October 2025 arXiv papers — page 37

Xingyu Liu, Kun Ming Goh

Convolutional Neural Networks (CNNs) has revolutionized computer vision, but training very deep networks has been challenging due to the vanishing gradient problem. This paper explores Residual Networks (ResNet), introduced by He et al. (2015), which overcomes this limitation by using skip connections. ResNet enables the training of networks with hundreds of

Xinqi Li, Yiqun Liu, Shan Jiang, Enrong Zheng

We introduce GraphNet, a dataset of 2.7K real-world deep learning computational graphs with rich metadata, spanning six major task categories across multiple deep learning frameworks. To evaluate tensor compiler performance on these samples, we propose the benchmark metric Speedup Score S(t), which jointly considers runtime speedup and execution correctness

Yufan Liu, Wanqian Zhang, Huashan Chen, Lin Wang

Despite rapid advancements in text-to-image (T2I) models, their safety mechanisms are vulnerable to adversarial prompts, which maliciously generate unsafe images. Current red-teaming methods for proactively assessing such vulnerabilities usually require white-box access to T2I models, and rely on inefficient per-prompt optimization, as well as inevitably gen

Lizhen Qin, Yang Su

A topological space is called self-covering if it is a nontrivial cover of itself. We prove that, under mild assumptions, a closed self-covering manifold with an abelian fundamental group fibers over a torus in various senses. As a corollary, if its dimension is above $5$ and its fundamental group is free abelian, then it is a fiber bundle over a circle. We

Peng Cai, Reza Ryan, Nickson M. Karie

System logs are a cornerstone of cybersecurity, supporting proactive breach prevention and post-incident investigations. However, analyzing vast amounts of diverse log data remains significantly challenging, as high costs, lack of in-house expertise, and time constraints make even basic analysis difficult for many organizations. This study introduces LLMLogA

Ahmet Akkaya Melih, Yamuna Singh, Kunal L. Agarwal, Priya Mukherjee

The rapid advancements in large foundation models and multi-agent systems offer unprecedented capabilities, yet current Human-in-the-Loop (HiTL) paradigms inadequately integrate human expertise, often leading to cognitive overload and decision-making bottlenecks in complex, high-stakes environments. We propose the "Human-Machine Social Hybrid Intelligence" (

Andrew Gerstenslager, Bekarys Dukenbaev, Ali A. Minai

Boundary Vector Cells (BVCs) are a class of neurons in the brains of vertebrates that encode environmental boundaries at specific distances and allocentric directions, playing a central role in forming place fields in the hippocampus. Most computational BVC models are restricted to two-dimensional (2D) environments, making them prone to spatial ambiguities i

Tingyue Pan, Mingyue Cheng, Shilong Zhang, Zhiding Liu

Cross-domain time series forecasting is a valuable task in various web applications. Despite its rapid advancement, achieving effective generalization across heterogeneous time series data remains a significant challenge. Existing methods have made progress by extending single-domain models, yet often fall short when facing domain-specific trend shifts and i

Yingzhao Jian, Zhongan Wang, Yi Yang, Hehe Fan

Humanoid agents often struggle to handle flexible and diverse interactions in open environments. A common solution is to collect massive datasets to train a highly capable model, but this approach can be prohibitively expensive. In this paper, we explore an alternative solution: empowering off-the-shelf Vision-Language Models (VLMs, such as GPT-4) to control

Zibo Liu, Zhe Jiang, Zelin Xu, Tingsong Xiao

Spatio-Temporal Multivariate time series Forecast (STMF) uses the time series of $n$ spatially distributed variables in a period of recent past to forecast their values in a period of near future. It has important applications in spatio-temporal sensing forecast such as road traffic prediction and air pollution prediction. Recent papers have addressed a prac

Jiaqi Luo, Shixin Xu, Zhouwang Yang

The accuracy of Physics-Informed Neural Networks (PINNs) critically depends on the placement of collocation points, as the PDE loss is approximated through sampling over the solution domain. Global sampling ensures stability by covering the entire domain but requires many samples and is computationally expensive, whereas local sampling improves efficiency by

Tianqi Guo, Liping Chen, Ciyuan Peng, Jingjing Zhou

Understanding the evolution of brain functional networks over time is of great significance for the analysis of cognitive mechanisms and the diagnosis of neurological diseases. Existing methods often have difficulty in capturing the temporal evolution characteristics of connections between specific functional communities. To this end, this paper proposes a n

Yuchi Ishikawa, Toranosuke Manabe, Tatsuya Komatsu, Yoshimitsu Aoki

Audio-visual captioning aims to generate holistic scene descriptions by jointly modeling sound and vision. While recent methods have improved performance through sophisticated modality fusion, it remains unclear to what extent the two modalities are complementary in current audio-visual captioning models and how robust these models are when one modality is d

Saujas Vaduguru, Yilun Hua, Yoav Artzi, Daniel Fried

Humans leverage shared conversational context to become increasingly successful and efficient at communicating over time. One manifestation of this is the formation of ad hoc linguistic conventions, which allow people to coordinate on short, less costly utterances that are understood using shared conversational context. We present a method to train large mul

Xiao-Ping Chen, Chun-Lei Tang

Based on some new vector inequalities established by Figalli and Zhang [\emph{Duke Math. J.} \textbf{171} (2022), 2407--2459], we study the stability of the scale invariant and the scale non-invariant $L^p$-Caffarelli-Kohn-Nirenberg inequalities, which fills the recent work of Do \emph{et al.} [$L^p$-Caffarelli-Kohn-Nirenberg inequalities and their stabiliti

Haiduo Huang, Jiangcheng Song, Yadong Zhang, Pengju Ren

Knowledge distillation (KD) is a standard route to compress Large Language Models (LLMs) into compact students, yet most pipelines uniformly apply token-wise loss regardless of teacher confidence. This indiscriminate supervision amplifies noisy, high-entropy signals and is especially harmful under large teacher-student capacity gaps. We introduce SelecTKD, a

Hao An, Yang Xu

Mitigating hallucinations in Large Language Models (LLMs) is critical for their reliable deployment. Existing methods typically fine-tune LLMs to abstain from answering questions beyond their knowledge scope. However, these methods often rely on coarse-grained signals to guide LLMs to abstain, such as overall confidence or uncertainty scores on multiple samp

Xing Xing, Wei Wang, Lipeng Ma, Weidong Yang

Recent progress in large language models (LLMs) has advanced automatic code generation, yet most approaches rely on direct, single-step translation from problem descriptions to code, disregarding structured software engineering practices. We introduce a lifecycle-aware framework that systematically incorporates intermediate artifacts such as requirements ana

Brian Skoglind, Travis Roberts, Sourabh Karmakar, Cameron Turner

Modern consumer products are full of interconnected electrical and electronic modules to fulfill direct and indirect needs. In an automated assembly line still, most of these interconnections are required to be done manually due to the large variety of connector types, connector positions, and the soft, flexible nature of their structures. The manual connect

Aaron Winn, Eleni Katifori

Peristalsis is the driving mechanism behind a broad array of biological and engineered flows. In peristaltic pumping, a wave-like contraction of the tube wall produces local changes in volume which induce flow. Net flow arises due to geometric nonlinearities in the momentum equation, which must be properly captured to compute the flow accurately. While most

Mehrnoosh Tahani, Laura Fissel, Enrique Lopez Rodriguez, Kate Pattle

We propose to use the unprecedented polarization sensitivity of PRIMA's PRIMAger Polarization Imager and its high resolution in Band 1 (92 $\mu$m) to map magnetic fields across two contrasting molecular cloud environments: the well-studied Perseus cloud and the isolated Musca filament. This comparative study will leverage the existing VLA radio observations

Yizhu Jiao, Sha Li, Sizhe Zhou, Heng Ji

The task of information extraction (IE) is to extract structured knowledge from text. However, it is often not straightforward to utilize IE output due to the mismatch between the IE ontology and the downstream application needs. We propose a new formulation of IE TEXT2DB that emphasizes the integration of IE output and the target database (or knowledge base

İbrahim Oğuz Çetinkaya, İ. Esra Büyüktahtakın, Parshin Shojaee, Chandan K. Reddy

Our study contributes to the scheduling and combinatorial optimization literature with new heuristics discovered by leveraging the power of Large Language Models (LLMs). We focus on the single-machine total tardiness (SMTT) problem, which aims to minimize total tardiness by sequencing n jobs on a single processor without preemption, given processing times an

Byeonghu Na, Mina Kang, Jiseok Kwak, Minsang Park

Text-to-image models have recently made significant advances in generating realistic and semantically coherent images, driven by advanced diffusion models and large-scale web-crawled datasets. However, these datasets often contain inappropriate or biased content, raising concerns about the generation of harmful outputs when provided with malicious text promp

Hiroki Nakano, Jo Takezawa, Fabrice Matulic, Chi-Lan Yang

As AI writing support becomes ubiquitous, how disclosing its use affects reader perception remains a critical, underexplored question. We conducted a study with 261 participants to examine how revealing varying levels of AI involvement shifts author impressions across six distinct communicative acts. Our analysis of 990 responses shows that disclosure genera

Mirali Purohit, Bimal Gajera, Vatsal Malaviya, Irish Mehta

Foundation models have enabled rapid progress across many specialized domains by leveraging large-scale pre-training on unlabeled data, demonstrating strong generalization to a variety of downstream tasks. While such models have gained significant attention in fields like Earth Observation, their application to Mars science remains limited. A key enabler of

Yuan Jin, Antonio Pepe, Gian Marco Melito, Yuxuan Chen

The automated analysis of the aortic vessel tree (AVT) from computed tomography angiography (CTA) holds immense clinical potential, but its development has been impeded by a lack of shared, high-quality data. We launched the SEG.A. challenge to catalyze progress in this field by introducing a large, publicly available, multi-institutional dataset for AVT seg

Chi-Ming Chang, Ying-Hsuan Lin

We study the cohomology of a chiral supercharge $Q$ in the $\mathcal{N}=4$ super-Yang-Mills (SYM) theory at tree level. The cohomology classes correspond one-to-one to the $\frac1{16}$ Bogomol'nyi-Prasad-Sommerfield (BPS) states at one-loop. We argue that monotone classes on the Coulomb branch respect the S-duality between the theories with $\mathrm{SO}(2N+1

Fa Peng Huang, Chikako Idegawa, Aidi Yang

We investigate the formation of primordial black holes (PBHs) induced by a first-order electroweak phase transition in a realistic renormalizable framework, the complex singlet extension of the Standard Model. We perform a quantitative analysis of the PBH abundance and identify parameter regions consistent with current microlensing constraints. Furthermore,

Ming-Hui Yu, Xian-Hui Ge

We investigate the information paradox in the four-dimensional Kerr-Newman black hole by employing the recently proposed island paradigm. We first consider the quantum field in the four-dimensional Kerr-Newman spacetime. By employing the near-horizon limit, we demonstrate that the field can be effectively described by a reduced two-dimensional field theory.

Srinivas Balaji Bollepalli

We announce two breakthrough results concerning important questions in the Theory of Computational Complexity. In this expository paper, a systematic and comprehensive geometric characterization of the Subset Sum Problem is presented. We show the existence of a universal geometric structure, comprised of a family of non-decreasing paths in the Cartesian plan

Yiran Wang

The integrated Sachs-Wolfe (ISW) effect describes how photons are gravitationally redshifted, producing anisotropies in the Cosmic Microwave Background. We study the inverse problem and show that primordial gravitational perturbations, in particular their polarizations in the transversally traceless (TT) gauge can be identified from the local observation of

Shane Dirksen, Radha Kumaran, You-Jin Kim, Yilin Wang

We study attention in mobile Augmented Reality (AR) using object recall as a proxy outcome. We observe that the ability to recall an object (physical or virtual) that was encountered in a mobile AR experience depends on many possible impact factors and attributes, with some objects being readily recalled while others are not, and some people recalling object

Mengzhou Sun, Sendong Zhao, Jianyu Chen, Haochun Wang

Evidence-based medicine (EBM) holds a crucial role in clinical application. Given suitable medical articles, doctors effectively reduce the incidence of misdiagnoses. Researchers find it efficient to use large language models (LLMs) techniques like RAG for EBM tasks. However, the EBM maintains stringent requirements for evidence, and RAG applications in EBM

Zehao Lin

Global climate warming and air pollution pose severe threats to economic development and public safety, presenting significant challenges to sustainable development worldwide. Corporations, as key players in resource utilization and emissions, have drawn increasing attention from policymakers, researchers, and the public regarding their environmental strateg

Trí Minh Lê, Khai-Hoan Nguyen-Dang

Intrinsic volumes are fundamental geometric invariants generalizing volume, surface area, and mean width for convex bodies. We establish a unified Laplace-Grassmannian representation for intrinsic and dual volumes of convex polynomial sublevel sets. More precisely, let $f$ be a convex $d$-homogeneous polynomial of even degree $d \ge 2$ which is positive exce

Heethanjan Kanagalingam, Thenukan Pathmanathan, Mokeeshan Vathanakumar, Tharmakulasingam Mukunthan

Diabetic retinopathy (DR) is a leading cause of vision loss worldwide, yet early and accurate detection can significantly improve treatment outcomes. While numerous Deep learning (DL) models have been developed to predict DR from fundus images, many face challenges in maintaining robustness due to distributional variations caused by differences in acquisitio

Siavash M. Alamouti, Fay Arjomandi, Michel Burger, Bashar Altakrouri

Industrial automation in the energy sector requires AI systems that can operate autonomously regardless of network availability, a requirement that cloud-centric architectures cannot meet. This paper evaluates the application of Device-First Continuum AI (DFC-AI) to critical energy sector operations. DFC-AI, a specialized architecture within the Hybrid Edge

Kevin Buck, Woojeong Kim

We propose an adaptive sampling method for the training of Physics Informed Neural Networks (PINNs) which allows for sampling based on an arbitrary problem-specific heuristic which may depend on the network and its gradients. In particular we focus our analysis on the Allen-Cahn equations, attempting to accurately resolve the characteristic interfacial regio

Mengzhou Sun, Sendong Zhao, Jianyu Chen, Bin Qin

Evidence-based medicine (EBM) research has always been of paramount importance. It is important to find appropriate medical theoretical support for the needs from physicians or patients to reduce the occurrence of medical accidents. This process is often carried out by human querying relevant literature databases, which lacks objectivity and efficiency. Ther

Stanley Wu, Mohamad H. Danesh, Simon Li, Hanna Yurchyk

Recent advancements in legged robot locomotion have facilitated traversal over increasingly complex terrains. Despite this progress, many existing approaches rely on end-to-end deep reinforcement learning (DRL), which poses limitations in terms of safety and interpretability, especially when generalizing to novel terrains. To overcome these challenges, we in

Y. T. Zhu, Shibei Xue, Fangfang Ju, Haidong Yuan

Surface acoustic waves (SAWs), as Rayleigh waves generated by elastic media, have been used in gyroscopes for over 40 years due to their unique propagation characteristics. However, their working principle, based on Coriolis effects, has become increasingly ineffective for addressing modern sensing challenges in complex scenarios. Fortunately, recent advance

Mengzhou Sun, Sendong Zhao, Jianyu Chen, Haochun Wang

Retrieval-augmented Generation (RAG) has demonstrated potential in enhancing medical question-answering systems through the integration of large language models (LLMs) with external medical literature. LLMs can retrieve relevant medical articles to generate more professional responses efficiently. However, current RAG applications still face problems. They g

Anthony Carreon, Jagmohan Singh, Shivank Sharma, Shuzhi Zhang

High-speed chemically active flows present significant computational challenges due to their disparate space and time scales, where stiff chemistry often dominates simulation time. While modern supercomputing scientific codes achieve exascale performance by leveraging graphics processing units (GPUs), existing GPU-based compressible combustion solvers face c

Yuxiao Wen, Yanjun Han, Zhengyuan Zhou

Combinatorial bandits extend the classical bandit framework to settings where the learner selects multiple arms in each round, motivated by applications such as online recommendation and assortment optimization. While extensions of upper confidence bound (UCB) algorithms arise naturally in this context, adapting arm elimination methods has proved more challe

Dor Minzer, Kai Zhe Zheng

We show that for every $k\in\mathbb{N}$ and $\varepsilon>0$, for large enough alphabet $R$, given a $k$-CSP with alphabet size $R$, it is NP-hard to distinguish between the case that there is an assignment satisfying at least $1-\varepsilon$ fraction of the constraints, and the case no assignment satisfies more than $1/R^{k-1-\varepsilon}$ of the constraints

Maruf Ahmed Mridul, Oshani Seneviratne

The transformation of unstructured legal contracts into standardized, machine-readable formats is essential for automating financial workflows. The Common Domain Model (CDM) provides a standardized framework for this purpose, but converting complex legal documents like Credit Support Annexes (CSAs) into CDM representations remains a significant challenge. In

Shangde Gao, Zelin Xu, Zhe Jiang

Shifts in individual movement patterns following disruptive events can reveal changing demands for community resources. However, predicting such shifts before disruptive events remains challenging for several reasons. First, measures are lacking for individuals' heterogeneous social infrastructure resilience (SIR), which directly influences their movement pa

Phuc Nguyen Xuan, Thanh Nguyen Canh, Huu-Hung Nguyen, Nak Young Chong

This survey comprehensively reviews the evolving field of multi-robot collaborative Simultaneous Localization and Mapping (SLAM) using 3D Gaussian Splatting (3DGS). As an explicit scene representation, 3DGS has enabled unprecedented real-time, high-fidelity rendering, ideal for robotics. However, its use in multi-robot systems introduces significant challeng

Emre Parmaksiz, Ramon van Handel

A recent development in random matrix theory, the intrinsic freeness principle, establishes that the spectrum of very general random matrices behaves as that of an associated free operator. This reduces the study of such random matrices to the deterministic problem of computing spectral statistics of the free operator. In the self-adjoint case, the spectral

Hong Wang, Jiang Yixuan, Jie Wang, Xinyi Li

Operator eigenvalue problems play a critical role in various scientific fields and engineering applications, yet numerical methods are hindered by the curse of dimensionality. Recent deep learning methods provide an efficient approach to address this challenge by iteratively updating neural networks. These methods' performance relies heavily on the spectral

Adam Nordenhög, Akash Sharma

Score-based generative models based on stochastic differential equations (SDEs) achieve impressive performance in sampling from unknown distributions, but often fail to satisfy underlying constraints. We propose a constrained generative model using kinetic (underdamped) Langevin dynamics with specular reflection of velocity on the boundary defining constrain

Isabella Danhoni, Yumu Yang, Mauricio Hippert, Jacquelyn Noronha-Hostler

The symmetry energy expansion was developed to connect isospin symmetric matter probed in nuclear experiments to asymmetric matter found in neutron stars. Using the isospin asymmetry derived from the Gell-Mann-Nishijima formula, we derive the symmetry energy expansion for quark matter that has unique properties compared to hadronic matter. To test our method

Seonggyun Lee, Sungjun Lim, Seojin Park, Soeun Cheon

The field of preference optimization has made outstanding contributions to the alignment of language models with human preferences. Despite these advancements, recent methods still rely heavily on substantial paired (labeled) feedback data, leading to substantial resource expenditures. To address these challenges, we study the problem of Semi-Supervised Pref

Zhonghao Xia, Zhilong Yang, Yali Yang, Kaile Ren

The intrinsic coupling among electrical conductivity ($\sigma$), Seebeck coefficient ($S$), and lattice thermal conductivity ($\kappa_{\mathrm{L}}$) imposes a fundamental limit on the dimensionless figure of merit $ZT$ in thermoelectric (TE) materials. Increasing band degeneracy can effectively balance $\sigma$ and $S$, enabling a high power factor (PF, $S^{

Yihui Tong, Wenjie Liu, Zhichang Guo, Wenjuan Yao

This paper investigates a novel class of regularizations of the Perona-Malik equation with variable exponents, of forward-backward parabolic type, which possess a variational structure and have potential applications in image processing. The existence of Young measure solutions to the Neumann initial-boundary value problem for the proposed equation is establ

Jiaqi Yan, Ruilong Ren, Jingren Liu, Shuning Xu

Egocentric AI assistants in real-world settings must process multi-modal inputs (video, audio, text), respond in real time, and retain evolving long-term memory. However, existing benchmarks typically evaluate these abilities in isolation, lack realistic streaming scenarios, or support only short-term tasks. We introduce \textbf{TeleEgo}, a long-duration, st

Ying Chen, Bin Luo, Jian Huang

Super-Eddington accretion is a crucial phase in the growth of supermassive black holes. However, identifying super-Eddington accreting quasars observationally is challenging due to uncertain black-hole mass estimates and other complications. The Eddington ratio parameter does not represent accurately the accretion rate in the super-Eddington regime. On the o

Kazutoshi Akita, Norimichi Ukita

Cost-and-Quality (CQ) controllability in arbitrary-scale super-resolution is crucial. Existing methods predict Fourier components one by one using a recurrent neural network. However, this approach leads to performance degradation and inefficiency due to independent prediction. This paper proposes predicting multiple components jointly to improve both qualit

Alexandre St-Georges, Gabriel Richard, Maxime Toussaint, Christian Thibaudeau

Accurate quantification in positron emission tomography (PET) is essential for accurate diagnostic results and effective treatment tracking. A major issue encountered in PET imaging is attenuation. Attenuation refers to the diminution of photon detected as they traverse biological tissues before reaching detectors. When such corrections are absent or inadequ

Yohan Abeysinghe, Muhammad Akhtar Munir, Sanoojan Baliah, Ron Sarafian

Air pollution remains a leading global health and environmental risk, particularly in regions vulnerable to episodic air pollution spikes due to wildfires, urban haze and dust storms. Accurate forecasting of particulate matter (PM) concentrations is essential to enable timely public health warnings and interventions, yet existing models often underestimate r

Richard Berk

Using data from the Longyearbyen weather station, quantile gradient boosting ("small AI") is applied to forecast daily temperatures in Svalbard, Norway. Temperatures above 0 degrees Celsius are of special interest because of their impact on ice, snow, and tundra permafrost. To improve forecasting skill for warmer temperatures, the target quantile is 0.60; fo

Taha Ahmad

Background: Patients carrying MEF2C haploinsufficiency develop a recognizable neurodevelopmental syndrome featuring intellectual disability, treatment-resistant seizures, and autism spectrum behaviors. While MEF2C's critical roles in cardiac development and neuronal function are well-established, its specific transcriptional operations within microglia (the

Kevin Michalewicz, Mauricio Barahona, Barbara Bravi

Understanding the relationship between antibody sequence, structure and function is essential for the design of antibody-based therapeutics and research tools. Recently, machine learning (ML) models mostly based on the application of large language models to sequence information have been developed to predict antibody properties. Yet there are open direction

Byeonghu Na, Minsang Park, Gyuwon Sim, Donghyeok Shin

Text-to-image diffusion models rely on text embeddings from a pre-trained text encoder, but these embeddings remain fixed across all diffusion timesteps, limiting their adaptability to the generative process. We propose Diffusion Adaptive Text Embedding (DATE), which dynamically updates text embeddings at each diffusion timestep based on intermediate perturb

Victor Ayala, Adriano Da Silva

In this paper, we study linear control systems with positive bounded orbits. We show that the existence of positive bounded orbits imposes strong algebraic and topological constraints on the state space. In fact, a linear control system has bounded positive orbits if and only if it can be decomposed as the product of the stable and central subgroups of the d

Andraž Jelinčič, Owen Lockwood, Akhil Garlapati, Peter Schillinger

The proliferation of probabilistic AI has prompted proposals for specialized stochastic computers. Despite promising efficiency gains, these proposals have failed to gain traction because they rely on fundamentally limited modeling techniques and exotic, unscalable hardware. In this work, we address these shortcomings by proposing an all-transistor probabili

Zhuo-Hua Chen, Hou-Jian Duan, Ming-Xun Deng, Rui-Qiang Wang

Recent experimental and theoretical studies have revealed the emergence of a linear layer Hall effect (LHE) induced by hidden Berry curvature in \textrm{MnBi}$_{2}$\textrm{Te}$_{4}$ thin films. This phenomenon underscores the layer degree of freedom as a novel mechanism for generating Hall transport in layered materials, providing a new pathway to probe and

Xiaoyang Hu

A signature of human cognitive control is conflict adaptation: improved performance on a high-conflict trial following another high-conflict trial. This phenomenon offers an account for how cognitive control, a scarce resource, is recruited. Using a sequential Stroop task, we find that 12 of 13 vision-language models (VLMs) tested exhibit behavior consistent

Maria C. Borges, Julian Legler, Lucca Di Benedetto

Observability and alerting form the backbone of modern reliability engineering. Alerts help teams catch faults early before they turn into production outages and serve as first clues for troubleshooting. However, designing effective alerts is challenging. They need to strike a fine balance between catching issues early and minimizing false alarms. On top of

Lorenzo Carta, Fernando Spadea, Oshani Seneviratne

We present the first application of federated learning (FL) to the U.S. National Financial Capability Study, introducing an interpretable framework for predicting consumer financial distress across all 50 states and the District of Columbia without centralizing sensitive data. Our cross-silo FL setup treats each state as a distinct data silo, simulating real

Harshavardhana T. Gowda, Daniel C. Comstock, Lee M. Miller

We present a neuromuscular speech interface that translates electromyographic (EMG) signals recorded from orofacial muscles during speech articulation directly into audio. We find that self-supervised speech (S3) representations are strongly linearly related to the electrical power of muscle activity: a simple linear mapping predicts EMG power from S3 repres

Andriy Myronenko, Dong Yang, Baris Turkbey, Mariam Aboian

Vision-language models (VLMs) have shown strong promise for medical image analysis, but most remain opaque, offering predictions without the transparent, stepwise reasoning clinicians rely on. We present a framework that brings chain-of-thought (CoT) reasoning to chest X-ray interpretation. Inspired by reasoning-first training paradigms, our approach is desi

Mu-Lin Li, Sheng Rao, Kai Wang

Much inspired by J. A. Wi\'sniewski's nef-value function method, we prove that in a smooth projective family over the unit disk, if the adjoint bundle of the canonical line bundle with a relatively semiample line bundle is nef on one fiber, then it remains nef on all fibers. We further extend this result to the semiampleness of the adjoint canonical line bun

Scott Emmons, Roland S. Zimmermann, David K. Elson, Rohin Shah

While Chain-of-Thought (CoT) monitoring offers a unique opportunity for AI safety, this opportunity could be lost through shifts in training practices or model architecture. To help preserve monitorability, we propose a pragmatic way to measure two components of it: legibility (whether the reasoning can be followed by a human) and coverage (whether the CoT c

Ali Aouad, Aymane El Gadarri, Vivek F. Farias

Traditional LLM alignment methods are vulnerable to heterogeneity in human preferences. Fitting a na\"ive probabilistic model to pairwise comparison data (say over prompt-completion pairs) yields an inconsistent estimate of the population-average utility -a canonical measure of social welfare. We propose a new method, dubbed the sign estimator, that provides

Filipi N. Silva, Sadamori Kojaku, Alessandro Flammini, Filippo Radicchi

Most networks encountered in nature, society, and technology have weighted edges, representing the strength of the interaction/association between their vertices. Randomizing the structure of a network is a classic procedure used to estimate the statistical significance of properties of the network, such as transitivity, centrality and community structure. R

Hiroki Ishikawa, Kyosuke Ishibashi, Ko Yamamoto

This paper presents a soft robot finger capable of adaptive-twist deformation to grasp objects by wrapping them. For a soft hand to grasp and pick-up one object from densely contained multiple objects, a soft finger requires the adaptive-twist deformation function in both in-plane and out-of-plane directions. The function allows the finger to be inserted dee

Hangil Jang, Sanghoon Lim

Jet flavor tagging, the identification of jets originating from $c$-quarks, $b$-quarks, and other quarks (light quarks and gluons), is a crucial task in high-energy heavy-ion physics, as it enables the investigation of flavor-dependent responses within the hot and dense nuclear medium produced in heavy-ion collisions. Recently, several methods based on deep

Haining Pan, Jacob R. Taylor, Jay D. Sau, Sankar Das Sarma

Recent work on Majorana-bound states in semiconductor-superconductor hybrid structures has elucidated the key role of unintentional (and unknown) disorder (producing low-energy Andreev-bound states) in the system, which is detrimental to the emergence of Majorana-carrying topological superconductivity artificially engineered through the combination of superc

Peiyang Xu, Minzhou Pan, Zhaorun Chen, Shuang Yang

With the rapid proliferation of digital media, the need for efficient and transparent safeguards against unsafe content is more critical than ever. Traditional image guardrail models, constrained by predefined categories, often misclassify content due to their pure feature-based learning without semantic reasoning. Moreover, these models struggle to adapt to

Qiumeng Li, Chunhou Ji, Xinyue Liu

Understanding and replicating human mobility requires not only spatial-temporal accuracy but also an awareness of the cognitive hierarchy underlying real-world travel decisions. Traditional agent-based or deep learning models can reproduce statistical patterns of movement but fail to capture the semantic coherence and causal logic of human behavior. Large la

Geoffrey Janssens, Doryan Temmerman, François Thilmany

Let $\Gamma$ be a Zariski-dense subgroup of a reductive group $\mathbf{G}$ defined over a field $F$. Given a finite collection of finite subgroups $H_i$ ($i \in I$) of $\mathbf{G}(F)$ avoiding the center, we establish a criterion to ensure that the set of elements of $\Gamma$ that form a free product with every $H_i$ (the so-called simultaneous ping-pong par

Alejandro Escontrela, Shrinu Kushagra, Sjoerd van Steenkiste, Yulia Rubanova

Amazing progress has been made in controllable generative modeling, especially over the last few years. However, some challenges remain. One of them is precise and iterative object editing. In many of the current methods, trying to edit the generated image (for example, changing the color of a particular object in the scene or changing the background while k

Haoyang He, Eduardo Martínez-Pedroza

We show that there is a collection of subgroups of the mapping class group of a surface such that the associated coset intersection complex is quasi-isometric and homotopy equivalent to the curve complex. Moreover, we prove that these two complexes are combinatorially equivalent in the sense that one can be obtained from the other via taking a nerve. As an a

Pejman Kheradmand, Behnam Moradkhani, Raghavasimhan Sankaranarayanan, Kent K. Yamamoto

Tendon-actuated concentric tube mechanisms combine the advantages of tendon-driven continuum robots and concentric tube robots while addressing their respective limitations. They overcome the restricted degrees of freedom often seen in tendon-driven designs, and mitigate issues such as snapping instability associated with concentric tube robots. However, a c

Krishna Kumar Neelakanta Pillai Santha Kumari Amma

Dynamic pricing in retail requires policies that adapt to shifting demand while coordinating decisions across related products. We present a systematic empirical study of multi-agent reinforcement learning for retail price optimization, comparing a strong MAPPO baseline with a graph-attention-augmented variant (MAPPO+GAT) that leverages learned interactions

Yaxing Ma, Emmanuel Trélat, Lijuan Wang, Huaiqiang Yu

In this paper, we give sufficient conditions under which linear abstract control systems for which the semigroup is analytic are stabilizable with a bounded feedback. We obtain various characterizations of that property, which extend some earlier works. We illustrate our findings with several examples.

Jie Shi

We introduce a new class of asymptotic contractions that employs two quasi-metrics defined directly in terms of the underlying mapping. The contraction condition compares these two quantities via a sequence of bounding functions that converge locally uniformly to a Boyd-Wong function. This framework relaxes the hypotheses of Kirk's asymptotic fixed point the

Qingmin Liu, Yuyang Miao

This paper analyzes a dynamic interaction between a fully rational, privately informed sender and a boundedly rational, uninformed receiver with memory constraints. The sender controls the flow of information, while the receiver designs a decision-making protocol that uses a finite state space to learn and to provide incentives. We characterize optimal proto

Kyomin Hwang, Hyeonjin Kim, Seungyeon Kim, Sunghyun Wee

There have been a couple of studies showing that attempting to erase multilingual knowledge using only English data is insufficient for multilingual LLMs. However, their analyses remain highly performance-oriented. In this paper, we switch the point of view to evaluation, and address an additional blind spot which reveals itself when the multilingual LLM is

Qianfeng Wen, Zhenwei Tang, Ashton Anderson

Chess provides an ideal testbed for evaluating the reasoning, modeling, and abstraction capabilities of large language models (LLMs), as it has well-defined structure and objective ground truth while admitting a wide spectrum of skill levels. However, existing evaluations of LLM ability in chess are ad hoc and narrow in scope, making it difficult to accurate

Zihan Liu, Parisa Rabbani, Veda Duddu, Kyle Fan

LLM-powered multimodal systems are increasingly used to interpret human behavior, yet how researchers apply the models' 'social competence' remains poorly understood. This paper presents a systematic literature review of 176 publications across different application domains (e.g., healthcare, education, and entertainment). Using a four-dimensional coding fra

Madhav Vadlamani, Dyutimoy Chakraborty, Jianwei Jia, Halid Mulaosmanovic

Ferroelectric-based capacitive crossbar arrays have been proposed for energy-efficient in-memory computing in the charge domain. They combat the challenges like sneak paths and high static power faced by resistive crossbar arrays but are susceptible to thermal noise limiting the effective number of bits (ENOB) for the weighted sum. A direct way to reduce thi

Gian Marco Orlando, Jinyi Ye, Valerio La Gatta, Mahdi Saeedi

Generative agents are rapidly advancing in sophistication, raising urgent questions about how they might coordinate when deployed in online ecosystems. This is particularly consequential in information operations (IOs), influence campaigns that aim to manipulate public opinion on social media. While traditional IOs have been orchestrated by human operators a

Haochen Jiang, Enci Wang, Kai Wang, Chengyu Ma

The diminishing of star formation is accompanied by size differentiating, as quiescent galaxies are more compact than star-forming galaxies at fixed stellar mass. In order to understand how galaxy quenching is related to galaxy sizes, we performed a demographic study of 46 massive quiescent central galaxies with stellar mass from $10^{10.5}\rm M_\odot$ to $1

Oriel Kiss, Ivano Tavernelli, Francesco Tacchino, Denis Lacroix

The dynamical evolution of neutrino flavor in supernovae can be modeled by an all-to-all spin Hamiltonian with random couplings. Simulating such two-local Hamiltonian dynamics remains a major challenge, as methods with controllable accuracy require circuit depths that increase at least linearly with system size, exceeding the capabilities of current quantum

Christian Kremer, Marco Volpe

The goal of this paper is to put the theory of approximate fibrations into the framework of higher topos theory. We define the notion of an approximate fibration for a general geometric morphism of $\infty$-topoi, give several characterizations in terms of shape theory and compare it to the original definition for maps of topological spaces of Coram and Duva

J. Philippe, F. Elson, T. Arh, S. Sanz

In quantum magnetic materials it is common to observe both static and dynamic lattice effects on the magnetic excitation spectrum. Less common is to find that the magnetic correlations have a significant impact on the phonon spectrum. Can such an interplay occur in a structurally soft system with comparable elastic and magnetic energy scales? Here we study t

Benjamin Grando Moreira

Evaluating reasoning ability in Large Language Models (LLMs) is important for advancing artificial intelligence, as it transcends mere linguistic task performance. It involves understanding whether these models truly understand information, perform inferences, and are able to draw conclusions in a logical and valid way. This study compare logical and abstrac