April 2026 arXiv papers — page 103

Malvika Srivastava, Claudia Bank, Joachim Krug, Suman G. Das

Fitness landscapes are mappings between genotypes, phenotypes, and fitness that shape evolution. In recent years, empirical work and theoretical models have greatly advanced our understanding of how populations navigate rugged fitness landscapes. Here, we provide a timely review of this field. Its rapidly growing literature employs a wide range of terms, whi

Sergio Davis

Among the statistical mechanical frameworks able to describe systems in non-equilibrium steady states such as collisionless plasmas, self-gravitating systems and other complex systems, superstatistics have gained recent attention. Superstatistics postulates a superposition of canonical systems with inverse temperatures $\beta$ described by a probability dist

Tahir Cetin Akinci, Gokhan Gokmen, Alfredo A. Martinez-Morales

This study presents a physically informed hybrid time-frequency and machine learning (STFT-ML) framework for arc stability monitoring in electric arc welding systems. The primary current signal is modeled as a stochastic representation of plasma dynamics and transformed into a structured feature space using localized spectral energy distributions. Within thi

Mayank Chhaperwal, Amartyaraj Kumar, Kausik Majumdar

Single photon emitters (SPEs) are integral parts of several quantum technology implementations. Over the past decade or so, monolayers of transition metal dichalcogenides (TMDCs) have emerged as one of the promising candidates for SPE platforms with attractive characteristics. To move ahead, it is necessary to understand the atomistic origin of SPEs in TMDCs

Haoran Peng, Tong Wu, Hang Liu, Weijia Zheng

Ensuring strict safety guarantees is the paramount challenge for emerging 5G/6G wireless systems, particularly as they increasingly govern mission-critical applications ranging from autonomous UAV swarms to industrial automation. While deep reinforcement learning (DRL) offers a promising solution for complex resource allocation, standard algorithms frequentl

Pierre Beckmann, Patrick Butlin

The individuation problem for large language models asks which entities associated with them, if any, should be identified as minds. We approach this problem through mechanistic interpretability, engaging in particular with recent empirical work on persona vectors, persona space, and emergent misalignment. We argue that three views are the strongest candidat

Shaowen Wan, Yanjun Lv, Lu Zhang, Dajiang Zhu

Neurobiological and neurodegenerative diseases are inherently multifactorial, arising from coupled influences spanning genetic susceptibility, brain alterations, and environmental and behavioral factors. Multimodal modeling has therefore been increasingly adopted for disease diagnosis by integrating complementary evidence across data sources. However, in bot

Owais Ahmad, Jasifa Fayaz

In this article, we introduce the notion of Quaternion Boostlet Transform (QBT), a hypercomplex framework designed to unify the analysis of multi-component wavefields by merging the algebraic richness of quaternions with the relativistic, hyperbolic geometry of the boostlet system. By treating coupled physical phenomena such as acoustic pressure with particl

Lin Zhao, Qiaohui Gao, Elizabeth Martin, Kurt P. Schulz

Binge eating disorder (BED) is the most prevalent eating disorder. However, current diagnostic frameworks remain largely grounded in symptom-based criteria rather than underlying biological mechanisms, thereby limiting early detection and the development of biologically-informed interventions. Emerging studies have begun to investigate the neurobiological si

Victor Schmitt, Farzaneh Pourahmadi, Angela Flores-Quiroz, Pablo Apablaza

Transmission expansion planning (TEP) plays a critical role in ensuring power system reliability and facilitating the integration of renewable energy resources. However, this process requires planners to constantly deal with significant uncertainty. While multistage stochastic TEP models provide a robust framework for identifying investment plans under uncer

Tianbao Zhang

Large Language Models produce a controllability gap in safety-critical engineering: even low rates of undetected constraint violations render a system undeployable. Current orchestration paradigms suffer from sycophantic compliance, context attention decay, and stochastic oscillation during self-correction. We introduce the Convergent AI Agent Framework (CAA

Mingxi Pang, Dingheng Wang, Zekun Li, Zhenping Sun

Query-based 3D object detection methods using multi-view images often struggle to efficiently leverage dynamic multi-scale information, e.g., the relationship between the object features and the geometric of the queries are not sufficiently learned, directly exploring the multi-scale spatiotemporal features will pay too many costs. To address these challenge

Wei Roy Hua

For four decades, the QWERTY keyboard organized white-collar knowledge work. Typing's dominance was instrumental, not cognitively necessary. As multimodal AI achieves human-parity understanding of speech and gesture, this necessity dissolves. We introduce instrumental dissolution -- loss of institutional-default status while persisting in specialist niches.

Nisrine Rair, Alban Goupil, Valeriu Vrabie, Emmanuel Chochoy

Subjective NLP datasets typically aggregate annotator judgments into a single gold label, making it difficult to diagnose whether disagreement reflects unclear criteria, collapsed distinctions, or legitimate plurality. We propose a \emph{schema-level diagnostic} for auditing expert-designed annotation schemas \emph{prior to} gold-label commitment, using only

Weicheng Wang, Zhicheng Zhang, Zhongqi Zhang, Juncheng Zhou

Video editing aims to modify input videos according to user intent. Recently, end-to-end training methods have garnered widespread attention, constructing paired video editing data through video generation or editing models. However, compared to image editing, the high annotation costs of video data severely constrain the scale, quality, and task diversity o

Huije Lee, Jisu Shin, Hoyun Song, Changgeon Ko

Static benchmarks for harmful content detection face limitations in scalability and diversity, and may also be affected by contamination from web-scale pre-training corpora. To address these issues, we propose a framework for synthesizing harmful content, leveraging persona-guided large language model (LLM) agents. Our approach constructs two-dimensional use

Sukai Huang, Chenyuan Zhang, Fucai Ke, Zhixi Cai

Instruction granularity is an important yet poorly controlled variable in language-guided embodied AI. Existing benchmarks typically pair each task with a single static instruction, making it difficult to study how agent behavior changes when the same task is described at different levels of detail. We introduce Mini-BEHAVIOR-Gran, a new benchmark for contro

Alen Andrašek

A rational Diophantine $m$-tuple is a set $\{a_1,\ldots,a_m\}$ of distinct nonzero rational numbers such that $a_i a_j+1$ is a square for all $1\leq i < j\leq m$. Similarly, we may ask when $a_ia_j+1$ is a $k$-th power. Here, we study the case $k=4$ and produce some non-trivial infinite families of such triples. We show that there are infinitely many triples

Khomesh R. Patle, G. P. Singh

We present the cosmic expansion scenario in the framework of $f(T)$ gravity by employing a dark energy equation of state (EoS) parameter. Specifically, we proceed with the power-law form of the function $f(T) = \alpha$$(-T)^{n}$, in conjunction with the Gong-Zhang parametrization of the dark energy EoS. We derive the expansion rate in terms of the redshift f

Manpreet Singh, Siming Liu, N. S. Saini

Kinetic Alfven waves (KAWs) are fundamental to energy transport and small-scale structure formation in the turbulent, magnetized interstellar medium (ISM). While first-order Korteweg--de Vries (KdV) models describe weakly nonlinear KAW solitons, they fail in strongly inhomogeneous environments where higher-order effects become significant. We investigate hig

Xiaolei Ren

Prior work has demonstrated that functionally correct yet vulnerable outputs arise systematically in threat-oriented settings, where adversarial or implicit channels are used to induce security failures in code agents and automated patching workflows. This note introduces a complementary but distinct framing: False Security Confidence (FSC), which studies th

Jidong Kuang, Hongsong Wang, Jie Gui, Yuan Yan Tang

Skeleton data used for action recognition are acquired from a wide range of sources, including depth sensors, marker-based motion capture systems, and 2D/3D pose estimators. These sources yield skeletons that differ in joint number, skeletal topology, and coordinate dimensionality, making skeleton data inherently heterogeneous. However, previous works overlo

Oluwafolajimi Samuel Bolusteve, Linhan Fang, Xingpeng Li

Renewable energy adoption has increased significantly in recent years. However, the growing penetration of renewable generation that depends on the weather introduces additional variability, making accurate net load forecasting increasingly challenging. In this study, long short-term memory (LSTM) and fully connected neural network (FCNN) models are employed

David Dolžan, Bogdana Oliynyk

In this paper, we investigate structural properties of the Cayley graph of a quandle and describe this graph for several important classes of quandles, including conjugation, Takasaki, dihedral, and Alexander quandles. In particular, we prove that for an Alexander quandle $A_t(G)$ over a finite abelian group $G$, the connected components of the Cayley graph

Antonio Valerio Miceli Barone, Poon Tsz Nok

We introduce a self-play framework for semantic equivalence in Haskell, utilizing formal verification to guide adversarial training between a generator and an evaluator. The framework leverages Liquid Haskell proofs for validating equivalence and execution-based counterexamples for inequivalence, organized via a difficulty-aware curriculum. To facilitate thi

Wenzhen Yuan, Wutao Xiong, Fanchen Yu, Shengji Tang

Multi-agent systems (MAS) demonstrate clear advantages in tackling complex problems by coordinating diverse agents and external tools. However, most existing orchestration methods rely on static workflows or serial agent scheduling, and are further constrained by heterogeneous interface protocols between tools and agents. This leads to high system complexity

Yuxuan Ouyang, yingfeng luo, JingBo Zhu, Tong Xiao

Large Language Models (LLMs) are increasingly used to generate narrative content, including children's stories, which play an important role in social and cultural learning. Despite growing interest in AI safety and alignment, most existing evaluations focus primarily on English, leaving the cross-lingual generalization of aligned behavior underexplored. In

Sotiria Chatzimarkou, Panagiotis Dimakis

Inspired by Gaiotto's conformal limit construction for Higgs bundles we define and study a conformal limit construction for Nakajima quiver varieties. We prove that the conformal limit is indeed a limit of a one parameter family of points inside a specified quiver variety and that it gives a biholomorphic map between holomorphic Lagrangian submanifolds folia

Xinran Liu, Diptesh Kanojia, Wenwu Wang, Zhenhua Feng

Existing music-driven dance generation approaches have achieved strong realism and effective audio-motion alignment. However, they generally lack semantic controllability, making it difficult to guide specific movements through natural language descriptions. This limitation primarily stems from the absence of large-scale datasets that jointly align music, te

Juanda Kelana Putra, Richard Smolka

This paper reveals a categorical equivalence connecting two distinct quantum logic structures. The first is the orthomodular lattice, an algebraic system designed to formalize the properties of quantum systems. The second is a finitary orthomodular dynamic algebra, a specialized development of the orthomodular dynamic algebra where the underlying quantum act

Zhijun Zheng, Tian Qiu, Yuheng Zhao, Siming Chen

In visual analytics, applying filters to drill-down and extract higher-value insights is a common and important data analysis method. When the drill-down space becomes excessively large, analysts may lose orientation, leading to decreased efficiency in the drill-down process. To tackle these challenges, we propose the Intelligent Drill-Down Framework, in whi

Wei Li, Yuyang Li, Kaile Du, Yi Yu

The recently established Convolution Nuclear Norm Minimization (CNNM) addresses the problem of \textit{tensor completion with arbitrary sampling} (TCAS), which involves restoring a tensor from a subset of its entries sampled in an arbitrary manner. Despite its promising performance, the optimization procedure of CNNM needs performing Singular Value Decomposi

Estêvão V. B. de Oliveira, Muhammad Shaeer Moeed, Pierre-Nicholas Roy

In this work, we develop a theoretical description of the collective behavior of interacting dipolar planar rotors by using time independent perturbation theory and a small angle quadratic approximation. The ground state properties for both the ordered and disordered quantum phases of the system are directly calculated and analyzed. Time-independent perturba

Agissilaos Athanassoulis

The Alber equation is the mixed-state nonlinear Schr\"odinger equation with singular ($\delta$-interaction) kernel. It is used in the modeling of stochastic ocean waves, where it appears with the focusing sign in the nonlinearity, on $d=1.$ The main result of the paper is global well-posedness for self-adjoint, non-negative data in the Schatten-Sobolev space

Kun Ratha Kean, Yoritaka Iwata

Accurate determination of nuclear fission barrier heights is essential for understanding nuclear stability, fission dynamics, and nucleosynthesis. However, theoretical models such as the Extended Thomas-Fermi plus Strutinsky Integral (ETFSI) approach and the macroscopic-microscopic calculations of M\"oller et al. exhibit systematic deviations from experiment

Yifu Huo, Chenglong Wang, Ziming Zhu, Shunjie Xing

Reinforcement learning (RL) has emerged as a promising paradigm for training reasoning-oriented models by leveraging rule-based reward signals. However, RL training typically tends to improve single-sample success rates (i.e., Pass@1) while offering limited exploration of diverse reasoning trajectories, which is crucial for multi-sample performance (i.e., Pa

Pawan Kumar, Sobiya Ashraf, Naveen Tiwari, Dipin Pillai

Microorganisms navigating through narrow spaces encounter significant hydrodynamic challenges. To overcome these constraints and sustain efficient motion, they employ adaptive strategies, including adaptive oscillatory body deformations. While artificial microdroplets can traverse channels narrower than their diameter, studies of their locomotion have thus f

Xiaoyan Dai

This paper develops a semidefinite-programming-based method for online feedback control of nonlinear systems using a state-dependent representation. We formulate sequences of time-varying SDPs whose optimal solutions jointly yield a stabilizing feedback controller and a Lyapunov certificate satisfying stability conditions and quadratic performance specificat

Ganga Singh Manchanda

The admissibility of a gauge-fixing is governed by the invertibility of $\Delta=\{\sigma^a,\gamma_b\}$ where $\sigma^a$ are gauge-fixing conditions and $\gamma_b$ are independent first-class constraints. We prove, via the Schur complement, that the determinant of the combined constraint matrix $\mathcal{M}=\{\Phi_A, \Phi_B\}$ built from all constraints and g

Martin Balko, Jan Grebík, Pavel Hubáček, Martin Koutecký

We report new results on eight problems in mathematics and theoretical computer science, produced with the assistance of Bolzano, an open-source multi-agent LLM system. Bolzano orchestrates rounds of interaction between parallel prover agents and a verifier agent while maintaining a persistent knowledge base that is carried across rounds. Classified using th

Carson Dudley, Yutong Bi, Xiaofeng Liu, Samet Oymak

Non-stationary sequences arise naturally in control, forecasting, and decision-making. The data-generating process shifts at unknown times, and models must detect the change, discard or downweight obsolete evidence, and adapt to new dynamics on the fly. Transformer-based foundation models increasingly rely on in-context learning for time series forecasting,

Vittal Mirji

Schema drift in data pipelines is often caught only when a job touches real data. Typed-Dataset layers close part of this gap but require wholesale adoption; table-level enforcement systems close another part but operate at write time against a stored schema. We present a small Scala 3 framework that occupies the seam: it proves producer-to-contract structur

Ignasi Sole

The choice of visualisation in empirical performance analysis is not a neutral presentation decision but an analytical one: different graphical forms reveal different features of the same dataset, and reliance on any single type systematically conceals what the others expose. This paper presents and argues for a suite of five complementary visualisation tool

Sundaresan Jayanthi, Adonis Lupulescu, Julia Grinshtein, Lucio Frydman

Recently, Pang et al reported a novel polarization transfer scheme applicable to three-spin systems, whereby a rotating-frame NMR analogue of the cross effect could transfer polarization between; e.g., two 13Cs and an 15N in a single crystal. The present work furthers this scheme to the case of powder NMR under magic angle spinning (MAS) conditions, as well

Yiting Wang, Nolwenn Peyratout, Tim Brodermann, Jiahui Wang

This paper presents the report of the URVIS 2026 challenge on adverse-to-extreme panoptic segmentation. As the first challenge of its kind, it attracted 17 registered participants and 47 submissions, with 4 teams reaching the final phase. The challenge is based on the MUSES dataset, a multi-sensor benchmark for panoptic segmentation in adverse-to-extreme wea

Enwei Tong, Yao Zhu, Yuanchao Bai, Kai Wang

Large Language Models have revolutionized natural language processing, achieving unprecedented success across a vast range of tasks. However, their practical application in long-context scenarios is severely hampered by the formidable memory footprint of the Key-Value cache. While channel pruning has emerged as a promising compression strategy, existing meth

Adela Bâra, Simona-Vasilica Oprea

Current knowledge graph (KG) construction methods are confirmatory, focusing on recovering known relationships rather than identifying novel or context-dependent nodes. This paper proposes a phenotype-driven and evidence-governed framework that shifts the paradigm toward structured hypothesis discovery and controlled KG expansion. The approach integrates gra

Paul A. Constable, Dorothy A. Thompson, Irene O. Lee, Lynne Loh

The LEOPs (Light-ERG-Oscillatory Potentials) dataset provides light-adapted (LA) electroretinogram (ERG) and Oscillatory Potentials (OPs) waveforms for typically developing Control, Autism Spectrum Disorder (ASD) and ASD + Attention Deficit Hyperactivity Disorder (ADHD) childhood and adolescent populations. The ERGs were recorded in the Right And Left eyes w

Bruce A. Bassett, Amy Rouillard, Sitwala Mundia, Michael Cameron Gramanie

Background: Large language models (LLMs) are increasingly proposed for diagnostic support, but few evaluations use real-world multimodal inpatient data, particularly in low and middle-income country (LMIC) public hospitals. Methods: We conducted VALID, a retrospective evaluation of 539 multimodal inpatient cases from a tertiary public hospital in South Afric

Biao Wu, Yiwu Zhong, Meng Fang, Ling Chen

High-quality and diverse multimodal data are essential for improving vision-language models (VLMs), yet existing datasets often contain noisy, redundant, and poorly aligned samples. To address these problems, data filtering is commonly used to enhance the efficiency and performance of multimodal learning, but it introduces extra computational cost because fi

Manjul Bhargava, Arul Shankar, Xiaoheng Wang

We develop geometry-of-numbers methods to count orbits in coregular vector spaces having bounded invariants over any global field. We apply these techniques to bound the average ranks and determine average Selmer group sizes of elliptic curves and Jacobians of hyperelliptic curves over any base global field $F$ of characteristic not $2$, $3$ or $5$.

Dibakar Roychowdhury

We compute the rate of growth of operator size in matrix models by probing the Lin-Maldacena class of geometries with classical probes. We consider massive point particle probes whose proper momentum equals the size of the gauge invariant operator in the matrix model. We work out the example of the BMN Plane Wave Matrix Model using the electrostatic approach

Zhiyong Su, Jincan Wu, Yonghui Liu, Zheng Li

Point cloud denoising is a fundamental and crucial challenge in real-world point cloud applications. Existing quantitative evaluation metrics for point cloud denoising methods are implemented in a supervised manner, which requires both the denoised point cloud and the corresponding ground-truth clean point cloud to compute a representative geometric distance

Tetiana Obikhod

We present a review of supersymmetry, supergravity, and the non-perturbative dynamics of gauge theories, tracing a path from the supersymmetry algebra to moduli stabilisation and de~Sitter vacua in string theory. Representations of the supersymmetry algebra, the superspace formalism, and basic models including the Wess--Zumino model and $\mathcal{N}=1$ super

Yahya Saleh

Recent work has shown that parameterizing and optimizing coordinate transformations using normalizing flows, i.e., invertible neural networks, can significantly accelerate the convergence of spectral approximations. We present the first error estimates for approximating functions using Hermite expansions composed with adaptive coordinate transformations. Our

Panagiotis E. Theodorakis, Andrey Milchev

Applications such as digital microfluidics and bio-diagnostics rely on droplet locomotion. A prominent example of such motion is durotaxis, a phenomenon that requires a stiffness gradient along a surface for the transport of liquids, cells, or other nano-objects. Using surfaces with varying properties in specific directions can be exploited as a universal co

Yasushi Kawase, Warut Suksompong, Hanna Sumita, Yu Yokoi

In random assignment, fairness is often captured by stochastic-dominance envy-freeness (SD-EF). We observe that assignments satisfying SD-EF may admit decompositions that result in each agent envying another agent with high probability. To address this, we introduce decomposition envy-freeness (Dec-EF), which is a property of a decomposition rather than of a

Zhaokang Liao, Yingguo Gao, Yi Yang, Yongheng Hu

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a promising approach to improve the reasoning abilities of Large Language Models (LLMs). Among RLVR algorithms, Group Relative Policy Optimization (GRPO) and its variants have demonstrated strong performance and high training efficiency. However, GRPO-style objectives exhibit two issues on

Chetan Abhijnanam Bora, Badam Singh Kushvah, Kanak Saha

Long-term integrations of asteroid orbits with high-accuracy numerical integrators are essential for understanding dynamical evolution and ejection from the Solar System, but are computationally expensive. Here, we investigate the dynamical behaviour of asteroids and explore machine-learning (ML) and deep-learning (DL) approaches as efficient, scalable alter

Randall Ali, Thomas Dietzen, Matteo Scerbo, Enzo De Sena

We introduce a new framework for room acoustics modelling based on a state-space model of the boundary integral equation representing the sound field in a room. Whereas state-space models of linear time-invariant systems are traditionally constructed by means of a state vector and a 4-tuple of system matrices, the state-space representation introduced in thi

Joseph L. Garrett, P. S. Vishnu, Pauliina Salmi, Daniela Lupu

Unmixing reveals the spatial distribution and spectral details of different constituents, called endmembers, in a hyperspectral image. Because unmixing has limited ground truth requirements, can accommodate mixed pixels, and is closely tied to light propagation, it is a uniquely powerful tool for analyzing hyperspectral images. However, spectral variability

Weixiang Zhao, Yichen Zhang, Yingshuo Wang, Yang Deng

Experience-driven self-evolution has emerged as a promising paradigm for improving the autonomy of large language model agents, yet its reliance on self-curated experience introduces underexplored safety risks. In this study, we investigate how experience accumulation and utilization in self-evolving agents affect safety performance across web-based and embo

Daniel Fuertes, Andrea Cavallaro, Carlos R. del-Blanco, Fernando Jaureguizar

Path planning is usually solved by addressing either the (high-level) route planning problem (waypoint sequencing to achieve the final goal) or the (low-level) path planning problem (trajectory prediction between two waypoints avoiding collisions). However, real-world problems usually require simultaneous solutions to the route and path planning subproblems

Muhammad Israr Aslam, Manahil Ali, Abdul Malik Sultan, Xiao-Xiong Zeng

This study investigates the visual characteristics of a rotating black hole (BH) within the fabric of $4$D Einstein-Gauss-Bonnet gravity illuminated with two illumination models, such as a celestial light sphere and a thin accretion disk. To visualize the BH shadow images, we use a recent fisheye camera model and ray-tracing method. And then, we focus on inv

Maximilian Haug, Christian Stippel, Lukas Pscherer, Benjamin Schwendinger

Ensuring human safety is of paramount importance in warehouse environments that feature mixed traffic of human workers and autonomous mobile robots (AMRs). Current approaches often treat humans as generic dynamic obstacles, leading to conservative AMR behaviors like slowing down or detouring, even when workers are fully aware and capable of safely sharing sp

Pouya Esmaili-Dokht, Arash Yadegari, Victor Xirau, Julian Pavon

Memory simulators are used to estimate application performance on advanced memory systems, yet they may exhibit significant discrepancies compared to real hardware. This paper investigates two key questions: (1) what causes these inaccuracies, and (2) how can simulators be properly validated to ensure reliable performance predictions. We propose a methodolog

Prateek Goyal, Jatin Kumar Reddy Mothe, Swara Rajesh Shelke, Sujit Kumar Sahoo

Floating-point square-root computation is a power- and delay-critical operation in edge-AI, signal-processing, and embedded systems. Conventional implementations typically rely on multipliers or iterative pipelines, resulting in increased hardware complexity, switching activity, and energy consumption. This work presents E2AFS, a lightweight and fully multip

Francesco Veri

The Deliberative Reason Index (DRI) is increasingly used to assess the coherence between considerations and preferences in deliberative settings, including applications to LLM-generated data. Under low-signal conditions, however, the standard DRI can produce inflated scores by treating near-zero correlations as evidence of consistency. Monte Carlo simulation

Daniel Fuertes, Carlos R. del-Blanco, Fernando Jaureguizar, Juan José Navarro-Corcuera

Generating trajectories for synthetic aperture radar (SAR)-equipped aircraft poses significant challenges due to terrain constraints, and the need for straight-flight segments to ensure high-quality imaging. Related works usually focus on trajectory optimization for predefined straight-flight segments that do not adapt to the target visibility, which depends

Bo Wang, Jia Ni, Mengnan Zhao, Zhan Qin

The unauthorized use of personal data in model training has emerged as a growing privacy threat. Unlearnable examples (UEs) address this issue by embedding imperceptible perturbations into benign examples to obstruct feature learning. However, existing studies mainly evaluate UEs under from-scratch training settings, leaving their behavior under the widely a

Kiran Adhikari, Janis Nötzel

Recent work investigated the use of Stimulated Brillouin Scattering (SBS) to measure changes in fiber parameters, thereby enhancing the security of a Quantum Key Distribution (QKD) system. In this work, we focus solely on the impact of quantum technology on the task of intrusion-detection. We derive an effective input-output model for the SBS interaction, an

Biljana Mitreska

Semileptonic $b$-hadron decays proceed via charged-current interactions and provide powerful probes for testing the Standard Model and searching for New Physics effects. The advantages of studying such decays include the large branching fractions and reliable calculations of the hadronic matrix elements. Several features can be studied, such as the ratios of

Tianyi Chen, Haobo Wang, Kai Tang, Gengyu Lyu

Incomplete Multi-View Clustering (IMVC) faces the challenge of learning discriminative representations from fragmentary observations while maintaining robustness against missing views. However, prevalent Euclidean-based methods suffer from a geometric mismatch when modeling real-world data with intrinsic hierarchies, leading to semantic blurring where repres

Minyan Luo, Yuxin Zhang, Yifei Li, Xincan Wang

Narrative-driven product photography has become a prevalent paradigm in modern marketing, as coherent visual storytelling helps convey product value and establishes emotional engagement with consumers. However, existing image generation methods do not support structured narrative planning or cross-panel coordination, often resulting in weak storytelling and

Sai Vegasena

We present Open-TQ-Metal, the first implementation of fused compressed-domain attention on Apple Silicon, enabling 128K-context inference for Llama 3.1 70B on a single 64GB consumer Mac -- a configuration impossible with all existing inference frameworks. Open-TQ-Metal quantizes the KV cache to int4 on the fly and computes attention directly on the compresse

Matthieu Jonckheere, Seva Shneer

We study reaction-diffusion particle systems with several interaction mechanisms. As the number of particles tends to infinity, the system admits a mean-field limit describing the bulk behaviour. We focus on determining the propagation speed and the particle distribution around the centre of mass, which corresponds to the travelling wave of the limiting equa

Liyin Chen, Nazlee Zebardast, Mengyu Wang, Tobias Elze

Predicting disease progression from longitudinal imaging is useful for clinical decision making and trial design. Recent methods have moved toward increasing generative complexity, but the conditions under which this complexity is necessary remain unclear. We propose that generative complexity should match the entropy of the predictable component of a task's

Riza Alaudin Syah, Irwan Alnarus Kautsar, Gunawan Witjaksono, Haza Nuzly bin Abdull Hamed

Breast cancer diagnosis through thermographic image analysis remains a critical challenge in medical AI, with classical deep learning approaches facing limitations in complex thermal pattern classification tasks. This paper presents a novel Hybrid Quantum Neural Network (HQNN) architecture that integrates quantum computing principles with classical convoluti

Davit Kapanadze

This paper presents an algebraic-geometric construction of the derivative developed initially within the class of polynomial functions without introducing limits at the initial stage. Tangency is characterized by an algebraic condition: the difference between a function and a linear approximation has a double root at a given point. On this basis, the derivat

Bowen Peng, Yongxiang Liu, Jie Zhou, Xiaodong Chen

Learning robust representations across extremely heterogeneous modalities remains a fundamental challenge in multi-modal vision. As a critical and profound instantiation of this challenge, high-resolution (HR) joint optical and synthetic aperture radar (SAR) pretraining seeks modality synergy to mutually enhance single-source representations; its potential i

Guang Ping He

Although information, strictly speaking, is not a physical entity, it generally requires physical entities as its carriers, e.g., writing it down on paper, encoding it with quantum particles, or transmitting it using electro-magnetic fields. And it seems natural that these carriers cannot travel faster than light. Here we reveal that if we use quantum correl

Pollawat Hongwimol, Haoning Shang, Chutong Wang, Zhichao Wan

Product attribute extraction in e-commerce is bottlenecked by ontologies that are inconsistent, incomplete, and costly to maintain. We present AutoPKG, a multi-agent Large Language Model (LLM) framework that automatically constructs a Product-attribute Knowledge Graph (PKG) from multimodal product content. AutoPKG induces product types and type-specific attr

Yun-Peng Li, Hans-Andrea Loeliger

The paper considers the computation of L1 regularization paths in a state space setting, which includes L1 regularized Kalman smoothing, linear SVM, LASSO, and more. The paper proposes two new algorithms, which are duals of each other; the first algorithm applies to L1 regularization of independent variables while the second applies to L1 regularization of d

Vadim Bobylev, Anisa Bajkova

The characteristics of Gaia catalogues, such as trigonometric parallaxes and proper motions of stars, are discussed. Radial velocities of stars are also important for studying spatial motions. The most important results of the kinematics analysis of stellar groups from the nearest vicinity of the Sun are noted, where Gaia data allow estimating a number of pa

Mario Aragonés Lozano, Oscar Romero, Antonio León

This work introduces Structured 3D-SVD as a practical framework for the reconstruction, compression, and analysis of biological volumetric data. Inspired by the logic of matrix singular value decomposition (SVD), the proposed approach represents third-order volumetric data in the spatial domain and supports progressive reconstruction through ordered quasi-si

Zhen Wang

The aim of this paper is to study Takesaki duality for weak* closed $L^p$-operator crossed product $W^*_p(G,A,\alpha)$, where $G$ is a countable discrete Abelian group, $A$ is a unital separable weak* closed $L^p$-operator algebra ($p>1$), and $\alpha$ is a weak* continuous $p$-completely isometric action of $G$ on $A$. In this paper, we construct a weak* co

Volodymyr Lyubashenko

We define biprops as a generalization of coloured props and of symmetric weak multicategories. These are bicategories whose objects form a free monoid. They are equipped with some structure resembling a symmetric strict tensor product. We prove that a symmetric weak multicategory gives rise to a biprop and a symmetric weak multifunctor gives rise to a morphi

Bo Li, Ningyuan Deng, Tianyu Dong, Shaobo Wang

Multimodal large language models (MLLMs) have shown impressive capabilities, yet they often struggle to effectively capture the fine-grained textual information within images crucial for accurate image translation. This often leads to a modality gap between visual text inputs and textual inputs/outputs for image translation. Existing methods, primarily relyi

Dao Sy Duy Minh, Tran Chi Nguyen, Trung Kiet Huynh, Pham Phu Hoa

We present MEMRES, an agentic system for Python dependency resolution that introduces a multi-level confidence cascade where the LLM serves as the last resort. Our system combines: (1) a Self-Evolving Memory that accumulates reusable resolution patterns via tips and shortcuts; (2) an Error Pattern Knowledge Base with 200+ curated import-to-package mappings;

Junlin Li, Shuangyong Song, Guodong Du, Ngai Wong

Supervised Fine-Tuning (SFT) accelerates taskspecific large language models (LLMs) development, but the resulting proliferation of finetuned models incurs substantial memory overhead. Delta compression addresses this by retaining a single pre-trained LLM with multiple compressed delta weights. However, existing methods fail on models fine-tuned with largesca

Lorenzo Mazzucchelli, Marco Zanotti, Luca Vincenzo Ballestra, Andrea Guizzardi

This study examines the disposition effect in both long and short exposure positions in FTSE MIB tracking ETFs using a unique dataset of almost 9 million individual transactions. Building on the integrated framing approach, we extend the analysis to explicitly incorporate leverage and long short exposures, allowing us to assess how portfolio context and syst

Kutay Kulbak, Mohamed Iyad Boualem, Charlie Masse, Mariana Delalibera de Toledo

The dynamics of magnetization near a stable equilibrium in ferromagnetic nanomagnets are examined within the Landau--Lifshitz--Gilbert (LLG) framework. For a small angle precession, the dependence of ferromagnetic resonance (FMR) frequency, the damping constant and the resulting quality factor $Q$ of the resonance on the local curvature around the free-energ

Wei-Chi Wu, Sheng-Lun Wei, Hen-Hsen Huang, Hsin-Hsi Chen

Translation-based prompting is widely used in multilingual LLMs, yet its effectiveness varies across languages and tasks. We evaluate prompting strategies across ten languages of different resource levels and four benchmarks. Our analysis shows that no single strategy is universally optimal. Translation strongly benefits low-resource languages even when tran

Alexis Carrillo, Salvatore Citraro, Ali Aghazhadeh Ardebili, Enrique Taietta

Scarce longitudinal evidence examines LLMs' persuasiveness and humanness along time-evolving psychological frameworks. We introduce Talk2AI, a longitudinal framework quantifying psycho-social, reasoning and affective dimensions of LLMs' persuasiveness about polarizing societal topics. In a four-way longitudinal setup, Talk2AI's 770 participants engaged in st

Yuto Urano, Ryo Tamura, Yui Tamogami, Toshikaze Kariyado

Reliable optical control of valley degrees of freedom in moir\'e excitons requires that the emitted polarization faithfully reflect the underlying valley state. Here, we show that linearly polarized photoluminescence from WSe2/WS2 moir\'e excitons is largely insensitive to the excitation polarization and therefore does not arise from valley coherence. Automa

Marcus Kessel

Behavioral Co-Versioning remains absent from mainstream practice: while developers routinely version source code with Git, they rarely persist and query how run-time behavior evolves across revisions. This paper argues that this mismatch contributes to a blind spot in software evolution analysis and CI, where rich execution information is discarded and typic

Noga Mudrik, Adam S. Charles

Across many scientific fields, measurements often represent the number of times an event occurs. For example, a document can be represented by word occurrence counts, neural activity by spike counts per time window, or online communication by daily email counts. These measurements yield high-dimensional count data that often approximate a Poisson distributio

Jiaxin Fang, Runyuan He, Sahil Bhatia, Neel Gajare

Recent advances in large language models (LLMs) have shown that test-time scaling can substantially improve model performance on complex tasks, particularly in the coding domain. Under this paradigm, models use a larger token budget during inference to generate intermediate reasoning traces before producing a final answer. However, current evaluations primar

Xiyin Zeng, Yi Lu, Hao Wang

Visual Question Answering (VQA) requires models to identify the correct answer options based on both visual and textual evidence. Recent Mixture-of-Experts (MoE) methods improve option reasoning by grouping similar concepts or routing based on examples. However, unstable routing can lead to inconsistent expert selection in the same question type, while overl

Ruijun Huang, Zhiqiao Kang, Yuxuan Zhu, Junxiong Li

The accurate extraction of scientific measurements from literature is a critical yet challenging task in AI4Science, enabling large-scale analysis and integration of quantitative research findings. However, Large Language Models (LLMs) frequently exhibit severe hallucinations, which significantly undermine the reliability of automated scientific document und