May 2025 arXiv papers — page 75

Ananth Muppidi, Tarak Das, Sambaran Bandyopadhyay, Tripti Shukla

The generation of presentation slides automatically is an important problem in the era of generative AI. This paper focuses on evaluating multimodal content in presentation slides that can effectively summarize a document and convey concepts to a broad audience. We introduce a benchmark dataset, RefSlides, consisting of human-made high-quality presentations

Che Liu, Haozhe Wang, Jiazhen Pan, Zhongwei Wan

Improving performance on complex tasks and enabling interpretable decision making in large language models (LLMs), especially for clinical applications, requires effective reasoning. Yet this remains challenging without supervised fine-tuning (SFT) on costly chain-of-thought (CoT) data distilled from closed-source models (e.g., GPT-4o). In this work, we pres

Haihong Xiao, Jianan Zou, Yuxin Zhou, Ying He

We present SplatCo, a structure-view collaborative Gaussian splatting framework for high-fidelity rendering of complex outdoor scenes. SplatCo builds upon three novel components: 1) a cross-structure collaboration module that combines global tri-plane representations, which capture coarse scene layouts, with local context grid features representing fine deta

Tom Bleckmann, Paul Tschisgale

In recent years, natural language processing (NLP) has become integral to educational data mining, particularly in the analysis of student-generated language products. For research and assessment purposes, so-called embedding models are typically employed to generate numeric representations of text that capture its semantic content for use in subsequent quan

Mieke Wessel, Svenja zur Verth

Let $f(\mathbf x)$ be a non-singular quadratic form with sufficiently many mixed terms and $t$ an integer. For a sequence of weights $\mathcal A$ we study the number of weighted solutions to $f(\mathbf x) = t$. In particular, we give conditions on both $\mathcal A$ and $f$ such that we can use the circle method to count such solutions of bounded height.

Jianheng Zhuo, Yifan Yang, Yiwen Shao, Yong Xu

Automatic speech recognition (ASR) has made remarkable progress but heavily relies on large-scale labeled data, which is scarce for low-resource languages like Vietnamese. While existing systems such as Whisper, USM, and MMS achieve promising performance, their efficacy remains inadequate in terms of training costs, latency, and accessibility. To address the

Siddhanta Parial, Sasthi C. Ghosh, Anil K. Ghosh

In unmanned aerial vehicle (UAV) assisted millimeter wave (mmWave) communication, appropriate user-UAV association is crucial for improving system performance. In mmWave communication, user throughput largely depends on the line of sight (LoS) connectivity with the UAV, which in turn depends on the mobility pattern of the users. Moreover, different traffic t

Reggie C. Pantig, Ali Övgün, Ángel Rincón

In this paper, we investigate the phenomenology of electrically charged black holes in a Lorentz-violating gravitational framework mediated by a background Kalb-Ramond (KR) antisymmetric tensor field. Employing the Gauss-Bonnet theorem in a non-asymptotically flat geometry, we derive analytic expressions for the weak deflection angle of light and massive par

Jinchi Dong, Richard S. J. Tol, Jinnan Wang

Estimating the effects of climate on economic output is crucial for formulating climate policy, but current empirical findings remain ambiguous. Using annual panel model and panel long-difference model with global subnational data from nearly all countries, we find robust evidence that weather shocks have a transient effect on output. The impact on economic

Antoine Houssard

In the field of artificial intelligence (AI) research, there seems to be a rapprochement between academics and industrial forces. The aim of this study is to assess whether and to what extent industrial domination in the field as well as the ever more frequent switch between academia and industry resulted in the adoption of industrial norms and practices by

J. A. Montanez-Barrera, Yanjun Ji, Michael R. von Spakovsky, David E. Bernal Neira

Mapping quantum approximate optimization algorithm (QAOA) circuits with non-trivial connectivity in fixed-layout quantum platforms, such as superconducting quantum processing units (QPUs), requires a transpilation process to match the circuit to the hardware layout. This step is critical for reducing error rates on noisy QPUs. Two approaches that improve the

Sahil Kundu, Surya Narayan Maharana, Manoranjan Mishra

The convection-diffusion-reaction system governing incompressible reactive fluids in porous media is studied, focusing on the \( A + B \to C \) reaction coupled with density-driven flow. The time-dependent Brinkman equation describes the velocity field, incorporating permeability variations modeled as an exponential function of the product concentration. Den

Di Chen, Natasa Vasiljevic, Andrei Sarua, Martin Kuball

Tuning the local film stress (and associated strain) provides a universal route towards exerting dynamic control on propagating fields in nanoscale geometries, and engineering controlled interactions between them. The majority of existing techniques are adapted for engineering either uniform stresses or fixed stress gradients, but there is a need to develop

Zigeng Chen, Xinyin Ma, Gongfan Fang, Ruonan Yu

Large Reasoning Models (LRMs) excel at complex tasks using Chain-of-Thought (CoT) reasoning. However, their tendency to overthinking leads to unnecessarily lengthy reasoning chains, dramatically increasing inference costs. To mitigate this issue, we introduce VeriThinker, a novel approach for CoT compression. Unlike conventional methods that fine-tune LRMs d

Francesca Balestrieri, Kevin Destagnol, Julian Lyczak, Jennifer Park

This paper initiates the systematic study of the number of points of bounded height on symmetric squares of weak Fano varieties. We provide a general framework for establishing the point count on $\text{Sym}^2 X$. In the specific case of surfaces, we relate this to the Manin--Peyre conjecture for $\text{Hilb}^2 X$, and prove the conjecture for an infinite fa

Manuel Lecha, Andrea Cavallo, Francesca Dominici, Ran Levi

Graph Neural Networks (GNNs) excel at learning from pairwise interactions but often overlook multi-way and hierarchical relationships. Topological Deep Learning (TDL) addresses this limitation by leveraging combinatorial topological spaces. However, existing TDL models are restricted to undirected settings and fail to capture the higher-order directed patter

Tianyou Li, Haijun Zou, Jiayuan Wu, Zaiwen Wen

Routing problems are canonical combinatorial optimization tasks with wide-ranging applications in logistics, transportation, and supply chain management. However, solving these problems becomes significantly more challenging when complex constraints are involved. In this paper, we propose LMask, a novel learning framework that utilizes dynamic masking to gen

Frank van der Meulen, Moritz Schauer, Stefan Sommer

We develop a general methodological framework for probabilistic inference in discrete- and continuous-time stochastic processes evolving on directed acyclic graphs (DAGs). The process is observed only at the leaf nodes, and the challenge is to infer its full latent trajectory: a smoothing problem that arises in fields such as phylogenetics, epidemiology, and

Zhihong Chen, Yiqian Yang, Jinzhao Zhou, Qiang Zhang

The rapid advancement of large language models (LLMs) raises critical concerns about their ethical alignment, particularly in scenarios where human and AI co-exist under the conflict of interest. This work introduces an extendable, asymmetric, multi-agent simulation-based benchmarking framework to evaluate the moral behavior of LLMs in a novel human-AI co-ex

Sebastian Gerstner, Hinrich Schütze

Interpretability researchers have attempted to understand MLP neurons of language models based on both the contexts in which they activate and their output weight vectors. They have paid little attention to a complementary aspect: the interactions between input and output. For example, when neurons detect a direction in the input, they might add much the sam

Stephon Alexander, Bruno Alexandre, Michael Fine, João Magueijo

Graviweak theory seeks to unify gravity (specifically in its self-dual formulation) with the weak interaction, preying on their parallel chiral $SU(2)$ structures. In this paper we further this idea by folding it with the concept of spontaneous symmetry breaking. We do this first with a standard Higgs field and potential, starting with a unifying parity-inva

Marcin Lawenda, Łukasz Szustak, László Környei, Flavio Cesar Cunha Galeazzo

In this work, the authors focus on assessing the impact of the AMD EPYC processor architecture on the performance of CFD applications. Several generations of architectures were analyzed, such as Rome, Milan, Milan X, Genoa, Genoa X and Bergamo, characterized by a different number of cores (64-128), L3 cache size (256 - 1152 MB) and RAM type (8-channel DDR4 o

Tom Britton, Andrea Pugliese

We consider a model for an influenza-like disease, in which, between seasons, the virus makes a random genetic drift $\delta$, (reducing immunity by the factor $\delta$) and obtains a new random transmissibility $\tau$ (closely related to $R_0$). Given the immunity status at the start of season $k$: $\textbf{p}^{(k)}$, describing community distribution of ye

Umberto Casti, Giacomo Baggio, Sandro Zampieri, Fabio Pasqualetti

A key claim in recent work on Selective State Space Models is that selectivity, the ability to focus on relevant information while filtering irrelevant inputs, requires breaking the Linear Time-Invariant (LTI) property through time-varying dynamics. We challenge this claim by demonstrating that LTI systems can achieve selectivity when designed using principl

Xingjian Li, Qifeng Wu, Adithya S. Ubaradka, Yiran Ding

Medical image segmentation is vital for clinical diagnosis, yet current deep learning methods often demand extensive expert effort, i.e., either through annotating large training datasets or providing prompts at inference time for each new case. This paper introduces a zero-shot and automatic segmentation pipeline that combines off-the-shelf vision-language

B. S. Cartwright, S. A. Wrathmall, R. M. Potvliege

The rapid turn-on of a strong, resonant, continuous wave laser field may trigger the formation of a transient oscillation akin to a train of damped solitons, before the vapor-field system relaxes into a stationary state. We study this transient dynamic on theoretical models of a rubidium vapor. We also consider doubly resonant V-systems, for which the transi

Alexander Gabitashvili, Philipp Kellmeyer

Intensive care unit (ICU) is a crucial hospital department that handles life-threatening cases. Nowadays machine learning (ML) is being leveraged in healthcare ubiquitously. In recent years, management of ICU became one of the most significant parts of the hospital functionality (largely but not only due to the worldwide COVID-19 pandemic). This study explor

Junyi Lu, Lili Jiang, Xiaojia Li, Jianbing Fang

The complexity of code reviews has driven efforts to automate review comments, but prior approaches oversimplify this task by treating it as snippet-level code-to-text generation and relying on text similarity metrics like BLEU for evaluation. These methods overlook repository context, real-world merge request evaluation, and defect detection, limiting their

Valentim Romão, Rafael Soares, Luís Rodrigues, Vasco Manquinho

The advent of microservices has led multiple companies to migrate their monolithic systems to this new architecture. When decomposing a monolith, a functionality previously implemented as a transaction may need to be implemented as a set of independent sub-transactions, possibly executed by multiple microservices. The concurrent execution of decomposed funct

Simone Ciani, Ugo Gianazza, Zheng Li

We study Phragm\'en-Lindel\"of-type theorems for functions $u$ in homogeneous De Giorgi classes, and we show that the maximum modulus $\mu_+(r)$ of $u$ has a power-like growth of order $\alpha\in(0,1)$ when $r\to\infty$. By proper counterexamples, we show that in general we cannot expect $\alpha$ to be $1$.

Jiancheng Wang, Mingjia Yin, Hao Wang, Enhong Chen

Modeling feature interactions is essential for accurate click-through rate (CTR) prediction in advertising systems. Recent studies have adopted the Mixture-of-Experts (MoE) approach to improve performance by ensembling multiple feature interaction experts. These studies employ various strategies, such as learning independent embedding tables for each expert

Shokoufe Faraji

We investigate the standard relativistic geometrically thin and optically thick accretion disk in the background of a deformed compact object. The main purpose of this work is to determine whether such a deformed object possesses its own observational fingerprint that can distinguish it from Schwarzschild and Kerr black holes. Our analysis reveals the proper

Yuekun Yao, Yupei Du, Dawei Zhu, Michael Hahn

Implicit reasoning is the ability of a language model to solve multi-hop reasoning tasks in a single forward pass, without chain of thought. We investigate this capability using GPT2-style language models trained from scratch on controlled $k$-hop reasoning datasets ($k = 2, 3, 4$). We show that while such models can indeed learn implicit $k$-hop reasoning,

Yu-Chien Ning, Xin Zhou, Francine Laden, Molin Wang

We introduce the SoftBart approach from Bayesian ensemble learning to estimate the relationship between multipollutant mixtures and health on chronic exposures in epidemiology research. This approach offers several key advantages over existing methods: (1) it is computationally efficient and well-suited for analyzing large datasets; (2) it is flexible in est

Carlos Salazar-Ruiz, Francisco Lopez-Tiro, Ivan Reyes-Amezcua, Clement Larose

Determining the type of kidney stones is crucial for prescribing appropriate treatments to prevent recurrence. Currently, various approaches exist to identify the type of kidney stones. However, obtaining results through the reference ex vivo identification procedure can take several weeks, while in vivo visual recognition requires highly trained specialists

Giuliano Angelone, Manuel Asorey, Fernando Ezquerro, Paolo Facchi

We consider a generalization of Dirac's comb model, describing a non-relativistic particle moving in a periodic array of generalized point interactions. The latter represent the most general point interactions rendering the kinetic-energy operator self-adjoint, and form a four-parameters family that includes the $\delta$-potential and the $\delta'$-potential

Mingquan Feng, Yifan Fu, Tongcheng Zhang, Yu Jiang

Despite the widely recognized success of residual connections in modern neural networks, their design principles remain largely heuristic. This paper introduces KITINet (Kinetics Theory Inspired Network), a novel architecture that reinterprets feature propagation through the lens of non-equilibrium particle dynamics and partial differential equation (PDE) si

Hangting Ye, Jinmeng Li, He Zhao, Dandan Guo

Tabular data have been playing a vital role in diverse real-world fields, including healthcare, finance, etc. With the recent success of Large Language Models (LLMs), early explorations of extending LLMs to the domain of tabular data have been developed. Most of these LLM-based methods typically first serialize tabular data into natural language descriptions

Xingyu Li, Qing Liu, Tony Jiang, Hong Amy Xia

We propose a novel method, termed the M-learner, for estimating heterogeneous indirect and total treatment effects and identifying relevant subgroups within a mediation framework. The procedure comprises four key steps. First, we compute individual-level conditional average indirect/total treatment effect Second, we construct a distance matrix based on pairw

Xujia Gong, Amar Fakhredine, Carmine Autieri

We study the magnetic and electronic properties of Ni-intercalated NbSe$_2$.We calculate the magnetic exchanges of Ni$_x$NbSe$_2$ ($x = 1/3, 1/4,$ and $1$) and find that the out-of-plane magnetic coupling depends on the Ni connectivity: it is ferromagnetic when Ni atoms stack on top of each other, and antiferromagnetic otherwise. Focusing on Ni$_{0.25}$NbSe$

Manjari Bagchi, Prasanta Bera, Aru Beri, Dipankar Bhattacharya

Astrophysical compact objects, viz., white dwarfs, neutron stars, and black holes, are the remnants of stellar deaths at the end of their life cycles. They are ideal testbeds for various fundamental physical processes under extreme conditions that are unique in nature. Observational radio astronomy with uGMRT and OORT facilities has led to several important

Lynn Karam, Yipei Wang, Veeru Kasivisvanathan, Mirabela Rusu

Automated segmentation of cancer on medical images can aid targeted diagnostic and therapeutic procedures. However, its adoption is limited by the high cost of expert annotations required for training and inter-observer variability in datasets. While weakly-supervised methods mitigate some challenges, using binary histology labels for training as opposed to

Nayoung Kim, Seongsu Kim, Sungsoo Ahn

Designing metal-organic frameworks (MOFs) with novel chemistries is a longstanding challenge due to their large combinatorial space and complex 3D arrangements of the building blocks. While recent deep generative models have enabled scalable MOF generation, they assume (1) a fixed set of building blocks and (2) known local 3D coordinates of building blocks.

Anna Duwenig

We identify which conditions on an open normal subgroupoid of a LCH \'etale groupoid with twist are necessary and sufficient for the subgroupoid's reduced twisted C*-algebra to be a C*-diagonal in the ambient groupoid C*-algebra. We do so by first giving an explicit description of the Weyl groupoid and Weyl twist associated to any non-traditional Cartan suba

Luohong Wu, Matthias Seibold, Nicola A. Cavalcanti, Giuseppe Loggia

Bone surface reconstruction is an essential component of computer-assisted orthopedic surgery(CAOS), forming the foundation for both preoperative planning and intraoperative guidance. Compared to traditional imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI),ultrasound, an emerging CAOS technology, provides a radiation-f

Zheyang Huang, Jagannath Aryal, Saeid Nahavandi, Xuequan Lu

Cross-view geo-localization determines the location of a query image, captured by a drone or ground-based camera, by matching it to a geo-referenced satellite image. While traditional approaches focus on image-level localization, many applications, such as search-and-rescue, infrastructure inspection, and precision delivery, demand object-level accuracy. Thi

Bin Wu, Wei Wang, Yahui Liu, Zixiang Li

Reward Feedback Learning (ReFL) has recently shown great potential in aligning model outputs with human preferences across various generative tasks. In this work, we introduce a ReFL framework, named DiffusionReward, to the Blind Face Restoration task for the first time. DiffusionReward effectively overcomes the limitations of diffusion-based methods, which

Bram Grooten, Farid Hasanov, Chenxiang Zhang, Qiao Xiao

Model ensembles have long been a cornerstone for improving generalization and robustness in deep learning. However, their effectiveness often comes at the cost of substantial computational overhead. To address this issue, state-of-the-art methods aim to replicate ensemble-class performance without requiring multiple independently trained networks. Unfortunat

Litao Guo, Xinli Xu, Luozhou Wang, Jiantao Lin

With the rapid advancement of generative models, general-purpose generation has gained increasing attention as a promising approach to unify diverse tasks across modalities within a single system. Despite this progress, existing open-source frameworks often remain fragile and struggle to support complex real-world applications due to the lack of structured w

Ka Long Keith Ho, Yoshinari Takeishi, Junichi Takeuchi

Properties of Fisher information matrices of 2-layer neural ReLU networks with random hidden weights are studied. For these networks, it is known that the eigenvalue distribution highly concentrates on several eigenspaces approximately. In particular, the eigenvalues for the first three eigenspaces account for 97.7% of the trace of the Fisher information mat

Rounak Chatterjee, Vikas S Bhat, Kiran Bajar, Sushil Mujumdar

High-dimensional entanglement in the form of transverse spatial correlation between a pair of photons generated via spontaneous parametric downconversion is not only a valuable resource in many academic and real-life applications but also provides access to several intriguing quantum phenomena. One such non-intuitive phenomenon is phase entanglement, in whic

Xie Ting, Ye Huang, Zhilin Liu, Lixin Duan

In real-world scenarios, pixel-level labeling is not always available. Sometimes, we need a semantic segmentation network, and even a visual encoder can have a high compatibility, and can be trained using various types of feedback beyond traditional labels, such as feedback that indicates the quality of the parsing results. To tackle this issue, we proposed

Bim Gustavsson, Stacey Law

Let $p$ be any prime. We determine precisely those irreducible characters of symmetric groups which contain at most $p$ distinct linear constituents in their restriction to a Sylow $p$-subgroup, answering a question of Giannelli and Navarro. Moreover, we identify all of the linear constituents of such characters, and in the case $p = 2$ explicitly calculate

Jan Hajer

The calculation of particle decay widths and scattering cross sections naturally decomposes into a quantum mechanical amplitude and a relativistic phase space (PS). This PS can be formulated in terms of parallelotopes providing frame independent invariants. We demonstrate how these invariants are related to frame dependent observables such as momenta, energi

Ignacio Cabrera Martin, Subhaditya Mukherjee, Almas Baimagambetov, Joaquin Vanschoren

In an era defined by rapid data evolution, traditional Machine Learning (ML) models often struggle to adapt to dynamic environments. Evolving Machine Learning (EML) has emerged as a pivotal paradigm, enabling continuous learning and real-time adaptation to streaming data. While prior surveys have examined individual components of evolving learning - such as

James Hughes, Agniva Roy

Let $\lambda$ be a Legendrian link in standard contact $\mathbb{R}^3$, such that $L_1$, $L_2$ are two exact fillings of $\lambda$ and $\varphi$ is a Legendrian loop of $\lambda$. We study fillability and isotopy characterizations of Legendrian surfaces in standard contact $\mathbb{R}^5$ built from the above data by doubling or twist spinning; denoting them $

Denis Sakhno, Pavel A. Belov

This paper revisits a model for the plasma frequency of a simple wire medium formed by a rectangular lattice of parallel metallic wires. We provide a comparative analysis of existing formulae for estimating the plasma frequency and derive a new expression. The proposed formula demonstrates superior accuracy for a square lattice of thin wires, with a relative

Romain Mussard, Fannia Pacheco, Maxime Berar, Gilles Gasso

Deep learning models have significantly improved the ability to detect novelties in time series (TS) data. This success is attributed to their strong representation capabilities. However, due to the inherent variability in TS data, these models often struggle with generalization and robustness. To address this, a common approach is to perform Unsupervised Do

Srishti Bhardwaj, T. Maitra

We propose a mechanism where the valley splitting is caused by an in-plane electric polarization and the coupling between the two makes it possible for an electric field to control the valley degree of freedom. We demonstrate this by considering Gd-substituted EuCl$_2$ monolayer in its 1T-phase using first-principles calculations. This monolayer exhibits an

Zi-Ao Ma, Tian Lan, Rong-Cheng Tu, Shu-Hang Liu

The rapid progress in diffusion-based text-to-image (T2I) generation has created an urgent need for interpretable automatic evaluation methods that can assess the quality of generated images, therefore reducing the human annotation burden. To reduce the prohibitive cost of relying on commercial models for large-scale evaluation, and to improve the reasoning

Debanshu Ratha, Madhu Koirala, Pål Gunnar Ellingsen

Volume estimation of onshore cargo piles is of economic importance for shipping and mining companies as well as public authorities for real-time planning of logistics, business intelligence, transport services by land or sea and governmental oversight. In remote sensing literature, the volume of pile is estimated by relying on the illumination property of ob

Dan A. Calian, Gregory Farquhar, Iurii Kemaev, Luisa M. Zintgraf

The quality of foundation models depends heavily on their training data. Consequently, great efforts have been put into dataset curation. Yet most approaches rely on manual tuning of coarse-grained mixtures of large buckets of data, or filtering by hand-crafted heuristics. An approach that is ultimately more scalable (let alone more satisfying) is to \emph{l

Khalil Hennara, Muhammad Hreden, Mohamed Motaism Hamed, Zeina Aldallal

We introduce Mutarjim, a compact yet powerful language model for bidirectional Arabic-English translation. While large-scale LLMs have shown impressive progress in natural language processing tasks, including machine translation, smaller models. Leveraging this insight, we developed Mutarjim based on Kuwain-1.5B , a language model tailored for both Arabic an

Shruti Atul Mali, Zohaib Salahuddin, Danial Khan, Yumeng Zhang

Purpose: This study evaluates the impact of harmonization and multi-region feature integration on survival prediction in non-small cell lung cancer (NSCLC) patients. We assess the prognostic utility of handcrafted radiomics and pretrained deep features from thoracic CT images, integrating them with clinical data using a multicentre dataset. Methods: Survival

Romina M. Arroyo, Gabriela P. Ovando, Mariel Sáez

In this work we study the existence of solutions to the Mean Curvature Flow for which the initial condition has the structure of a two-dimensional Lie subgroup within a Lie group of dimension three. We consider Lie groups with a fixed left-invariant metric and first observe that if the Lie group is unimodular, then every Lie subgroup is a minimal surface (he

Ignacio Amores-Sesar, Christian Cachin, Juan Villacis, Luca Zanolini

In protocols with asymmetric trust, each participant is free to make its own individual trust assumptions about others, captured by an asymmetric quorum system. This contrasts with ordinary, symmetric quorum systems and with threshold models, where all participants share the same trust assumption. It is already known how to realize reliable broadcasts, share

Frank Ball, Tom Britton, Peter Neal

We analyse a generalized stochastic household epidemic model defined by a bivariate random variable $(X_G, X_L)$, representing the number of global and local infectious contacts that an infectious individual makes during their infectious period. Each global contact is selected uniformly among all individuals and each local contact is selected uniformly among

Xixian Yong, Xiao Zhou, Yingying Zhang, Jinlin Li

The recent rise of Large Reasoning Models (LRMs) has significantly improved multi-step reasoning performance, but often at the cost of generating excessively long reasoning chains. This paper revisits the efficiency of such reasoning processes through an information-theoretic lens, revealing a fundamental trade-off between reasoning length and semantic effic

G. V. Afonin, S. L. Scherbakov, R. A. Konchakov, N. P. Kobelev

We performed calorimetric and torsion stress relaxation measurements upon linear heating of six conventional and high-entropy metallic glasses with the mixing entropy {\Delta}Smix ranging from 0.86R to 1.79R (R is the universal gas constant). It is shown that high-entropy metallic glasses ({\Delta}Smix > 1.5 R) exhibit significantly greater resistance to str

Mingbo Dou, Xianjie Wang, L. L. Tao

The altermagnet exhibits the nonrelativistic spin splitting that enables all-electrical generation of spin-polarized currents beyond the spin-orbit coupling. Here, we report on a study on the anisotropic spin-polarized conductivity in collinear altermagnets. Based on the Boltzmann transport theory, we first study this effect using the general group-theoretic

Lukas Lanza, Johannes Köhler, Dario Dennstädt, Thomas Berger

Control barrier functions (CBFs) are a popular approach to design feedback laws that achieve safety guarantees for nonlinear systems. The CBF-based controller design relies on the availability of a model to select feasible inputs from the set of CBF-based controls. In this paper, we develop a model-free approach to design CBF-based control laws, eliminating

Natalia Matuszczyk, Craig R. Barnes, Rohit Gupta, Bulent Ozel

Bias in geospatial artificial intelligence (GeoAI) models has been documented, yet the evidence is scattered across narrowly focused studies. We synthesize this fragmented literature to provide a concise overview of bias in GeoAI and examine how the EU's Artificial Intelligence Act (EU AI Act) shapes audit obligations. We discuss recurring bias mechanisms, i

Tan Liu, Wen-Fan Feng, Zong-Kuan Guo

Gravitational waves undergo redshift as they propagate through the expanding universe, and the redshift may exhibit time-dependent drift. Consequently, for any isolated gravitational wave sources, the mass parameter $\mathcal{M}$ and the redshift $z$ exhibit an observational degeneracy, typically manifesting in the waveform as the redshifted mass $\mathcal{M

Pranav Garimidi, Lioba Heimbach, Tim Roughgarden

We initiate the study of transaction fee mechanism design for blockchain protocols in which multiple block producers contribute to the production of each block. Our contributions include: - We propose an extensive-form (multi-stage) game model to reason about the game theory of multi-proposer transaction fee mechanisms. - We define the strongly BPIC property

Nikita Ivanov, Mark Klimov, Dmitry Glukhikh, Tatiana Chernysheva

Modern machine learning methods require significant amounts of labelled data, making the preparation process time-consuming and resource-intensive. In this paper, we propose to consider the process of prototyping a tool for annotating and generating training datasets based on video tracking and segmentation. We examine different approaches to solving this pr

Laines Schmalwasser, Niklas Penzel, Joachim Denzler, Julia Niebling

Concepts such as objects, patterns, and shapes are how humans understand the world. Building on this intuition, concept-based explainability methods aim to study representations learned by deep neural networks in relation to human-understandable concepts. Here, Concept Activation Vectors (CAVs) are an important tool and can identify whether a model learned a

Cole Wyeth, Marcus Hutter

We rigorously discuss the commonly asserted failures of the AIXI reinforcement learning agent as a model of embedded agency. We attempt to formalize these failure modes and prove that they occur within the framework of universal artificial intelligence, focusing on a variant of AIXI that models the joint action/percept history as drawn from the universal dis

Alexander Modell, Patrick Rubin-Delanchy, Nick Whiteley

There is a large ongoing scientific effort in mechanistic interpretability to map embeddings and internal representations of AI systems into human-understandable concepts. A key element of this effort is the linear representation hypothesis, which posits that neural representations are sparse linear combinations of `almost-orthogonal' direction vectors, refl

Wenjin Qin, Hailin Wang, Hao Shu, Feng Zhang

In recent years, tensor decomposition-based approaches for hyperspectral anomaly detection (HAD) have gained significant attention in the field of remote sensing. However, existing methods often fail to fully leverage both the global correlations and local smoothness of the background components in hyperspectral images (HSIs), which exist in both the spectra

Wen-Long Zhang, Chen-Ran Hu, Chen Du, Wen-Jun Tan

Recent observations of the repeating fast radio burst FRB 20201124A by the Five-hundred-meter Aperture Spherical radio Telescope (FAST) revealed a second-scale periodic modulation ($\sim$1.7\,s) in burst activity during two distinct observational windows. We find that these two periodic activity epochs temporally coincide with the transitional states of the

Zengrui Han, Lu Bai, Ziwei Huang, Xiang Cheng

In this paper, a novel large language model (LLM)-based method for scatterer generation (LLM4SG) is proposed for sixth-generation (6G) artificial intelligence (AI)-native communications. To provide a solid data foundation, we construct a new synthetic intelligent sensing-communication dataset for Synesthesia of Machines (SoM) in vehicle-to-vehicle (V2V) comm

Matthias Grätsch

We study families of analytic and meromorphic functions with bounded generalized Schwarzian derivative $S_k(f)$. We show that these families are quasi-normal. Further, we investigate associated families, such as those formed by derivatives and logarithmic derivatives, and prove several (quasi-)normality results. Moreover, we derive a new formula for $S_k(f)$

François Derrida, Shahar Lutati, Eliya Nachmani

Active Noise Cancellation (ANC) algorithms aim to suppress unwanted acoustic disturbances by generating anti-noise signals that destructively interfere with the original noise in real time. Although recent deep learning-based ANC algorithms have set new performance benchmarks, there remains a shortage of theoretical limits to rigorously assess their improvem

Arpita Goswami, Pallabi Chatterjee, Ranjan Modak, Shaon Sahoo

We consider a ladder system where one leg, referred to as the ``bath", is governed by an Aubry-André (AA) type Hamiltonian, while the other leg, termed the ``subsystem", follows a standard tight-binding Hamiltonian. We investigate the localization properties in the subsystem induced by its coupling to the bath. For the coupling strength larger than a

Yan Zhong, Xingyu Wu, Xinping Zhao, Li Zhang

In practical domains, high-dimensional data are usually associated with diverse semantic labels, whereas traditional feature selection methods are designed for single-label data. Moreover, existing multi-label methods encounter two main challenges in semi-supervised scenarios: (1). Most semi-supervised methods fail to evaluate the label correlations without

J. B. Cui, R. A. Konchakov, G. V. Afonin, A. S. Makarov

We performed calorimetric studies of 26 metallic glasses and calculated the excess entropy and excess enthalpy with respect to their counterpart crystals. On this basis, we introduced a dimensionless entropy-based parameter {\sigma}scl, which characterizes structural disordering and stability of the supercooled liquid state upon heating. A very good correlat

Wanhao Liu, Zonglin Yang, Jue Wang, Lidong Bing

Hypothesis ranking is vital for automated scientific discovery, especially in cost-intensive, throughput-limited natural science domains. Current methods focus on pre-experiment ranking, relying solely on language model reasoning without empirical feedback. We introduce experiment-guided ranking, which prioritizes hypotheses based on feedback from prior test

Licheng Pan, Zhichao Chen, Haoxuan Li, Guangyi Liu

Multi-task forecasting has become the standard approach for time-series forecasting (TSF). However, we show that it suffers from an Expressiveness Bottleneck, where predictions at different time steps share the same representation, leading to unavoidable errors even with optimal representations. To address this issue, we propose a two-stage framework: first,

Zezhi Shao, Yujie Li, Fei Wang, Chengqing Yu

The advent of universal time series forecasting models has revolutionized zero-shot forecasting across diverse domains, yet the critical role of data diversity in training these models remains underexplored. Existing large-scale time series datasets often suffer from inherent biases and imbalanced distributions, leading to suboptimal model performance and ge

Shaina Raza, Rizwan Qureshi, Azib Farooq, Marcelo Lotif

Large language models (LLMs) reproduce misinformation not by memorizing false facts alone, but by learning the linguistic patterns that make falsehoods persuasive, such as hedging, false presuppositions, and fabricated citations. We propose model immunization, a training paradigm based on supervised fine-tuning over curated (false claim, correction) pairs, i

Mohammad Shahverdikondori, Mohammad Reza Badri, Negar Kiyavash

We introduce the Best Group Identification problem in a multi-objective multi-armed bandit setting, where an agent interacts with groups of arms with vector-valued rewards. The performance of a group is determined by an efficiency vector which represents the group's best attainable rewards across different dimensions. The objective is to identify the set of

Konstantinos Spathis, Nikolaos Kardaris, Petros Maragos

In practical applications, computer vision tasks often need to be addressed simultaneously. Multitask learning typically achieves this by jointly training a single deep neural network to learn shared representations, providing efficiency and improving generalization. Although action and gesture recognition are closely related tasks, since they focus on body

Hongshu Guo, Zeyuan Ma, Yining Ma, Xinglin Zhang

Designing effective black-box optimizers is hampered by limited problem-specific knowledge and manual control that spans months for almost every detail. In this paper, we present \textit{DesignX}, the first automated algorithm design framework that generates an effective optimizer specific to a given black-box optimization problem within seconds. Rooted in t

Thiebout Delabie, Claudio Llosa Isenrich, Romain Tessera

We classify compactly generated locally compact groups of polynomial growth up to $L^p$ measure equivalence (ME) for all $p\leq 1$. To achieve this, we combine rigidity results (previously proved for discrete groups by Bowen and Austin) with new constructions of explicit orbit equivalences between simply connected nilpotent Lie groups. In particular, we prov

Faizal Amir, Alimuddin S. Miru, Edy Sabara

The 3R-based Zero Waste approach aims to minimize household solid waste through the principles of Reduce, Reuse, and Recycle. This study examines the relationship between household environmental knowledge, personal attitude, subjective norms, and perceived behavioral control as key behavioral predictors. A structured survey was conducted among 1,200 urban ho

Nicolas Zucchet, Francesco d'Angelo, Andrew K. Lampinen, Stephanie C. Y. Chan

Emergence is a fascinating property of large language models and neural networks more broadly: as models scale and train for longer, they sometimes develop new abilities in sudden ways. Despite initial studies, we still lack a comprehensive understanding of how and when these abilities emerge. To address this gap, we study the emergence over training of spar

Ziwei Zhou, Rui Wang, Zuxuan Wu, Yu-Gang Jiang

Recent Multimodal Large Language Models (MLLMs) achieve promising performance on visual and audio benchmarks independently. However, the ability of these models to process cross-modal information synchronously remains largely unexplored. We introduce Daily-Omni, a multiple-choice Audio-Visual QA benchmark featuring 684 real-world videos and 1,197 questions s

Jianghao Lin, Jiachen Zhu, Zheli Zhou, Yunjia Xi

Over the past decades, superplatforms, digital companies that integrate a vast range of third-party services and applications into a single, unified ecosystem, have built their fortunes on monopolizing user attention through targeted advertising and algorithmic content curation. Yet the emergence of AI agents driven by large language models (LLMs) threatens

Wenning Xu, Shiyu Fan, Paul Henderson, Edmond S. L. Ho

Generating realistic human motion with high-level controls is a crucial task for social understanding, robotics, and animation. With high-quality MOCAP data becoming more available recently, a wide range of data-driven approaches have been presented. However, modelling multi-person interactions still remains a less explored area. In this paper, we present Gr

Masahiro Fujisawa, Masaki Adachi, Michael A. Osborne

Despite the importance of aligning language models with human preferences, crowd-sourced human feedback is often noisy -- for example, preferring less desirable responses -- posing a fundamental challenge to alignment. A truly robust alignment objective should yield identical model parameters even under severe label noise, a property known as redescending. W