November 2025 arXiv papers — page 194

Vesna Poprcova, Iulia Lefter, Martijn Warnier, Frances Brazier

Social anxiety is a prevalent mental health condition that can significantly impact overall well-being and quality of life. Despite its widespread effects, adequate support or treatment for social anxiety is often insufficient. Advances in technology, particularly in social robotics, offer promising opportunities to complement traditional mental health. As a

Olivier Fercoq

In this paper, we minimize the self-centered smoothed gap, a recently introduced optimality measure, in order to solve convex-concave saddle point problems. The self-centered smoothed gap can be computed as the sum of a convex, possibly nonsmooth function and a smooth weakly convex function. Although it is not convex, we propose an algorithm that minimizes t

Doria Bonzi, Alexandre Guiggi, Frédéric Béchet, Carlos Ramisch

Critical appraisal of scientific literature is an essential skill in the biomedical field. While large language models (LLMs) can offer promising support in this task, their reliability remains limited, particularly for critical reasoning in specialized domains. We introduce CareMedEval, an original dataset designed to evaluate LLMs on biomedical critical ap

Lucas Slot, David Steurer, Manuel Wiedmer

Efficient algorithms for convex optimization, such as the ellipsoid method, require an a priori bound on the radius of a ball around the origin guaranteed to contain an optimal solution if one exists. For linear and convex quadratic programming, such solution bounds follow from classical characterizations of optimal solutions by systems of linear equations.

Matteo Barborini, Jorge Charry, Matej Ditte, Andronikos Leventis

We present the first open access version of the QMeCha (Quantum MeCha) code, a quantum Monte Carlo (QMC) package developed to study many-body interactions between different types of quantum particles, with a modular and easy-to-expand structure. The present code has been built to solve the Hamiltonian of a system that can include nuclei and fermions of diffe

V. A. Allakhverdyan, A. D. Avrorin, A. V. Avrorin, V. M. Aynutdinov

We present a fast data processing system for the Baikal-GVD neutrino telescope, designed for rapid identification of astrophysical neutrino events. Leveraging Baikal-GVD's modular cluster architecture, the system implements parallelized file processing where raw data files undergo concurrent analysis across dedicated virtual machines. The system implements t

Md Mizanur Rahaman Nayan, Zheyu Li, Flavio Ponzina, Sumukh Pinge

Database search and clustering are fundamental components of many data analytics problems, such as mass spectrometry-driven proteomics. Traditional full clustering and search algorithms suffer from high resource usage and long latencies. We introduce HERP, a lightweight incremental clustering method and a highly parallelizable database (DB) search platform t

Demetrio Vilardi, Pietro M. Bonetti

We consider the SU(2) gauge theory for spin fluctuations in the two-dimensional Hubbard model, where the electron field is fractionalized in terms of spinons and chargons. In this theory, spinons are described by a non-linear sigma model, while chargons are treated as fermions at a mean-field level. We investigate the instability to a superconducting state S

Maciej Korpalski, Grzegorz Plebanek

We present several results providing lower bounds for the Banach-Mazur distance \[d_{BM}(C(K), C(L))\] between Banach spaces of continuous functions on compact spaces. The main focus is on the case where $C(L)$ represents the classical Banach space $c$ of convergent sequences. In particular, we obtain generalizations and refinements of recent results from \c

Botao 'Amber' Hu, Helena Rong

As the "agentic web" takes shape-billions of AI agents (often LLM-powered) autonomously transacting and collaborating-trust shifts from human oversight to protocol design. In 2025, several inter-agent protocols crystallized this shift, including Google's Agent-to-Agent (A2A), Agent Payments Protocol (AP2), and Ethereum's ERC-8004 "Trustless Agents," yet thei

Arvid Siqveland

In the authors book, Associative Algebraic Geometry, 2023, and the following article Shemes of Associative Algebras,\\ https://doi.org/10.48550/arXiv.2410.17703,2024, we use an algebraization of the semi-local formal moduli of simple modules to construct associative schemes. Here, we consider a commutative ring for which we can use the localization in maxima

Wenfeng Zhou, Xin Wang, Xun Zhang, Yuqi Chen

The rise of artificial intelligence has triggered exponential growth in data volume, demanding rapid and efficient processing. High-speed, energy-efficient, and parallel-scalable computing hardware is thus increasingly critical. We demonstrate a wafer-scale non-volatile topological photonic computing chip using topological modulators. Leveraging the GHz-spee

Kwang-Wu Chen

We study two families of zeta-like multiple series -- the multiple $\rho$-values and the multiple $\eta$-values -- defined by nested sums with shifted denominators. An explicit factorial formula for $\rho$ reveals its intrinsic combinatorial structure and leads to closed expressions for fixed weight and depth. A remarkable identity emerges from a weighted-su

Seth Hardy, Max Wenqiang Xu

Let $\Psi(x,y)$ denote the count of $y$-smooth numbers below $x$ and $P(n)$ denote the largest prime factor of $n$. We prove that for $f$ a Steinhaus random multiplicative function, the partial sums over $y$-smooth numbers always enjoy better than squareroot cancellation, in the sense that $$ \mathbb{E} \Big|\sum_{\substack{1\leq n \leq x\\ P(n) \leq y}} f(n

Seema Chahal, Sugandha Maheshwary

In this article, we study the metacyclic p-group codes arising from finite semisimple group algebras. In [CM25], we studied group codes arising from metacyclic groups with order divisible by two distinct odd primes. In the current work, we focus on metacyclic p-group codes, as a result of which we are also able to extend the results of [CM25] for metacyclic

Zhichao Chen, Zelin Jia

The generalized Markov equations are deeply connected with the generalized cluster algebras of Markov type. We construct a deformed Fock-Goncharov tropicalization for the generalized Markov equations and prove that their tropicalized tree structure is essentially the same as that of the classical Euclid tree. We then define the generalized Euclid tree and pr

Florentia Afentaki, Maha Shatta, Konstantinos Balaskas, Georgios Panagopoulos

Flexible Electronics (FE) have emerged as a promising alternative to silicon-based technologies, offering on-demand low-cost fabrication, conformality, and sustainability. However, their large feature sizes severely limit integration density, imposing strict area and power constraints, thus prohibiting the realization of Machine Learning (ML) circuits, which

Guoqing Cui, Ling Wang, Bin Zhou

In this paper, we establish local potential estimates and H\"older estimates for solutions of linearized Monge-Amp\`ere equations with the right-hand side being a signed measure, under suitable assumptions on the data. In particular, the interior H\"older estimate holds for an inhomogeneous linearized Monge-Amp\`ere equation with right-hand side being the no

Ilya Borovik, Dmitrii Gavrilev, Vladimir Viro

Emotions are fundamental to the creation and perception of music performances. However, achieving human-like expression and emotion through machine learning models for performance rendering remains a challenging task. In this work, we present SyMuPe, a novel framework for developing and training affective and controllable symbolic piano performance models. O

Stuart Lane

The standard fuzzy regression discontinuity (FRD) estimator is a ratio of differences of local polynomial estimators. I show that this estimator does not possess any finite integer moments, regardless of local polynomial degree, kernel function, or bandwidth. The estimator is heavy-tailed in small samples or when the treatment probability discontinuity at th

Jiyoung Lee, Song Park, Sanghyuk Chun, Soo-Whan Chung

This paper proposes VoxStudio, the first unified and end-to-end speech-to-image model that generates expressive images directly from spoken descriptions by jointly aligning linguistic and paralinguistic information. At its core is a speech information bottleneck (SIB) module, which compresses raw speech into compact semantic tokens, preserving prosody and em

Haidong Wu, Shunzhe Zhang

A long-standing conjecture of Thomassen says that every longest cycle of a $3$-connected graph has a chord. Thomassen (2018) proved that if $G$ is a $2$-connected cubic graph, then any longest cycle must have a chord. He also showed that in any 3-connected graph with minimum degree at least four, some longest cycle must contain a chord. Harvey proved that ev

Shihai Wang, Tao Chen

Elicited performance requirements need to be quantified for compliance in different engineering tasks, e.g., configuration tuning and performance testing. Much existing work has relied on manual quantification, which is expensive and error-prone due to the imprecision. In this paper, we present LQPR, a highly efficient automatic approach for performance requ

Chaimaa Medjadji, Sadi Alawadi, Feras M. Awaysheh, Guilain Leduc

Federated Learning (FL) enables collaborative model training across decentralized clients while preserving data privacy by keeping raw data local. However, FL suffers from significant communication overhead due to the frequent exchange of high-dimensional model updates over constrained networks. In this paper, we present FedSparQ, a lightweight compression f

Alberto Caimo, Isabella Gollini

Understanding the structure of weighted signed networks is essential for analysing social systems in which relationships vary both in sign and strength. Despite significant advances in statistical network analysis, there is still a lack of statistical models that can jointly and rigorously account for both the sign and strength of relationships in networks.

G. S. Bisnovatyi-Kogan, E. A. Patraman

Stellar models are calculated in the approximation of a uniform density distribution. Variational method was used for determination of the boundary of a stability loss, for stellar masses in the range from 2 up to $10^5$ $M_{\odot}$. The effects of the general relativity had been taken into account. The equation of state in the temperature and density ranges

Tatiana Komarova, William Matcham

We analyze multivariate ordered discrete response models with a lattice structure, modeling decision makers who narrowly bracket choices across multiple dimensions. These models map latent continuous processes into discrete responses using functionally independent decision thresholds. In a semiparametric framework, we model latent processes as sums of covari

D. S. Bhattacharya, J. Bae, M. Bergevin, J. Boissevain

BUTTON-30 is a neutrino detector demonstrator located in the STFC Boulby underground facility in the north-east of England. The main goal of the project is to deploy and test the performance of the gadolinium-loaded water-based liquid scintillator for neutrino detection in an underground environment. This will pave the way for a future large-volume neutrino

Nikolai Herrmann, Marcella C. Zijta, Stefan Klein, Régine P. M. Steegers-Theunissen

Standardized alignment of the embryo in three-dimensional (3D) ultrasound images aids prenatal growth monitoring by facilitating standard plane detection, improving visualization of landmarks and accentuating differences between different scans. In this work, we propose an automated method for standardizing this alignment. Given a segmentation mask of the em

Xusheng Zhu, Farshad Rostami Ghadi, Tuo Wu, Kaitao Meng

Fluid antenna systems (FAS) offer a promising paradigm for enhancing wireless communication by exploiting spatial diversity, yet a rigorous analytical framework for their error probability has been notably absent. To this end, this paper addresses this critical gap by unveiling the \textbf{fundamental scaling laws} that govern the symbol error rate (SER) of

Mao-Jing Liu, Ying Li, Zhi-Tian Zou

We present a unified analysis of $\overline{B}^0 \to D^{(*)+}\ell^-\bar{\nu}_\ell$ and $\overline{B}^0 \to D^{(*)+}\pi^-$ decays using the perturbative QCD (PQCD) approach. The $B \to D^{(*)}$ transition form factors are calculated at low $q^2$ and extrapolated to the high-$q^2$ region using the latest lattice QCD results via a model-independent $z$-expansio

Yoojin Oh, Junhyug Noh

Class Activation Mapping (CAM) and its extensions have become indispensable tools for visualizing the evidence behind deep network predictions. However, by relying on a final softmax classifier, these methods suffer from two fundamental distortions: additive logit shifts that arbitrarily bias importance scores, and sign collapse that conflates excitatory and

Yiquan Yang, Xiaolong Hu, Wei Du, Shuhe Wu

Chiral discrimination of enantiomeric biomolecules is vital in chemistry, biology, and medicine. Conventional methods, relying on circularly polarized light, face weak chiroptical signals and potential photodamage. Despite extensive efforts to improve sensitivity under low-photon exposure, classical chiral probes remain fundamentally bounded by the shot-nois

Longpeng Qiu, Ting Li, Shuai Mao, Nan Yang

Input errors in question-answering (QA) systems often lead to incorrect responses. Large language models (LLMs) struggle with this task, frequently failing to interpret user intent (misinterpretation) or unnecessarily altering the original question's structure (over-correction). We propose QuestionRAG, a framework that tackles these problems. To address misi

Kseniia Mamaeva, Hodjat Haijan, Carolyn Elliott, Hannah Killeen

Achieving strong coupling between excitons of colloidal semiconductor quantum dots (QDs) and localized surface plasmon polaritons (LSPs) is critical for advanced room-temperature quantum emitter and sensing applications. A key challenge is to have precise control of the emitters position with respect to an individual plasmonic nanostructure. Here, we present

Canhui Wu, Qiong Cao, Chao Xue, Wei Xi

Recent advances in large language models (LLMs) have leveraged explicit Chain-of-Thought (CoT) prompting to improve reasoning accuracy. However, most existing methods primarily focus on compressing verbose reasoning outputs. These Long-to-Short transformations aim to improve efficiency, but require a large amount of short CoT data. In this work, we introduce

Célian Ringwald, Fabien Gandon, Catherine Faron, Franck Michel

Small language models (SLMs) have shown promises for relation extraction (RE) when extracting RDF triples guided by SHACL shapes focused on common datatype properties. This paper investigates how SLMs handle both datatype and object properties for a complete RDF graph extraction. We show that the key bottleneck is related to long-tail distribution of rare pr

Zsolt Baja, Tamás László, András Némethi

In this article, first we give two formulae for the delta invariant of a complex curve singularity that can be embedded as a ${\mathbb Q}$-Cartier divisor in a normal surface singularity with rational homology sphere link. Next, we consider representable numerical semigroups, they are semigroups associated with normal weighted homogeneous surface singulariti

Alexandros Vazaios, Jannis Brugger, Cedric Derstroff, Kristian Kersting

AlphaZero-like Monte Carlo Tree Search systems, originally introduced for two-player games, dynamically balance exploration and exploitation using neural network guidance. This combination makes them also suitable for classical search problems. However, the original method of training the network with simulation results is limited in sparse reward settings,

Qianhui Zhao, Li Zhang, Fang Liu, Junhang Cheng

In recent years, Large Language Models (LLMs) have achieved remarkable progress in automated code generation. In real-world software engineering, the growing demand for rapid iteration and continuous delivery underscores the importance of project-level code generation, where LLMs are expected to generate complete software projects directly from complex user

Kunrui Cao, Jingyu Chen, Panagiotis D. Diamantoulakis, Lei Zhou

Pinching antenna system (PAS) serves as a groundbreaking paradigm that enhances wireless communications by flexibly adjusting the position of pinching antenna (PA) and establishing a strong line-of-sight (LoS) link, thereby reducing the free-space path loss. This paper introduces the concept of wireless-powered PAS, and investigates the reliability of wirele

Minquan Gao, Xinyi Li, Qing Yan, Xiaojian Sun

Pre-trained robot policies serve as the foundation of many validated robotic systems, which encapsulate extensive embodied knowledge. However, they often lack the semantic awareness characteristic of foundation models, and replacing them entirely is impractical in many situations due to high costs and the loss of accumulated knowledge. To address this gap, w

Andrea Cremaschi, Manuele Leonelli, Gherardo Varando

We propose a fully Bayesian approach for causal inference with multivariate categorical data based on staged tree models, a class of probabilistic graphical models capable of representing asymmetric and context-specific dependencies. To account for uncertainty in both structure and parameters, we introduce a flexible family of prior distributions over staged

Zhonghao Liang, Chenlu Jia, Dongmei Huang, Qunying Liao

Maximum distance separable (in short, MDS), near MDS (in short, NMDS), and self-orthogonal codes play a pivotal role in algebraic coding theory, particularly in applications such as quantum communications and secret sharing scheme. Recently, the construction of non-generalized Reed-Solomon (in short, non-GRS) codes has emerged as a significant research front

Christopher Couzens, Alice Lüscher, James Sparks

We use equivariant localization and holography to study four-dimensional $\mathcal{N}=1$ superconformal field theories arising from M5-branes wrapped on a punctured Riemann surface. We explain how, given a Riemann surface with marked points, one can glue in a ``puncture geometry'' locally around each point. Using equivariant localization we show that the cen

Chuanyuan Ge, Shiping Liu

In this paper, we give a new proof of the Lemmens-Seidel conjecture on the maximum number of equiangular lines with a common angle $\arccos(1/5)$. This conjecture was previously resolved by Cao, Koolen, Lin, and Yu in 2022 through an analysis involving forbidden subgraphs for the smallest Seidel eigenvalue $-5$. Our new proof is based on bounds on eigenvalue

Rathin Chandra Shit, Sharmila Subudhi

The real-time performance, adversarial resiliency, and privacy preservation are the most important metrics that need to be balanced to practice collision avoidance in large-scale multi-UAV (Unmanned Aerial Vehicle) systems. Current frameworks tend to prescribe monolithic solutions that are not only prohibitively computationally complex with a scaling cost of

Stefan Franssen

We give a contributed discussion on "Model uncertainty and missing data: An Objective Bayesian Perspective", where we discuss frequentist perspectives on the proposed methodology.

Chiara Rucco, Motaz Saad, Antonella Longo

Traditional ETL and ELT design patterns struggle to meet modern requirements of scalability, governance, and real-time data processing. Hybrid approaches such as ETLT (Extract-Transform-Load-Transform) and ELTL (Extract-Load-Transform-Load) are already used in practice, but the literature lacks best practices and formal recognition of these approaches as des

Viktória Fröhlich, Zsolt Regály

Free-floating planets are thought to be numerous in the Galaxy and may retain their moons after ejection from their natal systems. If those satellites acquire or preserve orbital eccentricity, tidal dissipation could provide a long-lasting internal heat source, potentially creating urable environments (capable of enabling abiogenesis) in the absence of stell

João Victor Galvão da Mata, Anders Hansson, Martin S. Andersen

Maximum likelihood estimation is effective for identifying dynamical systems, but applying it to large networks becomes computationally prohibitive. This paper introduces a maximum likelihood estimation method that enables identification of sub-networks within complex interconnected systems without estimating the entire network. The key insight is that under

Yiyuan Wang

We propose a new algorithm for a broad class of periodic time-varying Stochastic Game-Theoretic Riccati Differential Equations arising in Zero-Sum Linear-Quadratic Stochastic Differential Games. The algorithm is constructed via dual-layer matrix-valued functions iteration sequences, which reformulate the original problem into a set of interconnected bilevel

Mihir Kumar Sahoo, Abhay Anand V S, Nihar Ranjan Sahoo, Anshuman Kumar

A plasmonic nanostructure-based substrate, serving as a surface-enhanced Raman scattering (SERS) substrate, enhances the Raman scattering of molecules. By employing an electron beam lithography followed by our recently developed nano-electroplating protocol, a gold nanorod array SERS substrate can be fabricated to detect lower molecular analyte concentration

Andrzej A. Zdziarski, Swadesh Chand, Gulab Dewangan, Ranjeev Misra

We study the transient black hole binary MAXI J1631--479 in its soft spectral state observed simultaneously by the NICER and NuSTAR instruments. Its puzzling feature is the presence of a strong and broad Fe K line, while the continuum consists of a strong disk blackbody and a very weak power-law tail. The irradiation of the disk by a power-law spectrum fitti

Tal Gottesman, Viktória Klász, Markus Kleinau, Rene Marczinzik

In a recent article, Iyama and Marczinzik showed that a lattice is distributive if and only if the incidence algebra is Auslander regular, giving a new connection between homological algebra and lattice theory. In this article we study when a distributive lattice has a pure minimal injective coresolution, a notion first introduced and studied in a work of Aj

Peter Markoš, Richard Hlubina

It is generally accepted that the most fundamental property of a superconductor is that it exhibits the Meissner effect. Of similar importance is the Becker-London effect, i.e. generation of magnetic field inside a rotating superconductor. Hirsch has recently pointed out that, within the conventional theory of superconductivity, the question about how these

Saumitra Yadav, Manish Shrivastava

Existing Machine Translation (MT) research often suggests a single, fixed set of hyperparameters for word segmentation models, symmetric Byte Pair Encoding (BPE), which applies the same number of merge operations (NMO) to train tokenizers for both source and target languages. However, we demonstrate that this uniform approach doesn't guarantee optimal MT per

Hari Mohan Pandey, Anshul Gupta, Subham Sarkar, Minakshi Tomer

Text-to-SQL systems enable users to interact with structured databases using natural language, eliminating the need for specialized programming knowledge. In this work, we introduce GEMMA-SQL, a lightweight and efficient text-to-SQL model built upon the open-source Gemma 2B architecture. Unlike many large language models (LLMs), GEMMA-SQL is fine-tuned in a

Zekai Qu, Yinxu Pan, Ao Sun, Chaojun Xiao

Reinforcement learning (RL) post-training has become a trending paradigm for enhancing the capabilities of large language models (LLMs). Most existing RL systems for LLMs operate in a fully synchronous manner, where training must wait for the rollout of an entire batch to complete. This design leads to severe inefficiencies, as extremely long trajectories ca

Pan He, Min Zhang, Yue-Xin Huang, Jingru Li

The nonlinear Hall effect (NHE) can enable rectification and energy harvesting, and its control by external fields, including gate, strain and magnetic field, has been pursued intensively. However, existing tuning pathways rely predominantly on fully quantum mechanical effects and are typically inefficient, resulting in weak NHE signals that limit further pr

Karyn Le Hur

Through the two-channel Kondo model, I address a magnetic spin-1/2 impurity interacting with a bound state of spin origin at the edge in a Luther-Emery liquid showing a spin gap in the bulk. The system presents two zero-energy Majorana fermions of magnetic origin, one on the impurity site and one at the edge. I derive the wavefunction at the edge for the pro

Samarth Toolhally, Joeri Roelofs, Siep Weiland, Amritam Das

In inkjet printing, optimal paper moisture is crucial for print quality, achieved through hot-air impingement in the fixation unit. This paper presents a modular digital twin of the fixation unit, modeling the thermo-fluidic drying process and monitoring its spatio-temporal performance. The novel approach formulates the digital twin as an infinite-dimensiona

Yangliu Fan, Kilian Buehling, Volker Stocker

Despite the importance of social science knowledge for various stakeholders, measuring its diffusion into different domains remains a challenge. This study uses a novel text-based approach to measure the idea-level diffusion of social science knowledge from the research domain to the journalism and policy-making domains. By doing so, we expand the detection

Jonas Cederlöf, Sara Roman

We study how job search behavior evolves over the unemployment spell and the extent to which job seekers experience duration dependence in callbacks. Leveraging data on 2.4 million monthly activity reports containing detailed information on job applications, interviews, and other search activities, we separate within-spell changes from dynamic selection with

Syed Muqeem Mahmood, Hassan Mohy-ud-Din

We present a framework that combines Large Language Models with computational image analytics for non-invasive, zero-shot prediction of IDH mutation status in brain gliomas. For each subject, coregistered multi-parametric MRI scans and multi-class tumor segmentation maps were processed to extract interpretable semantic (visual) attributes and quantitative fe

Mengyao Guo, Kexin Nie, Ze Gao, Black Sun

Creating meaningful visual narratives through human-AI collaboration requires understanding how text-image intertextuality emerges when textual intentions meet AI-generated visuals. We conducted a three-phase qualitative study with 15 participants using GPT-4o to investigate how novices navigate sequential visual narratives. Our findings show that users deve

Darryl Z. Seligman, Dušan Marčeta, Eloy Peña-Asensio

In this paper we calculate the expected orbital elements, radiants, and velocities of Earth-impacting interstellar objects. We generate a synthetic population of $\sim10^{10}$ interstellar objects with M-star kinematics in order to obtain $\sim10^4$ Earth-impactors. The relative flux of impactors arriving from the direction of the solar apex and the galactic

Marine De Clerck, Oleg Evnin

G\'erard and Grellier proposed, under the name of the cubic Szeg\H{o} equation, a remarkable classical field theory on a circle with a quartic Hamiltonian. The Lax integrability structure that emerges from their definition is so constraining that it allows for writing down an explicit general solution for prescribed initial data, and at the same time, the dy

Shenghao Li

For complex logical data augmentation, heavy reliance on human annotation is costly, whereas direct generation with large language models yields uninterpretable and logically homogeneous examples. To address this, we present LFC-DA, a symbolic-logic-controlled pipeline: logical text is first mapped to propositional expressions, a compact rule library is comp

Ivan Rubtsov, Ivan Dudakov, Yuri Kuratov, Vadim Korolev

Crystal graph neural networks are widely applicable in modeling experimentally synthesized compounds and hypothetical materials with unknown synthesizability. In contrast, structure-agnostic predictive algorithms allow exploring previously inaccessible domains of chemical space. Here we present a universal approach for enhancing composition-based materials p

Sébastien Aumaître, Nicolas Mujica

Forced granular matter in confined geometries presents phase transitions and coexistence. Depending on the system and forcing parameters, liquid-vapor and liquid-solid co-existing states are possible. For the solid-liquid coexistence that is observed in quasi-two-dimensional vibrated systems, both first- and second-order transitions have been reported. Exper

Yunbo Long, Yuhan Liu, Alexandra Brintrup

The deployment of large language models (LLMs) in automated negotiation has set a high performance benchmark, but their computational cost and data privacy requirements render them unsuitable for many privacy-sensitive, on-device applications such as mobile assistants, embodied AI agents or private client interactions. While small language models (SLMs) offe

Rom Himelstein, Amit LeVi, Brit Youngmann, Yaniv Nemcovsky

Safety-aligned large language models (LLMs) are becoming increasingly widespread, especially in sensitive applications where fairness is essential and biased outputs can cause significant harm. However, evaluating the fairness of models is a complex challenge, and approaches that do so typically utilize standard question-answer (QA) styled schemes. Such meth

Hongrun Ren, Yun Xiong, Lei You, Yingying Wang

The rise of the machine learning (ML) model economy has intertwined markets for training datasets and pre-trained models. However, most pricing approaches still separate data and model transactions or rely on broker-centric pipelines that favor one side. Recent studies of data markets with externalities capture buyer interactions but do not yield a simultane

Gahyeon Kim, Sohee Kim, Seokju Lee

Recent advances in large-scale vision and language models have led to significant progress in zero-shot learning tasks. Methods such as CoOp and CoCoOp have shown that replacing handcrafted prompts with learnable vectors, known as prompt learning, can result in improved performance. However, these models often struggle to generalize to entirely unseen catego

Kapila W. S. Palitharathna, Constantinos Psomas, Ioannis Krikidis

In this paper, we propose a lightwave power transfer-enabled underwater optical integrated sensing and communication (O-ISAC) system, where an access point (AP) mounted on a seasurface ship transmits lightwave signals to two nodes, namely ($i$) a seabed sensor that harvests energy and transmits uplink information to the AP, and ($ii$) a sensing target whose

Gabriela Fernandes

Ovarian cancer remains one of the most lethal gynecological malignancies, largely due to late diagnosis and extensive heterogeneity across subtypes. Current diagnostic methods are limited in their ability to reveal underlying genomic variations essential for precision oncology. This study introduces a novel hybrid deep learning pipeline that integrates quant

O. O. Horchakov, A. Yu. Shvets

The realization of novel scenario involving transitions between different types of chaotic attractors is investigated for the Rossler system. Characteristic features indicative of the presence of generalized intermittency scenario in this system are identified. The properties of "chaos-chaos" transitions following the generalized intermittency scenario are a

Xiaocai Zhang, Hur Lim, Ke Wang, Zhe Xiao

In this study, a modular, data-free pipeline for multi-label intention recognition is proposed for agentic AI applications in transportation. Unlike traditional intent recognition systems that depend on large, annotated corpora and often struggle with fine-grained, multi-label discrimination, our approach eliminates the need for costly data collection while

Hiroki Ono, Yusuke Doi, Akihiro Nakatani

We propose a novel type of umklapp-free lattice (UFL), where umklapp processes are completely absent. The proposed UFL incorporates cubic long-range nonlinearity, a feature not addressed in previous studies. In this paper, we derive an analytical expression for the cubic nonlinear coupling constants by imposing mathematical conditions such that the nonlinear

Gabriel Pirlogeanu, Alexandru-Lucian Georgescu, Horia Cucu

In this work, we present a new state-of-the-art Romanian Automatic Speech Recognition (ASR) system based on NVIDIA's FastConformer architecture--explored here for the first time in the context of Romanian. We train our model on a large corpus of, mostly, weakly supervised transcriptions, totaling over 2,600 hours of speech. Leveraging a hybrid decoder with b

Gianluca Crippa

In these lecture notes, we provide an introduction to the theory of mixing for incompressible flows from a PDE perspective. We discuss both the Lagrangian (ODE) and Eulerian (PDE, continuity equation) viewpoints, and introduce suitable notions of mixing scales that quantify the degree to which a scalar field transported by a velocity field becomes mixed. We

Alexander Alecio

We consider a model for systemic risk comprising of a system of diffusion processes, interacting through their empirical mean. Each process is subject to a confining double-well potential with some uncertainty in the coefficients, corresponding to fluctuations in height of the potential barrier seperating the two wells. This is equivalent to studying a singl

Koichi Nagasaki, Matsuo Sato

String geometry theory is one of the candidates of the non-perturbative formulation of superstring theory. In this paper, in string geometry theory, we identify perturbative heterotic vacua, which include general heterotic backgrounds. From fluctuations around these vacua, we derive the path-integrals of heterotic perturbative superstrings on the backgrounds

Colin Rogers, Adriana C. Briozzo

A novel extension of the canonical solitonic mKdV equation is introduced which admits hybrid Ermakov-Painlev\'e II symmetry reduction. Application of the latter is made to obtain exact solution of Airy-type to a class of moving boundary problems of Stefan kind for this extended mKdV equation. A reciprocal transformation is then applied to the latter to gener

Moran Yanuka, Paul Dixon, Eyal Finkelshtein, Daniel Rotman

Speculative decoding accelerates autoregressive speech generation by letting a fast draft model propose tokens that a larger target model verifies. However, for speech LLMs that generate acoustic tokens, exact token matching is overly restrictive: many discrete tokens are acoustically or semantically interchangeable, reducing acceptance rates and limiting sp

Guangyao Wang, Yue Wu, Yulin Pan, Kayhan Momeni

Consistency relations of internal gravity waves (IGWs) describe ratios of cross-spectral quantities as functions of frequency. It has been a common practice to evaluate the measured or simulated signals (e.g., time series of velocity, density, etc.) against the consistency relations, as a way to determine whether an oceanic field of interest is comprised of

Thomas Leblé

We prove that the hexagonal lattice is a local minimizer, among all point configurations, of the interaction energy per unit volume for pair potentials that are completely monotonic functions of the square distance. This includes Gaussian interactions and power laws.

Lorena Ballesteros Ferraz

This work demonstrates that repeated weak measurements together with post-selection can produce sharp dynamical discontinuities in meter observables, even in minimal quantum systems. The discontinuous behavior is governed by the polar angle of the post selected state, which serves as a continuous control parameter. As this angle is varied, the expectation va

Saba Latif, Fajar J. Ekaputra, Maxim Vidgof, Sabrina Kirrane

The Object-Centric Event Data (OCED) is a novel meta-model aimed at providing a common ground for process data records centered around events and objects. One of its objectives is to foster interoperability and process information exchange. In this context, the integration of data from different providers, the combination of multiple processes, and the enhan

Lukas Schweiger, Florian Spieckermann, Michael Burtscher, Stefan Wurster

Metal hydrides remain an intriguing alternative to conventional gaseous and liquid hydrogen storage methods, offering high volumetric storage density and enhanced hydrogen storage safety at ambient conditions. In this regard, the intermetallic compound FeTi is one of the most promising storage materials. However, its widespread industrial application remains

Changxi Zhu, Mehdi Dastani, Shihan Wang

In multi-agent deep reinforcement learning (MADRL), agents can communicate with one another to perform a task in a coordinated manner. When multiple tasks are involved, agents can also leverage knowledge from one task to improve learning in other tasks. In this paper, we propose Multi-task Communication Skills (MCS), a MADRL with communication method that le

Mario Ayala, Nicolas Dirr, Grigorios A. Pavliotis, Johannes Zimmer

We study the interplay between reversibility, geometry, and the choice of multiplicative noise (in particular It\^{o}, Stratonovich, Klimontovich) in stochastic differential equations (SDEs). Building on a unified geometric framework, we derive algebraic conditions under which a diffusion process is reversible with respect to a Gibbs measure on a Riemannian

Antonio Carvajal-Rodríguez

Living systems can be understood as organized entities that capture, transform, and reproduce information. Classical gene-centered models explain adaptation through frequency changes driven by differential fitness, yet they often overlook the higher-order organization and causal closure that characterize living systems. Here we revisit several evolutionary f

Sander Borst, Danish Kashaev

We study the online load balancing problem on unrelated machines, with the objective of minimizing the square of the $\ell_2$ norm of the loads on the machines. The greedy algorithm of Awerbuch et al. (STOC'95) is optimal for deterministic algorithms and achieves a competitive ratio of $3 + 2 \sqrt{2} \approx 5.828$, and an improved $5$-competitive randomize

Elif Arslan, Jacobus G. M. van der Linden, Serge Hoogendoorn, Marco Rinaldi

Sparse decision tree learning provides accurate and interpretable predictive models that are ideal for high-stakes applications by finding the single most accurate tree within a (soft) size limit. Rather than relying on a single "best" tree, Rashomon sets-trees with similar performance but varying structures-can be used to enhance variable importance analysi

Yundi Zhang, Daniel Rueckert, Jiazhen Pan

Motivation: CMR is the golden standard for cardiac diagnosis, and medical data annotation is time-consuming. Thus, screening techniques from unlabeled data can help streamline the cardiac diagnosis process. Goal: This work aims to enable cardiac function assessment from unlabeled cardiac MR images using an unsupervised approach with masked image modeling. Ap

Jannis Brugger, Viktor Pfanschilling, David Richter, Mira Mezini

Neural-guided equation discovery systems use a data set as prompt and predict an equation that describes the data set without extensive search. However, if the equation does not meet the user's expectations, there are few options for getting other equation suggestions without intensive work with the system. To fill this gap, we propose Residuals for Equation

Daniel Walsken, Pavel Petrov, Matthias Ehrhardt

In this study, a Fourier-based, split-step Pad\'e (SSP) method for solving the parabolic wave equation with applications in guided wave propagation in ocean acoustics is presented. Traditional SSP implementations rely in finite-difference discretizations of the depth-dependent differential operator. This approach limits accuracy in coarse discretizations as

Umut Çalıkyılmaz, Nitin Nayak, Jinghua Groppe, Sven Groppe

In recent years, the research of multi-agent systems has taken a direction to explore larger and more complex models to fulfill sophisticated tasks. We point out two possible pitfalls that might be caused by increasing complexity; susceptibilities to faults, and performance bottlenecks. To prevent the former threat, we propose a transaction-based framework t