April 2026 arXiv papers — page 141

Zhaoyang Jia, Naifu Xue, Zihan Zheng, Jiahao Li

Recent advanced diffusion methods typically derive strong generative priors by scaling diffusion transformers. However, scaling fails to generalize when adapted for real-time compression scenarios that demand lightweight models. In this paper, we explore the design of real-time and lightweight diffusion codecs by addressing two pivotal questions. First, does

BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson

A partial wave analysis of the process $\psi(2S)\rightarrow\gamma\chi_{c1}, \chi_{c1}\rightarrow\pi^+\pi^-\eta^{\prime}$ is performed using $(2712.4\pm14.3)\times10^{6}$ $\psi(2S)$ events collected with the BESIII detector. An isovector state with exotic quantum numbers $J^{PC}=1^{-+}$, denoted as $\pi_{1}(1600)$, is observed for the first time in the charmo

Sandip Das, Lavkesh Lachhvani, Kunal Singha, Rosh Roy

Energy transfer among various spectral components of fluctuating physical parameters in plasma occurs due to the nonlinear interactions, but these effects are typically not captured by the traditional linear spectral methods. Plasma density fluctuations measured in the Inverse Mirror Plasma Experimental Device (IMPED) have signatures of nonlinear mode intera

Kexin Zhang, Xiaosong Han, Ershuai Yin, Xin Qian

The integration of high-thermal-conductivity diamond films onto silicon carbide (SiC) substrates offers a promising pathway for thermal management in high-power electronic devices. Here, we investigate the depth-dependent thermal conductivity of a ~5 {\mu}m-thick diamond film grown on SiC by hot-filament chemical vapor deposition (HFCVD) using square-pulsed

Karolin Hymon, Alexander Chen, Meng-Xue Tsai, Wan-Ting Hseu

The cubic-kilometer neutrino telescopes have opened neutrino astronomy as an observational discipline. The recent detection of KM3-230213A, the highest-energy neutrino ever observed at ~220 PeV, as a near-horizontal muon track underscores that the ultra-high-energy frontier is accessed through horizontal directions where the Earth's opacity above ~100 TeV co

Rahul Kumar, Ramkumar Pathmanabhan

Potential flow theory remains a cornerstone of unsteady aerodynamics due to its computational efficiency in modeling complex flow phenomena. This study presents a significant advancement by integrating a viscous unsteady theory with established numerical vortex methods, creating a hybrid computational tool for low-to-moderate Reynolds number flows. We develo

Tianyi Lou

We study the simplicity of $C^{*}$-algebras built from group actions. For a faithful isometric action of a group $G$ on a countable metric space $X$, we use the associated action representation on $\ell^2(X)$ to define the action-based $C^{*}$-algebra $C^{*}_{X}G$. We define generalized versions of the properties $P_{\text{naive}}$ and $P_{\text{analytic}}$

Raghav Bongole, Tobias J. Oechtering, Mikael Skoglund

Recent work established a generalized-Fano framework for lower bounding prior-predictive (Bayesian) CVaR in interactive statistical decision making. In this paper, we show how to instantiate that framework in concrete interactive problems and derive explicit Bayesian CVaR lower bounds from its abstract corollaries. Our approach compares a hard model with a r

Armin Dadras, Robert Sablatnig, Franziska Proksa, Markus Seidl

The reliable computational assessment of photographic composition requires features that are discriminative of spatial layout yet robust to semantic content. This paper proposes a low-level representation grounded in the assumption that composition can be understood as the flow of visual attention across geometric structure. We introduce VFCNet, which fuses

Kang He, Yuzhe Ding, Xinrong Wang, Fei Li

Multimodal sentiment analysis (MSA) integrates heterogeneous text, audio, and visual signals to infer human emotions. While recent approaches leverage cross-modal complementarity, they often struggle to fully utilize weaker modalities. In practice, dominant modalities tend to overshadow non-verbal ones, inducing modality competition and limiting overall cont

Benjamin Vejnar

A fan is an arc-wise connected hereditarily unicoherent continuum with exactly one branching point. By a result of Borsuk, every fan is a 1-dimensional continuum that can be expressed as the union of a family of arcs, each pair of which intersects in the branching point. In this paper, we prove that the converse does not hold by providing a more general resu

Romain Guérout

We use the configuration-space Faddeev formalism to study scattering of three particles in the double continuum where all particles are free. All scattering processes, starting from and ending in both single and double continua, are collected in a unique matrix. We apply our method to the benchmark system of neutron-deuteron scattering.

Van Kien Nguyen

In this paper, we investigate the approximation problem for functions in Gaussian Sobolev spaces $W^s_p(\mathbb{R}^d, \gamma)$ of smoothness $s > 0$, where the approximation error is measured in the Gaussian Lebesgue space $L_q(\mathbb{R}^d, \gamma)$. Such function spaces naturally arise in the analysis of high-dimensional problems with Gaussian measures and

Zhiyuan Zeng, Jiameng Huang, Zhangyue Yin, Jiashuo Liu

Reinforcement learning with verifiable rewards (RLVR) has become a central paradigm for improving reasoning and code generation in large language models, and GRPO-style training is widely adopted for its simplicity and effectiveness. However, an important design choice remains underexplored: how token-level policy gradient terms are aggregated within each sa

Yves Ruffenach

We study whether a variational language model can support a minimal and measurable form of agentic control grounded in its own internal evidence. Our model combines local variational hidden computation (EVE), a homeostatic latent regulator, structurally aware checkpoint retention and a calibrated uncertainty-aware controller operating on top of the retained

Guanyi Qin, Jie Liang, Bingbing Zhang, Lishen Qu

In this paper, we present an overview of the NTIRE 2026 challenge on the 3rd Restore Any Image Model in the Wild, specifically focusing on Track 1: Professional Image Quality Assessment. Conventional Image Quality Assessment (IQA) typically relies on scalar scores. By compressing complex visual characteristics into a single number, these methods fundamentall

Víctor Blanco, Ricardo Gázquez, Juan Francisco Ocaña-Rivas

This paper proposes an optimization-based framework for the analysis of multiperiod directed multihypergraphs aimed at identifying self-amplifying structures that sustain endogenous growth in complex systems. The approach captures the progressive and nested activation of nodes and hyperarcs, providing a dynamic representation of evolving production and react

Seungwoo Lee, Minjun Park, Yunsang Noh, Sung Jin An

We report gate-controlled quantum-dot transport in a trilayer MoSe2 device that combines a graphite back gate beneath the active region, a separate global gate for conductive access regions, and local top finger gates. In the low-backgate regime, bias spectroscopy shows regular Coulomb-blockade diamonds characteristic of single-dot transport. As backgate is

Snehal Jauhri, Vignesh Prasad, Georgia Chalvatzaki

Mobile Manipulation (MoMa) of articulated objects, such as opening doors, drawers, and cupboards, demands simultaneous, whole-body coordination between a robot's base and arms. Classical whole-body controllers (WBCs) can solve such problems via hierarchical optimization, but require extensive hand-tuned optimization and remain brittle. Learning-based methods

Jilong Zhu, Yang Feng

While Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in general visual understanding, they frequently falter in fine-grained perception tasks that require identifying tiny objects or discerning subtle visual relationships. We attribute this limitation to Visual Attenuation: a phenomenon where sparse fine-grained visual sig

Lapo Rubini

We show that hyperplane sections of strongly formal manifolds inherit strong formality. In particular, this property holds for generalized complete intersections defined by positive line bundles with trivial first de Rham cohomology group. Furthermore, we establish the strong formality of compact K\"ahler manifolds with central cohomology of width $\frac{n}{

Linhao Zhang, Yuhan Song, Aiwei Liu, Chuhan Wu

Recent Audio Large Language Models (AudioLLMs) exhibit a striking performance inversion: while excelling at complex reasoning tasks, they consistently underperform on fine-grained acoustic perception. We attribute this gap to a fundamental limitation of ASR-centric training, which provides precise linguistic targets but implicitly teaches models to suppress

E. Javier Olucha, Valentin Preda, Amritam Das, Roland Tóth

This paper aims to enhance the efficiency of validation and verification campaigns involving fuel sloshing phenomena. Our first contribution is the development of an open-source, high-fidelity and computationally efficient two-dimensional smoothed-particle hydrodynamics-based fuel sloshing simulator that reproduces the dynamics of a spacecraft with a partial

Saeed shaabanian

Cover time, in the context of dynamical systems, quantifies the rate at which orbits cover the system. We prove that for countable full shifts with a Gibbs measure, equipped with a natural metric, the rate of covering of orbits of points behaves according to the minimum measure of balls. Moreover, this rate exhibits sensitivity to changes in the metric.

Junbin Su, Ziteng Xue, Shihui Zhang, Kun Chen

Parameter-efficient fine-tuning (PEFT) in multimodal tracking reveals a concerning trend where recent performance gains are often achieved at the cost of inflated parameter budgets, which fundamentally erodes PEFT's efficiency promise. In this work, we introduce SEATrack, a Simple, Efficient, and Adaptive two-stream multimodal tracker that tackles this perfo

Leon Eshuijs, Shihan Wang, Antske Fokkens

Specification gaming under Reinforcement Learning (RL) is known to cause LLMs to develop sycophantic, manipulative, or deceptive behavior, yet the conditions under which this occurs remain unclear. We train 11 instruction-tuned LLMs (0.5B-14B) with on-policy RL across 3 environments and find that model size acts as a safety buffer in some environments but en

Angela Aguglia, Gábor Korchmáros

Cyclic AG-codes $C_L(D,G)$ on the Hermitian curve $H_q$ over $\mathbb{F}_{q^2}$ are constructed such that $G = m(P_2 + \ldots + P_q)$, where $2 \le m \le q-1$ and $\mathrm{supp}(G)$ is the intersection of $H_q$ with a chord $\ell$ minus two points $P_1, P_{q+1}$. The divisor $D = Q_1 + \ldots + Q_{q^2-1}$ consists of all $q^2 - 1$ points in a single orbit un

Mahmoud Amiri, Jamile Mohammad Jafari, Sara Mostafapour, Thomas Bocklitz

We present Lit2Vec, a reproducible workflow for constructing and validating a chemistry corpus from the Semantic Scholar Open Research Corpus using conservative, metadata-based license screening. Using this workflow, we assembled an internal study corpus of 582,683 chemistry-specific full-text research articles with structured full text, token-aware paragrap

Karol Dykiert, Mateusz Stolarski, Michał Czuba, Wojciech Cieżak

Accurate pipe roughness estimation in large-scale water distribution networks is often hindered by the high cost of traditional field methods. This study investigates whether network partitioning, by utilizing hydraulic and graph-derived attributes, can enhance the calibration of these parameters. Using a high-fidelity model of a real network as a benchmark,

Louis-Brahim Beaufort

We extend two results from the theory of geodesic flows to the magnetic setting on manifolds of arbitrary dimension. First, we investigate the magnetic ray transform and establish a tensor tomography result. Second, we define and analyze the ergodicity of the magnetic frame flow under a pinching condition, building on work of Ceki\'{c}-Lefeuvre-Moroianu-Semm

D. Bisht, Ing-Guey Jiang, K. Belwal, D. C. Cınar

We present a focused multiwavelength study of blue straggler stars (BSSs) in the intermediate-age open cluster Tombaugh 2, located in the outer Galactic disk, to constrain the dominant formation pathways of BSSs in a low-density environment. Cluster members are identified using Gaia DR3 astrometry through a Gaussian Mixture Model, yielding a clean sample of

Michael Hanna, Emmanuel Ameisen

LLMs can perform seemingly planning-intensive tasks, like writing coherent stories or functioning code, without explicitly verbalizing a plan; however, the extent to which they implicitly plan is unknown. In this paper, we define latent planning as occurring when LLMs possess internal planning representations that (1) cause the generation of a specific futur

Marcus L. Marcussen, Mikkel N. Lund, Frank Grundahl, Daniel Huber

$\mu$ Herculis is a bright, nearby quadruple system. Its brightest member, $\mu$ Her Aa, displays solar-like oscillations, establishing the system as a crucial benchmark for asteroseismology, provided that its mass can be determined independently of stellar models. We aim to resolve the full hierarchical architecture of the system and determine precise, mode

Lukas Voss

Large language models (LLMs) are increasingly deployed for tabular question answering, yet calibration on structured data is largely unstudied. This paper presents the first systematic comparison of five confidence estimation methods across five frontier LLMs and two tabular QA benchmarks. All models are severely overconfident (smooth ECE 0.35-0.64 versus 0.

James Ravi Kirkpatrick

Deepfakes are synthetic media that superimpose or generate someone's likeness on to pre-existing sound, images, or videos using deep learning methods. Existing accounts of the wrongs involved in creating and distributing deepfakes focus on the harms they cause or the non-normative interests they violate. However, these approaches do not explain how deepfakes

Bin Wang, Yantao Cao, Andi Liu, Guoliang Wu

Triangular-lattice systems host a variety of ground states, ranging from quantum spin liquids to magnetically ordered phases, the latter of which can exhibit a sequence of magnetic phase transitions under applied magnetic fields. Here, we report magnetic and thermodynamic measurements, combined with powder and single-crystal neutron diffraction, on a high-sp

Jiaxin Li, Dancheng Lu, Thanh Vu, Guangjun Zhu

We provide exact formulas for the depth of the quotient ring of powers of the edge ideal of an increasing weighted path.

Shuai Wang, Yinan Yu

Large Language Models (LLMs) exhibit strong abilities in natural language understanding and generation, yet they struggle with knowledge-intensive reasoning. Structured Knowledge Graphs (KGs) provide an effective form of external knowledge representation and have been widely used to enhance performance in classical Knowledge Base Question Answering (KBQA) ta

Sunyao Zhou, Yunzi Wu, Tianhang Wang, Xinhai Li

Long-horizon collaborative vision-language navigation (VLN) is critical for multi-robot systems to accomplish complex tasks beyond the capability of a single agent. CoNavBench takes a first step by introducing the first collaborative long-horizon VLN benchmark with relay-style multi-robot tasks, a collaboration taxonomy, along with graph-grounded generation

Andreu Queralt McBride, Dmitry Krasnov, Yuri Kolesnikov, Jörg Schumacher

The flow of a liquid metal (LM) in a rectangular duct segment, subject to a uniform transverse magnetic field and uniform heating at the side walls is explored in an ample parameter space using Direct Numerical Simulation (DNS). We modify electrical wall conductivity, (either highly conducting or perfectly insulating) and investigate the effects of the buoya

Dennis Trautwein, Cornelius Ihle, Moritz Schubotz, Corinna Breitinger

The promise of decentralized peer-to-peer (P2P) systems is fundamentally gated by the challenge of Network Address Translation (NAT) traversal, with existing solutions often reintroducing the very centralization they seek to avoid. This paper presents the first large-scale measurement study of a fully decentralized NAT traversal protocol, Direct Connection U

Mahmoud Fakhry, Ascensión Gallardo-Antolín

In this article, we propose the optimization of the resolution of time-frequency atoms and the regularization of fitting models to obtain better representations of heart sound signals. This is done by evaluating the classification performance of deep learning (DL) networks in discriminating five heart valvular conditions based on a new class of time-frequenc

K. Ege de Bruin, Kyrre Glette, Kai Olav Ellefsen, Giorgia Nadizar

Optimizing the body and brain of a robot is a coupled challenge: the morphology determines what control strategies are effective, while the control parameters influence how well the morphology performs. This joint optimization can be done through nested loops of evolutionary and learning processes, where the control parameters of each robot are learned indep

Nihal Jaiswal, Siddhartha Arjaria, Gyanendra Chaubey, Ankush Kumar

Text-to-image (T2I) generative models achieve impressive visual fidelity but inherit and amplify demographic imbalances and cultural biases embedded in training data. We introduce T2I-BiasBench, a unified evaluation framework of thirteen complementary metrics that jointly captures demographic bias, element omission, and cultural collapse in diffusion models

Mahmoud Fakhry, Piergiorgio Svaizer, Maurizio Omologo

In Gaussian model-based multichannel audio source separation, the likelihood of observed mixtures of source signals is parametrized by source spectral variances and by associated spatial covariance matrices. These parameters are estimated by maximizing the likelihood through an Expectation-Maximization algorithm and used to separate the signals by means of m

Shanyong Wang, Shuhang Lin, Yining Zhao, Xi Zhu

Recent advances in large language models (LLMs) have significantly improved the alignment of models with general human preferences. However, a major challenge remains in adapting LLMs to individual preferences, which are not only diverse but also dynamic. In this paper, we introduce a novel framework, Preference-Paired Fine-Tuning (PFT), designed to align mo

Stefan S. Nixon, Philipp P. Vieweg

As one of the cornerstones of fluid mechanics, stability analyses provide essential physical insights into the growth of perturbations and eventual transition to turbulence. However, classical \enquote{frozen-time} stability analyses implicitly assume a time-independence of their base flow and thus fail for \enquote{rapidly} diffusing shear layers. Here, we

Mahounan Pericles Adjovi, Roald Eiselen, Prasenjit Mitra

Large language models (LLMs) are trained on data contributed by low-resource language communities, yet the linguistic knowledge encoded in these models remains accessible only through commercial APIs. This paper investigates whether strategic prompting can extract usable text data from LLMs for two West African languages: Hausa (Afroasiatic, approximately 80

Hsiang-Wei Huang, Shen-Liang Yang, Chuan-Chi Huang, Yueh-Nan Chen

The quantum kernel method, a promising quantum machine learning algorithm, possesses substantial potential for demonstrating quantum advantage. Although the majority of the quantum kernel is constructed in the context of gate-based quantum circuits, inspired by the idea of analog quantum computing, here we construct an analog quantum kernel and a hybrid quan

Giuliano Orso

We provide a fully analytical derivation of Anderson's orthogonality catastrophe for the one dimensional Fermi polaron integrable model, describing a system of $N$ spin-up fermions, with fixed density $n=N/L$, interacting with a single spin-down fermion via an attractive contact potential. The proof combines the determinant representations of the norm of

Christopher D. Wallbridge, Erwin Jose Lopez Pulgarin

This position paper looks briefly at the way we attempt to program robotic AI systems. Many AI systems are based on the idea of trying to improve the performance of one individual system to beyond so-called human baselines. However, these systems often look at one shot and one-way decisions, whereas the real world is more continuous and interactive. Humans,

Yi Zhao, Yang Chenggang, Yuzhuo Wang, Tong Bao

Scientific novelty drives advances at the research frontier, yet it is also associated with heightened uncertainty and potential resistance from incumbent paradigms, leading to complex patterns of scientific impact. Prior studies have primarily ex-amined the relationship between a single dimension of novelty -- such as theoreti-cal, methodological, or result

Mohamed A. Abdelwahab, Zaynab Al-Ariny, Mahmoud Fakhry, El-Sayed Hasaneen

Accurate vehicle counting through video surveillance is crucial for efficient traffic management. However, achieving high counting accuracy while ensuring computational efficiency remains a challenge. To address this, we propose a fully automated, video-based vehicle counting framework designed to optimize both computational efficiency and counting accuracy.

Lingfang Li, Procheta Sen

Fine-tuning is the dominant paradigm for adapting pretrained large language models (LLMs) to downstream NLP tasks. In practice, fine-tuning datasets may contain various forms of noise arising from annotation errors, preprocessing artifacts, or automated data collection. While prior work has focused on designing robust learning algorithms to mitigate performa

Aadrita Paul, Anwesh Ray

We study real numbers defined by multidimensional automatic arrays weighted by multiplicatively independent bases. Let $a_1, \dots, a_r\geq 2$ be integers such that $\log a_1, \dots, \log a_r$ are $\mathbb Q$-linearly independent. Given bounded automatic sequences $(p_n(i))_{n\geq 0}$ with $i=1, \dots , r$ and a function $f:\mathbb Z^r\rightarrow \mathbb Z$,

P. Lassègues, A. Ajayakumar, M. Athanasakis-Kaklamanakis, O. Ahmad

We present the first measurement of 9p 2P1/2,3/2 and 10p 2P1/2,3/2 excited levels absolute wavenumbers and radiative lifetime in francium. We used the Collinear Resonance Ionization Spectroscopy (CRIS) technique, applied on a beam of 221Fr atoms. Prior to this work, no experimental data existed for francium p-states with n > 8. The results provide a precisio

Sergio Micó-Rosa, Alvaro Villaescusa-Tebar, Saúl Fenollosa, Carlos Villena-Jiménez

The transition toward 6G networks demands energy-efficient hardware capable of active interaction with the environment. Reconfigurable Intelligent Surfaces (RIS) have emerged as a key technology for Integrated Sensing and Communications (ISAC), enabling geometric environment recognition with minimal power consumption. However, achieving targeted 3D spatial m

Jakub Zakrzewski

We present an introductory review of nonergodic dynamics in interacting many-body quantum systems, focusing on the phenomenon of many-body localization (MBL). We describe aspects of MBL and summarize the evidence for a crossover from the ergodic to the MBL regime in finite systems, using the paradigmatic XXZ model as an example. We then broaden the scope to

Anqi Zhu, Mengting Ma, Yizhen Jiang, Xiangdong Li

Pansharpening aims to synthesize high-resolution multispectral (HR-MS) images by fusing the spatial textures of panchromatic (PAN) images with the spectral information of low-resolution multispectral (LR-MS) images. While recent deep learning paradigms, especially diffusion-based operators, have pushed the performance boundaries, they often encounter spectra

Yuu Hariya

Beckner's inequality is a family of inequalities that interpolates the two fundamental functional inequalities, the logarithmic Sobolev and Poincar\'e's inequalities. It is parametrized by exponent $p\in (1,2]$ and it implies the logarithmic Sobolev inequality as $p\to 1$ and agrees with Poincar\'e's inequality when $p=2$. In this paper, employing a stochast

Yongxuan Wu, Xixun Lin, He Zhang, Nan Sun

LLM-based Multi-Agent Systems (MAS) have demonstrated remarkable capabilities in solving complex tasks. Central to MAS is the communication topology which governs how agents exchange information internally. Consequently, the security of communication topologies has attracted increasing attention. In this paper, we investigate a critical privacy risk: MAS com

Motaz Ben Hassine, Saïd Jabbour

Machine learning has become a central research area, with increasing attention devoted to explainable clustering, also known as conceptual clustering, which is a knowledge-driven unsupervised learning paradigm that partitions data into $\theta$ disjoint clusters, where each cluster is described by an explicit symbolic representation, typically expressed as a

Esen Kurt, Haithem Afli

Large Language Models (LLMs) are increasingly deployed in politically sensitive environments, where memorisation of personal data or confidential content raises regulatory concerns under frameworks such as the GDPR and its Right to be Forgotten. Translating such legal principles into large-scale generative systems presents significant technical challenges. W

Yujie Hou, Xinbing Kong, Yalin Wang, Bin Wu

Expected Shortfall (ES) is a coherent measure of tail risk that captures the average loss beyond a quantile threshold. Despite the growing literature on ES regression conditional on covariates, no existing work considers ES modeling in panel data settings where both cross-sectional and temporal dependencies are present. This paper introduces the panel ES reg

Nicolas Tacheny

In multi-criteria graph traversal, paths are compared via Pareto dominance, an ordering that identifies which paths are non-dominated, but says nothing about which path to expand next or when the search may stop. As a result, existing approaches rely on external mechanisms-heuristics, scalarization, or population-based exploration while Pareto dominance rema

Laurent Bienvenu, Santiago Cifuentes, Hugo Gimbert

For a fixed alphabet A, an infinite sequence X is said to be normal if every word w over A appears in X with the same frequency as any other word of the same length. A classical result relates normality to finite automata as follows: a sequence X is normal if and only if all gambling strategies implementable with finite deterministic automata lose all their

Qixi Zheng, Yuxiang Zhao, Tianrui Wang, Wenxi Chen

Zero-shot voice conversion (VC) aims to convert a source utterance into the voice of an unseen target speaker while preserving its linguistic content. Although recent systems have improved conversion quality, building zero-shot VC systems for interactive scenarios remains challenging because high-fidelity speaker transfer and low-latency streaming inference

Jie Shi

We introduce a weak asymptotic version of nonlinear contraction, termed \emph{asymptotic pointwise contraction}. For a mapping on a metric space, this notion requires the existence of a sequence of functions that dominate the distances between the $n$-th iterates of any two points. The sequence is assumed to converge pointwise to a limit function, and the co

Zeng You, Yaofo Chen, Qiuwu Chen, Ying Sun

Large language models (LLMs) face significant challenges in processing long contexts due to the linear growth of the key-value (KV) cache and quadratic complexity of self-attention. Existing approaches address these bottlenecks separately: Multi-head Latent Attention (MLA) reduces the KV cache by projecting tokens into a low-dimensional latent space, while s

Christoph Zwahr, Frederic Schell, Tobias Steege, Andrés Fabián Lasagni

Laser material processing has emerged as a versatile and indispensable tool in various industries, including manufacturing, healthcare, and materials science. However, the interaction of a lasers with surfaces is highly dependent on a large number of factors, including properties of the laser source such as pulse duration, wavelength and pulse form, as well

Yongxu Fu

We analytically describe the dynamically separated $\mathbb{Z}_{2}$ skin channels (wavepacket evolutions) under periodic boundary condition (PBC) in non-Hermitian systems with anomalous time-reversal symmetry (ATRS), by combining the semiclassical worldline perspective with an enhanced understanding of skin effects. These channels, tied to the initial state

Ba Tuong Vo, Ba-Ngu Vo

This work presents a tractable approach to multi-object posterior computation under a generic measurement likelihood function. While filtering is a popular solution, valuable historical information is discarded. Posterior inference, which captures the full history of the multi-object states, provides a more comprehensive solution but is notoriously difficult

Guy Hay Kalifa, Dor Kopelevitch, Amir Stern, Yoav Sagi

Neutral-atom arrays offer a promising path for quantum simulation, yet the potential of fermionic $^{40}$K remains largely constrained by state-dependent light shifts that degrade cooling and detection fidelities. This problem can be resolved by working at a magic wavelength, where the differential light shift vanishes. We report the first experimental deter

Zixing Chen, Yifeng Gao, Li Wang, Yunhan Zhao

Vision-Language-Action (VLA) models inherit rich world knowledge from vision-language backbones and acquire executable skills via action demonstrations. However, existing evaluations largely focus on action execution success, leaving action policies loosely coupled with visual-linguistic semantics. This decoupling exposes a systematic vulnerability whereby c

Zida Li, Jun Li, Yuzhe Sha, Ziqiang Li

Text-to-image (T2I) diffusion models have achieved remarkable success in image synthesis, but their reliance on large-scale data and open ecosystems introduces serious backdoor security risks. Existing defenses, particularly input-level methods, are more practical for deployment but often rely on observable anomalies that become unreliable under stealthy, se

Rémi Buffe, Alessandro Duca, Hugo Parada

We study the controllability of a linear KdV-Schr{\"o}dinger equation on the one-dimensional torus via purely imaginary bilinear controls. Considering controls spanning a suitable finite number of Fourier modes, we prove small-time global approximate controllability in L2(T). The result holds between any pair of states with the same norm and is obtained via

Paschalis Giakoumoglou, Symeon Papadopoulos

Modern diffusion-based inpainting models pose significant challenges for image forgery localization (IFL), as their full regeneration pipelines reconstruct the entire image via a latent decoder, disrupting the camera-level noise patterns that existing forensic methods rely on. We propose DiffusionPrint, a patch-level contrastive learning framework that learn

Mohammad Nasir Uddin

U.S. financial institutions deploying AI-based fraud detection face a fragmented compliance landscape spanning four regulatory frameworks -- OCC Bulletin 2011-12, SR 11-7, the CFPB AI circular, and FinCEN BSA/SAR requirements -- with no integrated governance life cycle connecting these requirements to model development, validation, and monitoring practice. T

Hathout Nabil, Basilio Calderone, Fiammetta Namer, Franck Sajous

In derivational morphology, what mechanisms govern the variation in form-meaning relations between words? The answers to this type of questions are typically based on intuition and on observations drawn from limited data, even when a wide range of languages is considered. Many of these studies are difficult to replicate and generalize. To address this issue,

Rosario Di Bartolo, Valeria Cimini, Giorgio Minati, Danilo Zia

Model-independent estimation of the properties of quantum states is a central challenge in quantum technologies, as experimental imperfections, drifts, and imprecise models of the actual quantum dynamics inevitably hinder accurate reconstructions. Here, we introduce a training strategy for photonic quantum extreme learning machines in which both the learning

Haoyu Zheng, Tianwei Lin, Wei Wang, Zhuonan Wang

Real-world industrial inspection requires not only localizing defects, but also explaining them in natural language and generating controlled defect edits. However, existing approaches fail to jointly support all three capabilities within a unified framework and evaluation protocol. We propose IAD-Unify, a dual-encoder unified framework in which a frozen DIN

Haopeng Jin

Long-sequence video diffusion transformers hit a quadratic self-attention cost that dominates runtime and memory for very long token sequences. Most efficient attention methods use one approximation everywhere, yet video features are spectrally structured: low frequencies carry global layout and coarse motion; high frequencies carry texture and fine detail.

James Brooks-Park, Søren Bech, Jan Østergaard, Steven van de Par

Room compensation aims to improve the accuracy of loudspeaker reproduction in reverberant environments. Traditional methods, however, are limited to improving only spectral (timbral) and temporal accuracy, neglecting the spatial accuracy of loudspeaker reproduction. Proposed is a method that compensates for both spectral and spatial properties of loudspeaker

Tianhui Su, Tien-Ping Tan, Salima Mdhaffar, Yannick Estève

Real-time speech synthesis requires balancing inference latency and acoustic fidelity for interactive applications. Conventional continuous text-to-speech pipelines require computationally intensive neural vocoders to reconstruct phase information, creating a significant streaming bottleneck. Furthermore, regression-based acoustic modeling frequently induces

Mohammed Asad, Mohit Bajpai, Sudhir Singh, Rahul Katarya

Accurate characterization of suspicious breast lesions in mammography is important for early diagnosis and treatment planning. While Convolutional Neural Networks (CNNs) are effective at extracting local visual patterns, they are less suited to modeling long-range dependencies. Vision Transformers (ViTs) address this limitation through self-attention, but th

Benxu Tang, Yixi Cai, Fanze Kong, Longji Yin

Efficient and scalable 3D occupancy mapping is essential for autonomous robot applications in unknown environments. However, traditional occupancy grid representations suffer from two fundamental limitations. First, explicitly storing all voxels in three-dimensional space leads to prohibitive memory consumption. Second, exhaustive ray casting incurs high upd

Riasat Sheikh, Takashi Toma, Koji Tsumura

We study a two-component pseudo-Nambu-Goldstone-boson (pNGB) dark matter (DM) model motivated by boosted dark matter (BDM). The model is based on a complex scalar field charged under a dark $\text{U}(1)_V$ gauge symmetry, with a softly broken global $\text{SU}(3)_g$ symmetry that is spontaneously broken. The pNGB nature suppresses DM--Nucleon scattering, whi

Abdessamed Qchohi, Simone Rossi

In-context learning enables transformers to adapt to new tasks from a few examples at inference time, while grokking highlights that this generalization can emerge abruptly only after prolonged training. We study task generalization and grokking in in-context learning using a Bayesian perspective, asking what enables the delayed transition from memorization

Qingying Deng, Xian'an Jin, Qi Yan

The study of partial-twuality polynomials originates from the classical operations of geometric duality and Petrie duality on cellularly embedded graphs. These involutions generate the symmetric group $S_3$, and applying them to subsets of edges yields the notions of partial-(geometric) duality, partial-Petriality, and more generally, partial-twuality. In th

Amit Pradhan

We introduce a generalized version of the Biswas-Chatterjee-Sen (BChS) model \cite{Biswas} with group interactions of size $q$, extending the original pairwise interaction dynamics. Within a mean-field framework, we derive an exact expression for the critical noise $p_c(q)$, showing that it increases monotonically with $q$ and approaches $1/2$ in the large-$

Chenyi Sun, Ziting Zhang, Kai Wan, Xiang Zhang

This paper studies the information theoretic secure aggregation problem in a three-layer hierarchical network with arbitrary heterogeneous data assignment, where clustered users communicate with an aggregation server through an intermediate layer of relays. We consider a more general setting with arbitrary heterogeneous data assignment across users, where `a

Leonie Wolf, Samson Umezulike, Gurur Öndarö, Sebastian Schinzel

Cryptographic agility is a key prerequisite for maintaining the long-term security of digital communication, particularly in light of the transition to post-quantum cryptography. To systematically assess this capability, Hohm et al. proposed the Crypto Agility Maturity Model (CAMM). In this work, we present the first evaluation of the CAMM against establishe

Pengyuan Han, Huaiyuan Lu, Bing Lyu, Jiancheng Wu

The physical origin of optical changing-look AGNs (CLAGNs), characterized by the appearance or disappearance of broad emission lines, is thought to be mainly driven by the variation of the black-hole (BH) accretion rate. In this work, we explore this issue based on a sample of {224} CLAGNs with UV-to-optical continua, where the UV radiation is more sensitive

Sihao Liu, YuFan Xiong, Zhonghua Jiang, Zhaode Wang

Multimodal Large Language Models face severe challenges in computational efficiency and memory consumption due to the substantial expansion of the visual KV cache when processing long visual contexts. Existing KV cache compression methods typically rely on the "persistence of importance" hypothesis to prune tokens. However, this approach proves fragile in mu

Alicia Curth, Rachel Lawrence, Sushrut Karmalkar, Niranjani Prasad

We investigate whether transformers use their depth adaptively across tasks of increasing difficulty. Using a controlled multi-hop relational reasoning task based on family stories, where difficulty is determined by the number of relationship hops that must be composed, we monitor (i) how predictions evolve across layers via early readouts (the logit lens) a

Michele De Vita, Julian Wiederer, Vasileios Belagiannis

Trajectory prediction models often fail in real-world automated driving due to distributional shifts between training and test conditions. Such distributional shifts, whether behavioural or environmental, pose a critical risk by causing the model to make incorrect forecasts in unfamiliar situations. We propose a self-supervised method that trains a decoder i

Sihang Jia, Shuliang Liu, Songbo Yang, Yibo Yan

Multimodal Large Language Models frequently suffer from inference hallucinations, partially stemming from language priors dominating visual evidence. Existing training-free mitigation methods either perturb the visual representation and deviate from the natural image distribution, or enforce intrusive manipulations that compromise the model's inherent genera

Jinlu Li, Yanghai Yu

In this paper, we consider the Cauchy problem for the rod equation in the line. By constructing an explicit smooth initial data, we present a new method to prove that this problem is ill-posed in $H^s(\R)$ with $1< s<3/2$ in the sense of {\it norm inflation}, i.e., an initial data is smooth and arbitrarily small in $H^s(\R)$ with $1< s<3/2$, but the solution

Savvas Panagi, Chrysovalantis Spanias, Petros Aristidou

The growing penetration of distributed energy resources (DERs), electric vehicles (EVs), and heat pumps (HPs) in distribution networks underscores the need for secure, computationally efficient optimal power flow (OPF) solutions. Traditional OPF formulations often suffer from scalability limitations and may rely on relaxations/approximations whose exactness

Micha Selak, Dirk Krechel, Adrian Ulges, Sven Spieckermann

Extracting actionable insights from complex value stream map simulations can be challenging, time-consuming, and error-prone. Recent advances in large language models offer new avenues to support users with this task. While existing approaches excel at processing raw data to gain information, they are structurally unfit to pick up on subtle situational diffe