October 2025 arXiv papers — page 78

Mahitha Pulivathi, Ana Fontes Rodrigues, Isibor Kennedy Ihianle, Andreas Oikonomou

Deep learning is widely applied to modern problems through neural networks, but the growing computational and energy demands of these models have driven interest in more efficient approaches. Spiking Neural Networks (SNNs), the third generation of neural networks, mimic the brain's event-driven behaviour, offering improved performance and reduced power use.

Bojan Nikolic

In this article I consider type II superstring in the pure spinor formulation with constant background fields in the context of T-dualization. First I prove that bosonic and fermionic T-dualization commute using already known T-dual transformation laws for bosonic and fermionic T-dualization. Consequently, the T-dual transformation laws of the full T-dualiza

Markus Bujotzek, Evelyn Trautmann, Calum Hand, Ian Hales

AI methods are increasingly shaping pharmaceutical drug discovery. However, their translation to industrial applications remains limited due to their reliance on public datasets, lacking scale and diversity of proprietary pharmaceutical data. Federated learning (FL) offers a promising approach to integrate private data into privacy-preserving, collaborative

Jean Van Schaftingen, Leon Winter

Morrey--Sobolev inequalities are established for functions in weighted Sobolev spaces on the $n$-dimensional half-space, where the weight is a power of the distance to the boundary, as well as for Sobolev spaces on the $n$-dimensional hyperbolic space. All the estimates are optimal up to a multiplicative constant.

Clio Johnson, Neil D. Drummond, James P. Hague, Calum MacCormick

Fermionic cold atoms in optical traps provide viable quantum simulators of correlation effects in electronic systems. For dressed Rydberg atoms in two-dimensional traps with out-of-plane dipole moments, a realistic model of the pairwise interaction is of repulsive dipolar $1/r^3$ form at long range, softened to a constant at short range. This study provides

Selena Luo, Mark S. Keller, Tabassum Kakar, Lisa Choy

EasyVitessce is a Python package that turns existing static Scanpy and SpatialData plots into interactive visualizations by virtue of adding a single line of Python code. The package uses Vitessce internally to render interactive plots, and abstracts away technical details involved with configuration of Vitessce. The resulting interactive plots can be viewed

Waris Radji, Odalric-Ambrym Maillard

We revisit the problem of controlling linear systems with quadratic cost under unknown dynamics with model-based reinforcement learning. Traditional methods like Optimism in the Face of Uncertainty and Thompson Sampling, rooted in multi-armed bandits (MABs), face practical limitations. In contrast, we propose an alternative based on the Confusing Instance (C

Ruiyao Miao, Junren Xiao, Shiya Tsang, Hui Xiong

Existing Bayesian Optimization (BO) methods typically balance exploration and exploitation to optimize costly objective functions. However, these methods often suffer from a significant one-step bias, which may lead to convergence towards local optima and poor performance in complex or high-dimensional tasks. Recently, Black-Box Optimization (BBO) has achiev

Sean Dewar, Bernd Schulze, Shin-ichi Tanigawa, Louis Theran

We construct infinite periodic versions of the stress matrix and establish sufficient conditions for periodic tensegrity frameworks to be globally rigid in $\mathbb{R}^d$ in the cases when the lattice is either fixed, fully flexible, or flexible with a volume constraint for the fundamental domain. For the fixed and fully flexible lattice variants, we also es

Qing Mao, Tianxin Huang, Yu Zhu, Jinqiu Sun

Pairwise camera pose estimation from sparsely overlapping image pairs remains a critical and unsolved challenge in 3D vision. Most existing methods struggle with image pairs that have small or no overlap. Recent approaches attempt to address this by synthesizing intermediate frames using video interpolation and selecting key frames via a self-consistency sco

Rohan Senthil, Swee Liang Wong

Anomaly detection in cybersecurity is a challenging task, where normal events far outnumber anomalous ones with new anomalies occurring frequently. Classical autoencoders have been used for anomaly detection, but struggles in data-limited settings which quantum counterparts can potentially overcome. In this work, we apply Quantum Autoencoders (QAEs) for anom

V. Apinyan, T. Kopeć

In this study, we consider a two-layer graphene structure stacked in the AA form and exposed to the influence of two different electric fields applied to different layers. The graphene layers are also subjected to an external magnetic field perpendicular to the planes of the layers. We investigate the possible effects of the applied in-plane fields and the m

Vincent Savaux, Hyeon Seok Rou, Zeping Sui, Giuseppe Thadeu Freitas de Abreu

We investigate the robustness of affine frequency division multiplexing (AFDM) and orthogonal time frequency space (OTFS) waveforms against passive eavesdroppers performing brute-force demodulation to intercepted signals, under the assumption that eavesdroppers have no knowledge of chirp parameters (in AFDM) or the delay-Doppler grid configuration (in OTFS),

Perla Kfoury, Stefan Le Coz, Tai-Peng Tsai

We study the quasi-periodic standing wave solutions of the focusing and defocusing cubic nonlinear Schr{\"o}dinger equations in dimension one. In the defocusing case, we establish a diffeomorphic correspondence between the invariants of the ordinary differential equation of the wave profiles and the conserved quantities of the evolution equation. We introduc

Xiaoqi Feng, Bingzhe Hou, Kui Ji

In 1978, M. J. Cowen and R. G. Douglas introduced a class of geometric operators (known as Cowen-Douglas class of operators) and associated a Hermitian holomorphic vector bundle to such operators. In this paper, after giving some basic properties of $S$-spectrum and right eigenvalues of bounded right linear operators on separable quaternionic Hilbert spaces,

Charlotte Bäcker, Krishna Palaparthy, Walter T. Strunz

We investigate memory effects in non-Markovian dynamics on superconducting quantum processors provided by IBM Quantum. We use a collision-model approach to implement suitable single- and two-qubit dynamics with a gate-based quantum circuit. Coupling the system of interest to an ancilla allows for a characterization of the process with respect to non-Markovia

Zexin Fang, Bin Han, Zhu Han, Yufei Zhao

This paper investigates security vulnerabilities and countermeasures for the 3rd Generation Partnership Project (3GPP) Fifth Generation New Radio (5G-NR) Time Difference of Arrival (TDoA)-based unmanned aerial vehicle (UAV) localization in low-altitude urban environments. We first optimize node selection strategies under Air to Ground (A2G) channel condition

Xiangyu Li, Haojie Yang, Kaimiao Hu, Runzhi Wu

Accurate prediction of drug-target interactions (DTI) is pivotal for drug discovery, yet existing methods often fail to address challenges like cross-domain generalization, cold-start prediction, and interpretability. In this work, we propose CDI-DTI, a novel cross-domain interpretable framework for DTI prediction, designed to overcome these limitations. By

O. O. Svystunov, I. V. Beznosenko, A. V. Vasyliev, R. R. Kniaziev

The study investigated transparent on-chip structures with a rectangular profile and triangular profiles with grating ridge base angles of $\alpha = 36^\circ$, $30^\circ$, and $20^\circ$. Each triangular structure had both left- and right-handed profile orientations. For all variants, a modified version with a reflective gold coating was additionally conside

Benjamin Carrel, Daniel Kressner, Hei Yin Lam, Bart Vandereycken

Dynamical low-rank approximation (DLRA) is a widely used paradigm for solving large-scale matrix differential equations, as they arise, for example, from the discretization of time-dependent partial differential equations on tensorized domains. Through orthogonally projecting the dynamics onto the tangent space of a low-dimensional manifold, DLRA achieves a

Shuli Zhang, Hao Zhou, Jiaqi Zheng, Guibin Jiang

Online Internet platforms require sophisticated marketing strategies to optimize user retention and platform revenue -- a classical resource allocation problem. Traditional solutions adopt a two-stage pipeline: machine learning (ML) for predicting individual treatment effects to marketing actions, followed by operations research (OR) optimization for decisio

Yaoming Zhen, Piotr Zwiernik

In probabilistic principal component analysis (PPCA), an observed vector is modeled as a linear transformation of a low-dimensional Gaussian factor plus isotropic noise. We generalize PPCA to tensors by constraining the loading operator to have Tucker structure, yielding a probabilistic multilinear PCA model that enables uncertainty quantification and natura

Raffaele Resta

The theory of the intrinsic Hall effect, both linear and nonlinear, is rooted in a geometry which is defined in the Bloch-vector parameter space; the formal expressions are mostly derived from semiclassical concepts. When disorder and interaction are considered there is no Bloch vector to speak of; one needs a more general quantum geometry, defined in a diff

Maciej Mozolewski, Betül Bayrak, Kerstin Bach, Grzegorz J. Nalepa

In eXplainable Artificial Intelligence (XAI), instance-based explanations for time series have gained increasing attention due to their potential for actionable and interpretable insights in domains such as healthcare. Addressing the challenges of explainability of state-of-the-art models, we propose a prototype-driven framework for generating sparse counter

T. S. Kosmas, R. Sahu, V. K. B. Kota

The recent observations of the coherent neutrino- and antineutrino-nucleus scattering have opened up a plethora of opportunities to probe physics within standard and non-standard theories of the electroweak interactions. In the present article, our goal is to explore the possibility of using the molybdenum material as detection medium for coherent and incohe

Cedric Faas, Sophie Kerstan, Richard Uth, Markus Langer

As AI systems become increasingly capable and autonomous, domain experts' roles are shifting from performing tasks themselves to overseeing AI-generated outputs. Such oversight is critical, as undetected errors can have serious consequences or undermine the benefits of AI. Effective oversight, however, depends not only on detecting and correcting AI errors b

Ishita Modak, Rajesh Narayanan, Ferdinand Evers, Soumya Bera

Hilbert space fragmentation, as it is currently investigated, primarily originates from specific kinematic constraints or emergent conservation laws in many-body systems with translation invariance. It leads to non-ergodic dynamics and possible breakdown of the eigenstate thermalization hypothesis. Here, we demonstrate that also in disordered systems, such a

Maureen de Seyssel, Eeshan Gunesh Dhekane

Speech foundation models have recently achieved remarkable capabilities across a wide range of tasks. However, their evaluation remains disjointed across tasks and model types. Different models excel at distinct aspects of speech processing and thus require different evaluation protocols. This paper proposes a unified taxonomy that addresses the question: Wh

Jaya Krishna Mandivarapu

Learning a set of tasks over time, also known as continual learning (CL), is one of the most challenging problems in artificial intelligence due to catastrophic forgetting. Large language models (LLMs) are often impractical to frequent re-training and continual learning , due to high cost of computational resources for training. Moreover, LLM are not suitabl

Hoang Phi Dung, Vu The Khoi

In this study, we investigate the problem of determining the maximum purity for absolutely separable and absolutely PPT quantum states. From the geometric viewpoint, this problem is equivalent to asking for the exact Euclidean radius of the smallest ball around the maximally mixed state that encompasses the set of all absolutely separable or absolutely PPT s

Demian Till, John Smeaton, Peter Haubrick, Gouse Saheb

Recent work has demonstrated state-of-the-art results in large language model (LLM) hallucination detection and mitigation through consistency-based approaches which involve aggregating multiple responses sampled from a single LLM for a given prompt. These approaches help offset limitations stemming from the imperfect data on which LLMs are trained, which in

Dávid Csercsik, Ádám Sleisz

When the traded energy and reserve products between zones are co-allocated to optimize the infrastructure usage, both deterministic and stochastic flows have to be accounted for on interconnector lines. We focus on allocation models, which guarantee deliverability in the context of the portfolio bidding European day-ahead market framework, assuming a flow-ba

Canbin Huang, Tianyuan Shi, Yuhua Zhu, Ruijun Chen

Large language model (LLM) routers improve the efficiency of multi-model systems by directing each query to the most appropriate model while leveraging the diverse strengths of heterogeneous LLMs. Most existing approaches frame routing as a classification problem based solely on the input query. While this reduces overhead by avoiding inference across all mo

Y. -C. Soong, H. Li, Y. Fu, J. Tong

The electrochemical hydrogenation of graphene induces a robust and reversible conductor-insulator transition, of strong interest in logic-and-memory applications. However, its mechanism remains unknown. Here we show that it proceeds as a reduction reaction in which proton adsorption competes with the formation of H2 molecules via an Eley-Rideal process. Grap

Wei-Bo He, Yun-Tong Yang, Hong-Gang Luo

The development of novel quantum many-body computational algorithms relies on robust benchmarking. However, generating such benchmarks is often hindered by the massive computational resources required for exact diagonalization or quantum Monte Carlo simulations, particularly at finite temperatures. In this work, we propose a new algorithm for obtaining therm

Tim E. Veenstra, René van Roij, Marjolein Dijkstra

Harnessing the intrinsic dynamics of physical systems for information processing opens new avenues for computation embodied in matter. Using simulations of a model system, we show that assemblies of DNA tiles capable of self-organizing into multiple target structures can perform basic computational tasks analogous to those of finite-state automata when equip

Anvarbek Meirmanov

Mathematical models of joint filtration of liquids is the main part of mathematical models of oil displacement by suspension. Since mining is a very important and urgent economic task, exact modeling of joint filtration of two different fluids is also an urgent economic task. For example, mathematical models of oil displacement by suspension to create a hydr

Christian Carrick

We define a $t$-structure on the category of filtered $G$-spectra such that for a Borel $G$-spectrum $X$ the slice filtration of $X$ is the connective cover of the homotopy fixed-point filtration of $X$. Using this, we show that the slice spectral sequence for the norm $N_{C_2}^GMU_{\mathbb{R}}$ of Real bordism theory refines canonically to a $\mathbb{E}_\in

Lang Zhou, Amrish Jhingoer, Yinghao Luo, Klaske Vliegenthart--Jongbloed

Efficient screening and early diagnosis of HIV are critical for reducing onward transmission. Although large scale laboratory testing is not feasible, the widespread adoption of Electronic Health Records (EHRs) offers new opportunities to address this challenge. Existing research primarily focuses on applying machine learning methods to structured data, such

Mahiro Shirotori, Yutaka Fujita

Jets from active galactic nuclei (AGNs) are expected to heat the surrounding intracluster medium (ICM). We investigate how the interaction between jets and the ICM appears in high-resolution X-ray observations using mock X-ray observations based on two-dimensional hydrodynamic simulations. We constructed a model of an active galactic nucleus (AGN) similar to

Nayan Kumar Singh

Manually generating catchy descriptions and names is labor intensive and a slow process for retailers. Although generative AI provides an automation solution in form of Vision to Language Models (VLM), the current VLMs are prone to factual "hallucinations". Siloed, single task models are not only inefficient but also fail to capture interdependent relationsh

Mark S. Keller, Nicholas Lucarelli, Yijiang Chen, Samuel Border

Application of machine learning techniques enables segmentation of functional tissue units in histology whole-slide images (WSIs). We built a pipeline to apply previously validated segmentation models of kidney structures and extract quantitative features from these structures. Such quantitative analysis also requires qualitative inspection of results for qu

Chenyu Wang, Zhanglu Yan, Zhi Zhou, Xu Chen

In the era of large language models (LLMs), weight-activation quantization helps fit models on edge device by reducing memory and compute bit-widths. However, three challenges persist for energy constrained hardware: (1) even after quantization, multiply-accumulate (MAC) operations remain unavoidable and continue to dominate energy consumption; (2) dequantiz

Trung-Dung Vu, Benoit Gaudou, Kamaldeep Singh Oberoi

Modeling realistic human behaviour to understand people's mode choices in order to propose personalised mobility solutions remains challenging. This paper presents an architecture for modeling realistic human mobility behavior in complex multimodal transport systems, demonstrated through a case study in Toulouse, France. We apply Large Language Models (LLMs)

Kevin Huang, Rosario Scalise, Cleah Winston, Ayush Agrawal

Imitation learning has proven effective for training robots to perform complex tasks from expert human demonstrations. However, it remains limited by its reliance on high-quality, task-specific data, restricting adaptability to the diverse range of real-world object configurations and scenarios. In contrast, non-expert data -- such as play data, suboptimal d

Fernando Alonso, Álvaro Leitao, Carlos Vázquez

The ongoing progress in quantum technologies has fueled a sustained exploration of their potential applications across various domains. One particularly promising field is quantitative finance, where a central challenge is the pricing of financial derivatives-traditionally addressed through Monte Carlo integration techniques. In this work, we introduce two h

Suchir Salhan, Richard Diehl Martinez, Zébulon Goriely, Paula Buttery

Transformer language models typically operate with a fixed-length context window, which has grown in step with large-scale pretraining datasets. In the BabyLM Challenge, however, many past submissions have defaulted to using much shorter sequence lengths. We examine the impact of sequence length on BabyLM pretraining, to answer the simple question: what sequ

Apolline Chappard, José Alberto Rubiño-Martín, Ricardo Tanausú Génova Santos

QUIJOTE is a CMB experiment composed of two telescopes, QT1 and QT2, located at the Teide Observatory in Tenerife, Spain. The MFI instrument (2012-2018), installed on QT1, observed the sky at four frequency bands (11, 13, 17, and 19 GHz) with one degree angular resolution. Its successor, MFI2, began operations in 2024 and operates in the same bands. This pap

Ryuto Koike, Liam Dugan, Masahiro Kaneko, Chris Callison-Burch

Although membership inference attacks (MIAs) and machine-generated text detection target different goals, their methods often exploit similar signals based on a language model's probability distribution, and the two tasks have been studied independently. This can result in conclusions that overlook stronger methods and valuable insights from the other task.

Jonas Gebele, Timm Mutzel, Burak Oez, Florian Matthes

Sealed-bid auctions ensure fair competition and efficient allocation but are often deployed on centralized infrastructure, enabling opaque manipulation. Public blockchains eliminate central control, yet their inherent transparency conflicts with the confidentiality required for sealed bidding. Prior attempts struggle to reconcile privacy, verifiability, and

Jasem Hamoud, Alexey Belov Yakovlevich, Muaadh Almahalebi, Duaa Abdullah

This study explores the irregularity properties of trees with prescribed degree sequences by analyzing two prominent topological indices: the Albertson index and the sigma index. With a particular emphasis on caterpillar trees -frequently used to model molecular chains- we derive a closed-form expression for the Albertson index: \[ \mathrm{irr}(\mathscr{C}(n

Hemendra M. Naik

The exponential growth of multimedia streaming services over the Internet emphasizes the increasing significance of ensuring a seamless and high-quality streaming experience for users. Dynamic Adaptive Streaming over HTTP (DASH) has emerged as a popular solution for delivering multimedia content over variable network conditions. However, challenges such as n

František Bartoš, Samuel Pawel, Björn S. Siepe

Simulation studies are widely used to evaluate statistical methods. However, new methods are often introduced and evaluated using data-generating mechanisms (DGMs) devised by the same authors. This coupling creates misaligned incentives, e.g., the need to demonstrate the superiority of new methods, potentially compromising the neutrality of simulation studie

Dunjie Lu, Yiheng Xu, Junli Wang, Haoyuan Wu

Training computer-use agents requires massive amounts of GUI interaction data, but manually annotating action trajectories at scale is prohibitively expensive. We present VideoAgentTrek, a scalable pipeline that automatically mines training data from publicly available screen-recorded videos at web scale, eliminating the need for manual annotation. Our appro

Chen Li, Huiying Xu, Changxin Gao, Zeyu Wang

Single-source Domain Generalized Object Detection (SDGOD), as a cutting-edge research topic in computer vision, aims to enhance model generalization capability in unseen target domains through single-source domain training. Current mainstream approaches attempt to mitigate domain discrepancies via data augmentation techniques. However, due to domain shift an

Jacqueline Hidalgo-Jimenez, Taner Akbay, Tatsumi Ishihara, Kaveh Edalati

Photocatalytic hydrogen (H2) production and carbon dioxide (CO2) conversion to methane (CH4) are considered promising solutions for reducing CO2 emissions. However, the development of highly active photocatalysts is essential to efficiently drive these reactions without harming the environment. In this study, we introduce a strategy that incorporates element

Owais Ullah Faiz, Mushahid Hussain, Shashank Shalgar

In the vicinity of neutron star mergers (NSMs), it is possible for the neutrino self-interaction potential to cancel with the matter potential leading to matter neutrino resonance (MNR). MNR is one of the most interesting mechanisms by which neutrino flavor evolution can occur in dense astrophysical environments. Previous studies have typically assumed that

Zaifei Yang, Hong Chang, Ruibing Hou, Shiguang Shan

The molecular large language models have garnered widespread attention due to their promising potential on molecular applications. However, current molecular large language models face significant limitations in understanding molecules due to inadequate textual descriptions and suboptimal molecular representation strategies during pretraining. To address the

Xin Nie, Liang Dong, Haicheng Zhang, Jiawang Xiao

Weight quantization effectively reduces memory consumption and enable the deployment of Large Language Models on edge devices, yet existing hardware-friendly methods often rely on uniform quantization, which suffers from poor weight-distribution fitting and high dequantization overhead under low-bit settings. In this paper, we propose ELUTQ, an efficient qua

Xiaoge Bao, Wei P. Dai, Jan Nagler, Wei Lin

Understanding how transient dynamics unfold in response to localized inputs is central to predicting and controlling signal propagation in network systems, including neural processing, epidemic intervention, and power-grid resilience. Existing theoretical frameworks typically assume homogeneous network structures and constant or pulse-like inputs, overlookin

Iasonas Nikolaou, Miltiadis Stouras, Stratis Ioannidis, Evimaria Terzi

Given a collection of monotone submodular functions, the goal of Two-Stage Submodular Maximization (2SSM) [Balkanski et al., 2016] is to restrict the ground set so an objective selected u.a.r. from the collection attains a high maximal value, on average, when optimized over the restricted ground set. We introduce the Online Two-Stage Submodular Maximization

Qiang Chen, Zhongze Wu, Ang He, Xi Lin

Recent advancements in graph unlearning models have enhanced model utility by preserving the node representation essentially invariant, while using gradient ascent on the forget set to achieve unlearning. However, this approach causes a drastic degradation in model utility during the unlearning process due to the rapid divergence speed of gradient ascent. In

Julia Wąsala, Joannes D. Maasakkers, Ilse Aben, Rochelle Schneider

Most satellite images have systematically missing pixels (i.e., missing data not at random (MNAR)) due to factors such as clouds. If not addressed, these missing pixels can lead to representation bias in automated feature extraction models. In this work, we show that spurious association between the label and the number of missing values in methane plume det

B. M. Walsh, D. T. Welling, Z. Huang

While humans become more reliant on Earth's space environment, the potential for significant harm from severe space weather continues to grow. As structures from the sun reach Earth's magnetosphere and space environment, they deposit energy that fuels geomagnetic storms. Currently, space weather researchers work to predict the timing and intensity of space w

Peter Ván

In this note we compare the entropy principle and the objectivity arguments in the methodologies of Dunn and Serrin [1] and in the more recent weakly nonlocal thermodynamic analysis of Korteweg-type fluids in [2]. It is concluded that the different objectivity approaches lead to the same constitutive functions, and that the difference in the thermodynamicall

Julian Schulz

As AI systems approach dangerous capability levels where inability safety cases become insufficient, we need alternative approaches to ensure safety. This paper presents a roadmap for constructing safety cases based on chain-of-thought (CoT) monitoring in reasoning models and outlines our research agenda. We argue that CoT monitoring might support both contr

Constance Ferragu, Jonathan D. Ziegler, Nicolas Deutschmann, Arthur Lindoulsi

Direct Preference Optimization (DPO) is an effective approach for aligning protein language models with experimental design goals. However, DPO faces a scalability bottleneck: the number of possible training pairs grows quadratically with the number of labeled sequences, leading to prohibitive training times even for modestly sized datasets. We introduce g-D

Martin Hasenbusch

We study the $2$-component $\phi^4$ model on the simple cubic lattice in the presence of a cubic, or equivalently, a $\mathbb{D}_4$ invariant perturbation. To this end, we perform Monte Carlo simulations in conjunction with a finite size scaling analysis of the data. We follow previous work on the $3$-component case. We study the RG flow from the decoupled I

Juhyung Park, Rokgi Hong, Roh-Eul Yoo, Jaehyeon Koo

Recent advancements in artificial intelligence have created transformative capabilities in image synthesis and generation, enabling diverse research fields to innovate at revolutionary speed and spectrum. In this study, we leverage this generative power to introduce a new paradigm for accelerating Magnetic Resonance Imaging (MRI), introducing a shift from im

Weihao Yang, Hao Huang, Donglei Wu, Ningke Li

Mixture-of-Experts (MoE) has become a popular architecture for scaling large models. However, the rapidly growing scale outpaces model training on a single DC, driving a shift toward a more flexible, cross-DC training paradigm. Under this, Expert Parallelism (EP) of MoE faces significant scalability issues due to the limited cross-DC bandwidth. Specifically,

Tamara Pardo Yanguas

Numerous astronomical and cosmological observations point to the existence of dark matter, which constitutes about 27% of the Universe. Despite extensive efforts, only the DAMA/LIBRA experiment, using NaI(Tl) detectors at Gran Sasso National Laboratory, has reported a positive dark matter signal. To independently verify this result, using the same NaI target

Aritra Ghosh

In this article we show simultaneous non-vanishing of two Rankin-Selberg $L$-functions by proving an asymptotic result in weight aspect. The main input of this paper is to remove the $t$-integral dependence from the result of Blomer-Harcos (see \cite{BH2}) and getting a square root exponent for the error term.

Debarthi Pal, Ritajit Majumdar

Current quantum computers suffer from noise due to lack of error correction. Several techniques to mitigate the effect of noise have been studied, in particular to extract the expectation value of observables. One such technique, circuit cutting, partitions large circuits into smaller, less noisy subcircuits, but the exponential increase in the number of cir

T. O'Beirne, V. A. Kilborn, M. E. Cluver, O. I. Wong

Using the Widefield ASKAP L-band Legacy All-sky Blind surveY (WALLABY) we performed an untargeted search for H I-bearing ultra-diffuse galaxies (UDGs). We identified a core sample of 10 UDGs defined by $\mu_{g,0}\ge24$ mag arcsec$^{-2}$ and $R_{e}\ge1.5$ kpc, and a broader sample including 12 additional faint diffuse galaxies ($\mu_{g,0}\ge23.7$ mag arcsec$^

Khushboo Soni, S. Thirumugam, John Rozario Jegaraj, Nithyanadan Kanagaraj

High-power laser technologies are essential in precision manufacturing, defense, and scientific research, where accurate control of the beam profile is paramount. Although several beam-shaping methods exist, they often face implementation and scalability challenges. To address these limitations, we introduce a comprehensive and versatile framework for on-dem

Hao-Chiang Shao, Chun-Hao Chang, Yu-Hsien Lin, Chia-Wen Lin

Despite significant advancements in deep classification techniques and in-lab automatic optical inspection models for long-tailed or highly imbalanced data, applying these approaches to real-world IC defect classification tasks remains challenging. This difficulty stems from two primary factors. First, real-world conditions, such as the high yield-rate requi

Zhonghao Zhan, Amir Al Sadi, Krinos Li, Hamed Haddadi

In this work, we study security of Model Context Protocol (MCP) agent toolchains and their applications in smart homes. We introduce AegisMCP, a protocol-level intrusion detector. Our contributions are: (i) a minimal attack suite spanning instruction-driven escalation, chain-of-tool exfiltration, malicious MCP server registration, and persistence; (ii) NEBUL

Thomas I. Strasser, Edmund Widl, Carlos Ayon Mac Gregor, Mirko Ginocchi

The ongoing transformation of the European energy landscape, driven by the integration of renewable energy sources, digital technologies, and decentralized systems, requires a high degree of interoperability across diverse components and systems. Ensuring that these elements can exchange information and operate together reliably is essential for achieving a

Clemens Niederegger, Tatiana Vovk, Elias Starchl, Lukas M. Sieberer

Continuous monitoring of one-dimensional free fermionic systems can generate phenomena reminiscent of quantum criticality, such as logarithmic entanglement growth, algebraic correlations, and emergent conformal invariance, but in a nonequilibrium setting. However, whether these signatures reflect a genuine phase of nonequilibrium quantum matter or persist on

Kailin Jiang, Ning Jiang, Yuntao Du, Yuchen Ren

Large Multimodal Models (LMMs) encode rich factual knowledge via cross-modal pre-training, yet their static representations struggle to maintain an accurate understanding of time-sensitive factual knowledge. Existing benchmarks remain constrained by static designs, inadequately evaluating LMMs' ability to understand time-sensitive knowledge. To address this

Jonathan M Fraser, Yunlong Xu

Let $ E $ be a non-empty compact subset of the Riemann sphere and $T$ be a rational map of degree at least two. We study the associated \emph{orbital set}, that is, the backwards orbit of $E$ under $T$, and study the relationship between the upper box dimension of the orbital set and the upper box dimensions of the Julia set of $T$ and the initial set $ E$.

Banan Alnemri, Arwa Basbrain

Accurate segmentation and precise morphological analysis of neuronal cells in fluorescence microscopy images are crucial steps in neuroscience and biomedical imaging applications. However, this process is labor-intensive and time-consuming, requiring significant manual effort and expertise to ensure reliable outcomes. This work presents a pipeline for neuron

Elian Vanderborght, Henk A. Dijkstra

The Global Overturning Circulation (GOC) is a key component of the climate system, transporting heat, carbon, and salt throughout the global ocean. Previous reduced-dimensional models have sought to represent this three-dimensional circulation but often neglected three key observational features: (1) the meridional overturning circulation is in geostrophic b

Jean-Marie Le Ray

A policy-governed RAG architecture is specified for audit-ready generation in regulated workflows, organized as a triptych: (I) Contracts/Control (SHRDLU-like), which governs output adherence to legal and internal policies; (II) Manifests/Trails (Memex-like), which cryptographically anchors all cited source evidence to ensure verifiable provenance; and (III)

Dhanya Roy, Gabriele Di Stefano, Sandi Klavžar, Aparna Lakshmanan S

If $x\in V(G)$, then $S\subseteq V(G)\setminus\{x\}$ is an $x$-visibility set if for any $y\in S$ there exists a shortest $x,y$-path avoiding $S$. The $x$-visibility number $v_x(G)$ is the maximum cardinality of an $x$-visibility set, and the maximum value of $v_x(G)$ among all vertices $x$ of $G$ is the vertex visibility number ${\rm vv}(G)$ of $G$. It is p

Xueqi Ma, Yanbei Jiang, Sarah Erfani, James Bailey

Multimodal Large Language Models (MLLMs) have demonstrated exceptional performance across various objective multimodal perception tasks, yet their application to subjective, emotionally nuanced domains, such as psychological analysis, remains largely unexplored. In this paper, we introduce PICK, a multi-step framework designed for Psychoanalytical Image Comp

Clara Punzi

The rapid growth of the digital platform economy is transforming labor markets, offering new employment opportunities with promises of flexibility and accessibility. However, these benefits often come at the expense of increased economic exploitation, occupational segregation, and deteriorating working conditions. Research highlights that algorithmic managem

Steven Robertson, Noy Soffer Aranov

This is the first of a pair of papers, whose collective goal is to disprove a conjecture of Kemarsky, Paulin, and Shapira (KPS) on the escape of mass of Laurent series. This paper lays the foundations on which its sibling builds. In particular, the $p$-Cantor sequence is introduced. This generalises the classical Cantor sequence into a $p$-automatic sequence

Maria Lucia Sambataro, Salvatore Plumari, Santosh K. Das, Vincenzo Greco

We introduce the $p_T$-differential radial flow $v_0(p_T)$ in the heavy-quark sector. Within an event-by-event Langevin framework, we show that this observable exhibits a strong sensitivity to the heavy quark-bulk interaction. It provides a powerful and novel tool to constrain the transport coefficients of heavy quarks in the QGP and, more generally, to asse

Jakub Jasinski, Francesco Gucci, Thomas Brumme, Swaroop Palai

The initialization and control of a long-lived spin population in lead halide perovskites are prerequisites for their use in spintronic applications. Here, we demonstrate circular polarization of the interlayer exciton emission in a (BA)2PbI4/WSe2 monolayer heterostructure. The helicity of this emission is controlled by tuning the energy of the excitation la

Robert Brose, Iurii Sushch, Jonathan Mackey, Maria Arias

Early interaction of supernova blast waves with CSM has the potential to accelerate particles to PeV energies, although this has not yet been detected. Current models for this interaction assume the shock expands into a smooth stellar wind, although observations of many SNe do not support this assumption. We extend previous work by considering shocks expandi

Abdulkadyr Buchaev

The work proves that, for three-dimensional upper triangular groups over a field of odd characteristic with an abelian unipotent subgroup, the ring of invariants is polynomial if and only if the unipotent subgroup is generated by pseudoreflections or does not contain transvections.

Carles Roch I Carceller, Hanwool Lee, Jonatan Bohr Brask, Kieran Flatt

Quantum measurements under realistic conditions reveal only partial information about a system. Yet, by performing sequential measurements on the same system, additional information can be accessed. We investigate this problem in the context of semi-device-independent randomness certification using sequential maximum confidence measurements. We develop a gen

Nivar Anwer, Ezequiel López-Rubio, David Elizondo, Rafael M. Luque-Baena

The analysis and control of stochastic dynamical systems rely on probabilistic models such as (continuous-space) Markov decision processes, but large or continuous state spaces make exact analysis intractable and call for principled quantitative abstraction. This work develops a unified theory of such abstraction by integrating category theory, coalgebra, qu

Adithya Nandakumar, Ricarda S. Beckmann, Vid Irsic

This paper explores how time-varying increases in mass accretion onto rapidly spinning black holes influence their long-term spin evolution when affected by superradiance - a process where energy is extracted from the black hole by a surrounding axion field. Using simulations the study tracks how sudden accretion boosts affect a critical spin-down phase (the

Rundong Jiang, Jun Hu, Zhiyuan Xie, Yunqi Song

The growing number of wireless devices increases the need for secure network access. Radio Frequency Fingerprinting (RFF), a physical-layer authentication method, offers a promising solution as it requires no cryptography and resists spoofing. However, existing RFF approaches often lack a unified theory and effective feature extraction. Many methods use hand

Wageesha N. Manamperi, Thushara D. Abhayapala

The Relative Transfer Matrix (ReTM), recently introduced as a generalization of the relative transfer function for multiple receivers and sources, shows promising performance when applied to speech enhancement and speaker separation in noisy environments. Blindly estimating the ReTM of sound sources by exploiting the covariance matrices of multichannel recor

Linfeng Liang, Chenkai Tan, Yao Deng, Yingfeng Cai

Autonomous Driving Systems (ADS) are safety-critical, where failures can be severe. While Metamorphic Testing (MT) is effective for fault detection in ADS, existing methods rely heavily on manual effort and lack automation. We present AutoMT, a multi-agent MT framework powered by Large Language Models (LLMs) that automates the extraction of Metamorphic Relat

Zijian Feng, Hanzhang Zhou, Zixiao Zhu, Tianjiao Li

Pruning is a widely used technique to reduce the size and inference cost of large language models (LLMs), but it often causes performance degradation. To mitigate this, existing restoration methods typically employ parameter-efficient fine-tuning (PEFT), such as LoRA, to recover the pruned model's performance. However, most PEFT methods are designed for dens

Daria Perkowska, Szymon Żeberski

This paper explores the interplay between star operations, microscopic sets, and porous sets. The study focuses on the Galvin-Mycielski-Solovay theorem, which characterizes strongly measure zero sets and their interactions with meager sets. Results include the investigation of the star operation $\mathcal{F}^*$ and its properties. The paper also examines the