high-power density racks. While machine learning (ML) offers promise for intelligent management, its adoption is hindered by safety and reliability concerns. To address th

Michael Bersudsky, Nimish A. Shah

We study the asymptotic distribution of norm ball averages along orbits of a lattice $\Gamma \subset \text{SO}(n,1)$ acting on the moduli space of pairs of orthogonal discrete subgroups of $\mathbb{R}^{n+1}$ up to homothety. Our main result shows that, except for special $2$-lattices in $\mathbb{R}^3$ lying in hyperplanes tangent to the light cone, these mea

Nagare Katayama, Yuya Tanizaki

The Cardy-Rabinovici model is a toy model of the lattice $U(1)$ gauge theories to study various oblique confinement states associated with the nonzero $\theta$ angles. We reformulate the $2$d version of this model using the modified Villain lattice formalism, and we establish the exact $\theta$ periodicity for the Witten effect and the strong-weak duality at

Xiang Shi, Rui Zhang, Jiawei Liu, Yinpeng Liu

Multimodal fusion is susceptible to modality imbalance, where dominant modalities overshadow weak ones, easily leading to biased learning and suboptimal fusion, especially for incomplete modality conditions. To address this problem, we propose a Shapley-guided alternating training framework that adaptively prioritizes minor modalities to balance and thus enh

Yu Tang, Carlo Cenedese, Alessio Rimoldi, Florian Dórfler

The paper introduces Split-as-a-Pro, a control framework that integrates behavioral systems theory, operator splitting methods, and alternating projection algorithms. The framework reduces dynamic optimization problems - arising in both control and estimation - to efficient projection computations. Split-as-a-Pro builds on a non-parametric formulation that e

Jiajun Zhu, Ye Liu, Meikai Bao, Kai Zhang

Recently, large language models (LLMs) have demonstrated remarkable capabilities in natural language processing tasks, yet they remain prone to hallucinations when reasoning with insufficient internal knowledge. While integrating LLMs with knowledge graphs (KGs) provides access to structured, verifiable information, existing approaches often generate incompl

Ruihang Wang, Minghao Li, Zhiwei Cao, Jimin Jia

The explosion in artificial intelligence (AI) applications is pushing the development of AI-dedicated data centers (AIDCs), creating management challenges that traditional methods and standalone AI solutions struggle to address. While digital twins are beneficial for AI-based design validation and operational optimization, current AI methods for their creati

Lu Yi, Runlin Lei, Fengran Mo, Yanping Zheng

Future link prediction on temporal graphs is a fundamental task with wide applicability in real-world dynamic systems. These scenarios often involve both recurring (seen) and novel (unseen) interactions, requiring models to generalize effectively across both types of edges. However, existing methods typically rely on complex memory and aggregation modules, y

Jingyi Chen

Let $L$ be a compact oriented Lagrangian surface in a K\"ahler surface endowed with a complete Riemannian metric (compatible with the symplectic structure and the complex structure) with bounded sectional curvatures and a positive lower bound on injectivity radius. We show that for every nontrivial class $[\gamma]$ of the fundamental group $\pi_1(L)$ such th

Tianle Li, Jihai Zhang, Yongming Rao, Yu Cheng

While large language models (LLMs) demonstrate strong reasoning capabilities utilizing reinforcement learning (RL) with verifiable reward, whether large vision-language models (VLMs) can directly inherit such capabilities through similar post-training strategies remains underexplored. In this work, we conduct a systematic compositional probing study to evalu

Yan Wen, Junfeng Guo, Heng Huang

As large language models (LLMs) evolve into autonomous agents capable of collaborative reasoning and task execution, multi-agent LLM systems have emerged as a powerful paradigm for solving complex problems. However, these systems pose new challenges for copyright protection, particularly when sensitive or copyrighted content is inadvertently recalled through

Yuta Ono, Hiroshi Nakamura, Hideki Takase

Federated Active Learning (FAL) seeks to reduce the burden of annotation under the realistic constraints of federated learning by leveraging Active Learning (AL). As FAL settings make it more expensive to obtain ground truth labels, FAL strategies that work well in low-budget regimes, where the amount of annotation is very limited, are needed. In this work,

Camille Mondon, Huong Thi Trinh, Anne Ruiz-Gazen, Christine Thomas-Agnan

Invariant coordinate selection (ICS) is a dimension reduction method, used as a preliminary step for clustering and outlier detection. It has been primarily applied to multivariate data. This work introduces a coordinate-free definition of ICS in an abstract Euclidean space and extends the method to complex data. Functional and distributional data are prepro

Tai-Quan Peng, Xuzhen Yang

This paper examines how large language models (LLMs) are transforming core quantitative methods in communication research in particular, and in the social sciences more broadly-namely, content analysis, survey research, and experimental studies. Rather than replacing classical approaches, LLMs introduce new possibilities for coding and interpreting text, sim

Jangyeon Kim, Ui-Hyeop Shin, Jaehyun Ko, Hyung-Min Park

This paper presents an efficient speech enhancement (SE) approach that reuses a processing block repeatedly instead of conventional stacking. Rather than increasing the number of blocks for learning deep latent representations, repeating a single block leads to progressive refinement while reducing parameter redundancy. We also minimize domain transformation

Carlos Conde-Ocazionez, Tuojin Yin, Jaquelyn Noronha-Hostler, Nicolás Yunes

Future X-ray missions, such as NICER and LOFT, together with gravitational-wave observations from ground-based detectors, will provide new insights into neutron stars. Interpreting accurate observations in the future will require accurate models of their gravitational fields. In this first paper of a two-part series, we construct the perturbation equations f

Juan Sosa, Eleni Dilma, Brenda Betancourt

We introduce a flexible Bayesian framework for clustering nodes in undirected binary networks, motivated by the need to uncover structural patterns in complex environments. Building on the stochastic block model, we develop two hybrid extensions: the Class-Distance Model, which governs interaction probabilities through Euclidean distances between cluster-lev

Yiwei Xie, Ping Liu, Zheng Zhang

Text-to-Image (T2I) models have demonstrated impressive capabilities in generating high-quality and diverse visual content from natural language prompts. However, uncontrolled reproduction of sensitive, copyrighted, or harmful imagery poses serious ethical, legal, and safety challenges. To address these concerns, the concept erasure paradigm has emerged as a

Jiawen Zhang, Zhenwei Zhang, Shun Zheng, Xumeng Wen

Time-Series Foundation Models (TSFMs) are rapidly transitioning from research prototypes to core components of critical decision-making systems, driven by their impressive zero-shot forecasting capabilities. However, as their deployment surges, a critical blind spot remains: their fragility under adversarial attacks. This lack of scrutiny poses severe risks,

Futoshi Futami, Atsushi Nitanda

Calibration is a critical requirement for reliable probabilistic prediction, especially in high-risk applications. However, the theoretical understanding of which learning algorithms can simultaneously achieve high accuracy and good calibration remains limited, and many existing studies provide empirical validation or a theoretical guarantee in restrictive s

Ethan TS. Liu, Austin Wang, Spencer Mateega, Carlos Georgescu

Ensuring that large language models (LLMs) can effectively assess, detect, explain, and remediate software vulnerabilities is critical for building robust and secure software systems. We introduce VADER, a human-evaluated benchmark designed explicitly to assess LLM performance across four key vulnerability-handling dimensions: assessment, detection, explanat

C. Reichhardt, C. J. O. Reichhardt

We numerically examine the dynamics of a probe particle driven at a constant force through an assembly of particles with competing long-range repulsion and short-range attraction that forms a bubble or stripe state. In the bubble regime, we identify several distinct types of motion, including an elastic or pinned regime where the probe particle remains insid

Alexandru Chirvasitu

For a Coxeter system $(W,S)$ a self-map $τ$ of $W$ such that $τ_{σθ}\in \{τ_θ,\ στ_θ\}$ for all $θ\in W$ and all reflections $σ$ (analogous to being 1-Lipschitz with respect to the Bruhat order on $W$) is either constant or a right translation. A somewhat stronger version holds for $S_n$, where it suffices that $σ$ range over smaller, $θ$-dependent sets of r

Jade Robinson, Jonathan K. Kummerfeld

Code documentation is useful, but writing it is time-consuming. Different techniques for generating code summaries have emerged, but comparing them is difficult because human evaluation is expensive and automatic metrics are unreliable. In this paper, we introduce a simple new baseline in which we ask an LLM to give an overall score to a summary. Unlike n-gr

Tomoro Asai, Yoshihito Kohsaka

We study the Mullins' problem that was proposed by Mullins in 1957 and is one of the models of the thermal grooving by surface diffusion. Mathematically, this is the problem of evolving curves in the half space that is governed by the surface diffusion flow with the contact angle condition and the no-flux condition on the boundary. The no-flux condition is r

Mohammad Cheraghinia, Eli De Poorter, Jaron Fontaine, Merouane Debbah

Wireless Technology Recognition (WTR) and localization are essential in modern communication systems, enabling efficient spectrum management, seamless coexistence of diverse technologies, and accurate positioning in dynamic environments. In real-world conditions, solutions must handle signals from various resources with different sampling rates, capturing de

Jia Wei, Chun Ouyang, Bemali Wickramanayake, Zhipeng He

The global issue of overcrowding in emergency departments (ED) necessitates the analysis of patient flow through ED to enhance efficiency and alleviate overcrowding. However, traditional analytical methods are time-consuming and costly. The healthcare industry is embracing process mining tools to analyse healthcare processes and patient flows. Process mining

Takumi Goto, Yusuke Sakai, Taro Watanabe

We introduce gec-metrics, a library for using and developing grammatical error correction (GEC) evaluation metrics through a unified interface. Our library enables fair system comparisons by ensuring that everyone conducts evaluations using a consistent implementation. Moreover, it is designed with a strong focus on API usage, making it highly extensible. It

Botong Zhang, Shuo Li, Ignacio Hounie, Osbert Bastani

We study the problem of computing an optimal large language model (LLM) policy for the constrained alignment problem, where the goal is to maximize a primary reward objective while satisfying constraints on secondary utilities. Despite the popularity of Lagrangian-based LLM policy search in constrained alignment, iterative primal-dual methods often fail to c

Nate Gillman, Charles Herrmann, Michael Freeman, Daksh Aggarwal

Recent advances in video generation models have sparked interest in world models capable of simulating realistic environments. While navigation has been well-explored, physically meaningful interactions that mimic real-world forces remain largely understudied. In this work, we investigate using physical forces as a control signal for video generation and pro

Jiaqi Guo, Santiago Lopez-Tapia, Aggelos K. Katsaggelos

Limited Angle Computed Tomography (LACT) often faces significant challenges due to missing angular information. Unlike previous methods that operate in the image domain, we propose a new method that focuses on sinogram inpainting. We leverage MR-SDEs, a variant of diffusion models that characterize the diffusion process with mean-reverting stochastic differe

Seokgi Lee, Jungjun Kim

We present the gradual style adaptor TTS (GSA-TTS) with a novel style encoder that gradually encodes speaking styles from an acoustic reference for zero-shot speech synthesis. GSA first captures the local style of each semantic sound unit. Then the local styles are combined by self-attention to obtain a global style condition. This semantic and hierarchical

Varun Reddy, Yen-Ling Kuo

Large language models (LLMs) exhibit strong performance on factual recall and general reasoning but struggle to adapt to user-specific, commonsense knowledge, a challenge particularly acute in small-parameter settings where computational efficiency is prioritized. We introduce CaseEdit, a new dataset and generation pipeline for evaluating localized, personal

Albert S. Berahas, Raghu Bollapragada, Shagun Gupta

In this paper, we propose a framework based on the Retrospective Approximation (RA) paradigm to solve optimization problems with a stochastic objective function and general nonlinear deterministic constraints. This framework sequentially constructs increasingly accurate approximations of the true problems which are solved to a specified accuracy via a determ

Anqing Jiang, Yu Gao, Zhigang Sun, Yiru Wang

Research interest in end-to-end autonomous driving has surged owing to its fully differentiable design integrating modular tasks, i.e. perception, prediction and planing, which enables optimization in pursuit of the ultimate goal. Despite the great potential of the end-to-end paradigm, existing methods suffer from several aspects including expensive BEV (bir

Mohamed Kefi, Tien Dat Pham, Thin Nguyen, Mark G. Tjoelker

Olive production is an important tree crop in Mediterranean climates. However, olive yield varies significantly due to climate change. Accurately estimating yield using remote sensing and machine learning remains a complex challenge. In this study, we developed a streamlined pipeline for olive yield estimation in the Kairouan and Sousse governorates of Tunis

Peter Renkel

This paper proposes an experiment designed to distinguish between competing interpretations of quantum mechanics: those that involve wave function collapse and those that assume purely unitary evolution. The experiment tests whether an observer can measure a system without collapsing its wave function. To this end, we introduce the concept of an unconscious

Ji Chen, Yue Wu, Muyang Li, Zhongyi Yuan

Intelligent object detection, which extracts crucial information like targets categories and locations, plays a vital role in emerging technologies including autonomous driving, the Internet of Things, and next-generation mobile communication systems. With the advancement of intelligent object detectors towards higher integration and miniaturization, their p

William Cooperman, Gautam Iyer, James Nolen

Consider a discrete time Markov process $X^\epsilon$ on $\mathbf R^d$ that makes a deterministic jump based on its current location, and then takes a small Gaussian step of variance $\epsilon^2$. We study the behavior of the asymptotic variance as $\epsilon \to 0$. In some situations (for instance if there were no jumps), then the asymptotic variance vanishe

Zichong Meng, Zeyu Han, Xiaogang Peng, Yiming Xie

State-of-the-art text-to-motion generation models rely on the kinematic-aware, local-relative motion representation popularized by HumanML3D, which encodes motion relative to the pelvis and to the previous frame with built-in redundancy. While this design simplifies training for earlier generation models, it introduces critical limitations for diffusion mode

Lance Ying, Almog Hillel, Ryan Truong, Vikash K. Mansinghka

A key feature of human theory-of-mind is the ability to attribute beliefs to other agents as mentalistic explanations for their behavior. But given the wide variety of beliefs that agents may hold about the world and the rich language we can use to express them, which specific beliefs are people inclined to attribute to others? In this paper, we investigate

Peng Gao, Liangyi Zhao

We study the $2k$-th moment of the family of Dirichlet $L$-functions to a fixed prime modulus on the critical line and establish sharp lower bounds for all real $k \geq 0$ and sharp upper bounds for $k$ in the range $0 \leq k \leq 1$.

Josiah Keagy, Haoyu Liu, Junyu Tang, Weilun Tan

We explore the impact of spin disorder associated with structural defects on antiferromagnetic magnon transport by probing the spin-flop transition of Cr2O3 using spin Seebeck effect measurements. By fabricating homoepitaxial Cr2O3 films grown on smooth Cr2O3 crystals, we systematically vary the thickness of the films in which the presence of point defects m

Niloufar Alipour Talemi, Hossein Kashiani, Hossein R. Nowdeh, Fatemeh Afghah

Prompt learning has emerged as a powerful paradigm for adapting vision-language models such as CLIP to downstream tasks. However, existing methods often overfit to seen data, leading to significant performance degradation when generalizing to novel classes or unseen domains. To address this limitation, we propose DiSa, a Directional Saliency-Aware Prompt Lea

Dian Jin, Xin Bing, Yuqian Zhang

This paper studies robust one-shot aggregation for distributed and federated Independent Component Analysis (ICA). In this setting, each client computes a local ICA estimator, while the server aims to recover a common global mixing matrix without accessing raw data. The main difficulty is that local ICA estimators are identifiable only up to signed permutati

Federica Donnini, Matteo Lapucci, Pierluigi Mansueto

In this work, we deal with unconstrained nonlinear optimization problems. Specifically, we are interested in methods carrying out updates possibly along directions not of descent, like Polyak's heavy-ball algorithm. Instead of enforcing convergence properties through line searches and modifications of search direction when suitable safeguards are not sat

Yin-Long Liu, Rui Feng, Jia-Xin Chen, Yi-Ming Wang

Recent breakthroughs in Automatic Speech Recognition (ASR) have enabled fully automated Alzheimer's Disease (AD) detection using ASR transcripts. Nonetheless, the impact of ASR errors on AD detection remains poorly understood. This paper fills the gap. We conduct a comprehensive study on AD detection using transcripts from various ASR models and their sy

Zhihong Pan, Kai Zhang, Yuze Zhao, Yupeng Han

The inherent capabilities of a language model (LM) and the reasoning strategies it employs jointly determine its performance in reasoning tasks. While test-time scaling is regarded as an effective approach to tackling complex reasoning tasks, it incurs substantial computational costs and often leads to "overthinking", where models become trapped in &

Clive Elphick, Quanyu Tang, Shengtong Zhang

Let $A_G $ be the adjacency matrix of a simple graph $ G $, and let $ χ(G) $, $ χ_f(G) $, $ χ_q(G) $, $ ξ(G) $ and $ ξ_f(G) $ denote its chromatic number, fractional chromatic number, quantum chromatic number, orthogonal rank and projective rank, respectively. For $ p \geq 0 $, we define the positive and negative $ p $-energies of $ G $ by $$ \mathcal{E}_p^+

Bartosz Furmanek, Filip Oskar Łanecki, Mateusz Przybylski, Jim Wiseman

Linear relations, defined as submodules of the direct sum of two modules, can be viewed as objects that carry dynamical information and reflect the inherent uncertainty of sampled dynamics. These objects also provide an algebraic structure that enables the definition of subtle invariants for dynamical systems. In this paper, we prove that linear relations de

Jinyang Ni, Yuanjun Jin, Guoqing Chang

The efficient detection of the magnetism in 2D antiferromagnetic (AFM) insulators is crucial for the advancement of 2D AFM spintronics and remains a challenging problem. In this letter, we introduce the magnon nonlinear Hall current, a second-order Hall response of collective spin excitations in ordered magnets, as a novel probe for 2D layered AFM insulators

Boxuan Lyu, Hidetaka Kamigaito, Kotaro Funakoshi, Manabu Okumura

Maximum a posteriori decoding, a commonly used method for neural machine translation (NMT), aims to maximize the estimated posterior probability. However, high estimated probability does not always lead to high translation quality. Minimum Bayes Risk (MBR) decoding offers an alternative by seeking hypotheses with the highest expected utility. Inspired by Qua

Egor Yasinsky

Let $G$ be a finite group and $H\subseteq G$ be its subgroup. We prove that if a smooth del Pezzo surface over an algebraically closed field is $H$-birationally rigid then it is also $G$-birationally rigid, answering a geometric version of Kollár's question in dimension 2 by positive.

Haim Horowitz, Saharon Shelah

Using creature technology, we construct families of Suslin ccc non-sweet forcing notions $\mathbb Q$ such that $ZFC$ is equiconsistent with $ZF+$"every set of reals equals a Borel set modulo the $(\leq \aleph_1)$-closure of the null ideal associated with $\mathbb Q$"+"there is an $ω_1$-sequence of distinct reals". This answers a question of t

Affan Adly Nazri, Zamri Zainal Abidin, Mohamad Ridhauddin Mat Sabri, Zulfazli Rosli

Extensive radio frequency interference (RFI) monitoring is essential in the site selection process before constructing radio astronomy observatories, followed by mitigation strategies to minimize its adverse effects. Malaysia has an enormous prospect for radio astronomy due to its prominent location in the centre of Southeast Asia, but is challenged by its r

Georgy Noarov, Soham Mallick, Tao Wang, Sunay Joshi

Top-$k$ decoding is a widely used method for sampling from LLMs: at each token, only the largest $k$ next-token-probabilities are kept, and the next token is sampled after re-normalizing them to sum to unity. Top-$k$ and other sampling methods are motivated by the intuition that true next-token distributions are sparse, and the noisy LLM probabilities need t

E. R. Cardozo de Oliveira, C. Xiang, C. Borrazás, S. Sandeep

Superlattices presenting a spatial modulation of the elastic properties appear as a main tool to reach the THz regime in nanoacoustic devices. The exploration of alternative materials with multifunctional properties remains a fertile domain of research. In this work, we study the structural, magnetic, and acoustic characteristics of nanometric superlattices

Yunbo Liu, Xukui Qin, Yifan Gao, Xiang Li

Human Activity Recognition (HAR) using wearable sensor data has become a central task in mobile computing, healthcare, and human-computer interaction. Despite the success of traditional deep learning models such as CNNs and RNNs, they often struggle to capture long-range temporal dependencies and contextual relevance across multiple sensor channels. To addre

Dave Benson, Srikanth B. Iyengar, Henning Krause, Julia Pevtsova

For a point $\mathfrak{p}$ in the spectrum of the cohomology ring of a finite group $G$ over a field $k$, we calculate the spectrum for the subcategory of dualisable objects inside the tensor triangulated category of $\mathfrak{p}$-local and $\mathfrak{p}$-torsion objects in the (big) stable module category of the group algebra $kG$.

Iskander Azangulov, Peter Potaptchik, Qinyu Li, Eddie Aamari

Guidance is a cornerstone of modern diffusion models, playing a pivotal role in conditional generation and enhancing the quality of unconditional samples. However, current approaches to guidance scheduling--determining the appropriate guidance weight--are largely heuristic and lack a solid theoretical foundation. This work addresses these limitations on two

Nikhila Kalia, Ryley McConkey, Eugene Yee, Fue-Sang Lien

This study evaluates the generalization performance and representation efficiency (parsimony) of a previously introduced Tensor Basis Kolmogorov-Arnold Network (TBKAN) architecture for data-driven turbulence modeling. The TBKAN framework replaces the multi-layer perceptron (MLP) used in either the standard or modified Tensor Basis Neural Network (TBNN) with

Diana Barseghyan, Juan Bory-Reyes, Baruch Schneider

In this paper, we study the following magnetic Schr\"odinger operator in $\mathbb{R}^3$: \[ H=(i \nabla +A)^2- \tilde{V}, \] where $\tilde{V}$ is non-negative potential supported over the tube built along a curve which is a local deformation of a straight one, and $B:=\mathrm{rot}(A)$ is a non-zero and local (i.e., a compact supported) magnetic field. Based

Amit Chakraborty, Sayyed Farid Ahamed, Sandip Roy, Soumya Banerjee

Machine Learning as a Service (MLaaS) enables users to leverage powerful machine learning models through cloud-based APIs, offering scalability and ease of deployment. However, these services are vulnerable to model extraction attacks, where adversaries repeatedly query the application programming interface (API) to reconstruct a functionally similar model,

Sidney D. V. Williams, Matthew N. Gudorf, Dmitri M. Orlov

Plasma turbulence is a key challenge in understanding transport phenomena in magnetically confined plasmas. This work presents a novel approach using periodic orbit theory to analyze plasma turbulence, identifying fundamental structures that underpin chaotic motion. By applying numerical optimization techniques to the Kuramoto-Sivashinsky equation - a reduce

Yasha Savelyev

To paraphrase, part I constructs a bundle of $A _{\infty}$ categories given the input of a Hamiltonian fibration over a smooth manifold. Here we show that this bundle is generally non-trivial by a sample computation. One principal application is differential geometric, and the other is about algebraic $K$-theory of the integers and the rationals. We find new

Mario Leiva, Noel Ngu, Joshua Shay Kricheli, Aditya Taparia

The deployment of pre-trained perception models in novel environments often leads to performance degradation due to distributional shifts. Although recent artificial intelligence approaches for metacognition use logical rules to characterize and filter model errors, improving precision often comes at the cost of reduced recall. This paper addresses the hypot

Manan Suri, Puneet Mathur, Nedim Lipka, Franck Dernoncourt

The growing capabilities of multimodal large language models (MLLMs) have advanced tasks like chart understanding. However, these models often suffer from hallucinations, where generated text sequences conflict with the provided visual data. To address this, we introduce Post-Hoc Visual Attribution for Charts, which identifies fine-grained chart elements tha

Luran Bengono Mintogo, Emmanuel de Dieu Nkou, Guy Martial Nkiet

We investigate nonparametric estimation of sliced inverse regression (SIR) via the $k$-nearest neighbors approach with a kernel. An estimator of the covariance matrix of the conditional expectation of the explanatory random vector given the response is then introduced, thereby allowing to estimate the effective dimension reduction (EDR) space. Consistency of

Faissal El Bouanani, Elmehdi Illi, Marwa Qaraqe, Osamah Badarneh

The study examines the secrecy outage probability (SOP) and intercept probability (IP) of a reflecting intelligent surface (RIS)-enabled THz wireless network experiencing $\alpha-\mu$ fading with pointing errors. Specifically, the base station (BS) sends information to a legitimate user $\ell$ via the RIS while an eavesdropper $e$ tries to overhear the conve

Kidist Amde Mekonnen, Yosef Worku Alemneh, Maarten de Rijke

Neural retrieval methods using transformer-based pre-trained language models have advanced multilingual and cross-lingual retrieval. However, their effectiveness for low-resource, morphologically rich languages such as Amharic remains underexplored due to data scarcity and suboptimal tokenization. We address this gap by introducing Amharic-specific dense ret

Lin Tian, Marian-Andrei Rizoiu

Understanding true influence in social media requires distinguishing correlation from causation--particularly when analyzing misinformation spread. While existing approaches focus on exposure metrics and network structures, they often fail to capture the causal mechanisms by which external temporal signals trigger engagement. We introduce a novel joint treat

Mohammad Mahdi Moradi, Sudhir Mudur

Knowledge-Based Visual Question Answering (KB-VQA) methods focus on tasks that demand reasoning with information extending beyond the explicit content depicted in the image. Early methods relied on explicit knowledge bases to provide this auxiliary information. Recent approaches leverage Large Language Models (LLMs) as implicit knowledge sources. While KB-VQ

Camilo Chacón Sartori

With the rise of generative AI (GenAI), Large Language Models are increasingly employed for code generation, becoming active co-authors alongside human programmers. Focusing specifically on this application domain, this paper articulates distinct ``Architectures of Error'' to ground an epistemic distinction between human and machine code generation. Examined

Chenrui Ma, Xi Xiao, Tianyang Wang, Yanning Shen

Current text-driven image editing methods typically follow one of two directions: relying on large-scale, high-quality editing pair datasets to improve editing precision and diversity, or exploring alternative dataset-free techniques. However, constructing large-scale editing datasets requires carefully designed pipelines, is time-consuming, and often result

Sayyed Farid Ahamed, Sandip Roy, Soumya Banerjee, Marc Vucovich

Federated Learning (FL) is a collaborative learning framework designed to protect client data, yet it remains highly vulnerable to Intellectual Property (IP) threats. Model extraction (ME) attacks pose a significant risk to Machine Learning as a Service (MLaaS) platforms, enabling attackers to replicate confidential models by querying black-box (without inte

Hannah Friedman, Bernd Sturmfels, Maximilian Wiesmann

We study statistical models that are parametrized by squares of linear forms. All critical points of the likelihood function are real and positive. There is one critical point in each region of the projective hyperplane arrangement defined by the linear forms. We examine the ideal and singular locus of the model, and we give a determinantal presentation for

Filippo Bigi, Sanggyu Chong, Agustinus Kristiadi, Michele Ceriotti

Molecular dynamics (MD) provides insights into atomic-scale processes by integrating over time the equations that describe the motion of atoms under the action of interatomic forces. Machine learning models have substantially accelerated MD by providing inexpensive predictions of the forces, but they remain constrained to minuscule time integration steps, wh

Gerasimos Gerogiannis, Stijn Eyerman, Evangelos Georganas, Wim Heirman

To alleviate the memory bandwidth bottleneck in Large Language Model (LLM) inference workloads, weight matrices are stored in memory in quantized and sparsified formats. Hence, before tiles of these matrices can be processed by in-core generalized matrix multiplication (GeMM) hardware engines, they need to be dequantized and de-sparsified. This is currently

Shaojie Wang, Sirui Ding, Na Zou

Recent advances in large language models (LLMs) have revolutionized natural language processing, yet evaluating their intrinsic linguistic understanding remains challenging. Moving beyond specialized evaluation tasks, we propose an information-theoretic framework grounded in mutual information (MI) to achieve this. We formalize the understanding as MI betwee

Tânia S. C. Ferreira, Juan J. Alonso, Catherine Gorlé

Contrails have recently gained widespread attention due to their large and uncertain estimates of effective radiative forcing, i.e., warming effect on the planet, comparable to those of carbon dioxide. To study this aircraft-induced cloud formation in the context of current conventional fuels and future alternative fuels, we have developed a numerical framew

Yongmin Yoo, Qiongkai Xu, Longbing Cao

Patent similarity evaluation plays a critical role in intellectual property analysis. However, existing methods often overlook the intricate structure of patent documents, which integrate technical specifications, legal boundaries, and application contexts. We introduce PatentMind, a novel framework for patent similarity assessment based on a Multi-Aspect Re

Jing-Ya Li, Hugo M. Verhelst, Henk den Besten, Matthias Möller

This paper presents spline-based coupling methods for partitioned multiphysics simulations, specifically designed for isogeometric analysis (IGA) based solvers. Traditional vertex-based coupling approaches face significant challenges when applied to IGA solvers, including geometric accuracy issues, interpolation errors, and substantial communication overhead

Yongmin Yoo, Qiongkai Xu, Longbing Cao

High-stakes texts such as patent claims, medical records, and technical reports are structurally complex and demand a high degree of reliability and precision. While large language models (LLMs) have recently been applied to automate their generation in high-stakes domains, reliably evaluating such outputs remains a major challenge. Conventional natural lang

Hideto Iwata

H.L.Montgomery proved a relation for error terms in asymptotic formulas for the Euler totient function. J.Kaczorowski defined the associated Euler totient function which generalizes and obtained an asymptotic formula for it. In this paper, we prove a relation on error terms similar to H.L.Montgomery's result for a certain special case of the associated Euler

Chun-Sheng Hsueh

We build handle decompositions of n-manifolds that encode given open book decompositions and describe handle slides that reveal new open book decompositions on the same underlying manifold, for $n \geq 3$. This recovers known stabilization operations for open books. As an application, we show that any open book with trivial monodromy can be stabilized to an

Andrew Ralph Warwick, Pui-Wai Ma, Sergei Lvovich Dudarev

We present a comprehensive ab initio dataset of formation energies and elastic properties of intrinsic point defects across all the transition and rare earth hexagonal close packed (hcp) metals, as well as metalloid elements with hcp crystal structure. Point defect properties appear weakly correlated with the c/a ratio of the hcp lattice. Instead, it is the

Pakanun Dokyeesun, Csilla Bujtás

For a connected graph $G$ and $X\subseteq V(G)$, we say that two vertices $u$, $v$ are $X$-visible if there is a shortest $u,v$-path $P$ with $V(P)\cap X \subseteq \{u,v\}$. If every two vertices from $X$ are $X$-visible, then $X$ is a mutual-visibility set in $G$. The largest cardinality of such a set in $G$ is the mutual-visibility number $\mu(G)$. When th

Libo Wang

To address the gaps between the static pre-set "thinking-planning-action" of humanoid robots in unfamiliar scenarios and the highly programmed "call tool-return result" due to the lack of autonomous coding capabilities, this work designs a dynamic architecture connecting continuous thought machines (CTM) and model context protocol (MCP). It proposes a theore

Adeela Bashir, Zia Ush Shamszaman, Zhao Song, The Anh Han

In the evolving digital landscape, it is crucial to study the dynamics of cyberattacks and defences. This study uses an Evolutionary Game Theory (EGT) framework to investigate the evolutionary dynamics of attacks and defences in cyberspace. We develop a two-population asymmetric game between attacker and defender to capture the essential factors of costs, po

Kevin Li, Marinka Zitnik

Offline goal-conditioned reinforcement learning methods have shown promise for reach-avoid tasks, where an agent must reach a target state while avoiding undesirable regions of the state space. Existing approaches typically encode avoid-region information into an augmented state space and cost function, which prevents flexible, dynamic specification of novel

Fan Li, Jiaqi Tong, Xi Fang, Chao Cheng

In cluster-randomized trials, generalized linear mixed models and generalized estimating equations have conventionally been the default analytic methods for estimating the average treatment effect as routine practice. However, recent studies have demonstrated that their treatment effect coefficient estimators may correspond to ambiguous estimands when the mo

Dora Zhao, Diyi Yang, Michael S. Bernstein

Large language models are designed to encode general purpose knowledge about the world from Internet data. Yet, a wealth of information falls outside this scope -- ranging from personal preferences to organizational policies, from community-specific advice to up-to-date news -- that users want models to access but remains unavailable. In this paper, we propo

Emmanuel Akinrintoyo, Nadine Abdelhalim, Nicole Salomons

Whisper fails to correctly transcribe dementia speech because persons with dementia (PwDs) often exhibit irregular speech patterns and disfluencies such as pauses, repetitions, and fragmented sentences. It was trained on standard speech and may have had little or no exposure to dementia-affected speech. However, correct transcription is vital for dementia sp

Charles Godfrey, Ping Nie, Natalia Ostapuk, David Ken

Large language models (LLMs) obtain state of the art zero shot relevance ranking performance on a variety of information retrieval tasks. The two most common prompts to elicit LLM relevance judgments are pointwise scoring (a.k.a. relevance generation), where the LLM sees a single query-document pair and outputs a single relevance score, and listwise ranking

Zach Studdiford, Timothy T. Rogers, Siddharth Suresh, Kushin Mukherjee

Current evaluations of Large Language Model (LLM) steering techniques focus on task-specific performance, overlooking how well steered representations align with human cognition. Using a well-established triadic similarity judgment task, we assessed steered LLMs on their ability to flexibly judge similarity between concepts based on size or kind. We found th

Yifei He, Martin Kruczenski

The Gauge Theory Bootstrap [arXiv:2309.12402, arXiv:2403.10772] computes the strongly coupled pion dynamics by considering the most general scattering matrix, form factors and spectral densities and matching them with perturbative QCD at high energy and with weakly coupled pions at low energy. In this work, we show that further constraints on the spectral de

Onder Akcaalan, Melike Gumus Akcaalan

Interferometric Fiber Optic Gyroscopes (IFOGs) are widely used in precision navigation systems due to their high sensitivity, robustness, and solid-state nature. To ensure linear response and accurate angular velocity measurement, a fixed $\pi/2$ phase bias is typically introduced between the clockwise (CW) and counter-clockwise (CCW) beams using active modu

W. Jacob Wagner

This document serves as supplementary material for two International Society for Terrain-Vehicle Systems conference publications regarding in situ soil property estimation by Wagner et al. in 2023 and 2025. It covers the derivation of the fundamental equation of earthmoving for a flat blade moving through sloped soil and provides some information regarding t

Yue Wang, Kelvin Shum, Yuyang Song, Tian Chen

Transforming planar mesoscale devices into precise 3-D architectures is vital for next-generation flexible electronics, implants, and adaptive optics, yet wafer-based manufacturing to free-standing 3-D structures remain elusive. We fabricate polyimide architected 2-D precursors whose bistable unit cells deploy into stable 3-D mesoscale structures. Target Gau

Manuela González-González, Soufiane Belharbi, Muhammad Osama Zeeshan, Masoumeh Sharafi

Ambivalence and hesitancy (A/H), closely related constructs, are the primary reasons why individuals delay, avoid, or abandon health behaviour changes. They are subtle and conflicting emotions that sets a person in a state between positive and negative orientations, or between acceptance and refusal to do something. They manifest as a discord in affect betwe