October 2025 arXiv papers — page 204

Md Rezanur Islam, Mahdi Sahlabadi, Keunkyoung Kim, Kangbin Yim

Security measures are essential in the automotive industry to detect intrusions in-vehicle networks. However, developing a one-size-fits-all Intrusion Detection System (IDS) is challenging because each vehicle has unique data profiles. This is due to the complex and dynamic nature of the data generated by vehicles regarding their model, driving style, test e

Hajime Moriya

We present a general definition of quantum mutual entropy for infinitely extended quantum spin and fermion lattice systems. Using this, we establish a thermal area law in these infinitely extended quantum systems. The proof is based on the local thermodynamical stability (LTS), a variational principle in terms of the conditional free energy. Our thermal area

Michał Ćwiąkała, Julia Walter, Dariusz Baran, Gabriela Wojak

This study examines the influence of various leadership styles on project efficiency across diverse organizational contexts. Using a quantitative research design, data were collected through a survey of 100 project professionals representing multiple industries, and analyzed with statistical techniques, including Spearman correlation, to explore the relation

João Henrique Inacio de Souza, Fabio Saggese, Kun Chen-Hu, Petar Popovski

This paper focuses on communication, radar search, and tracking task scheduling in multi-cell integrated sensing and communication (ISAC) networks under quality-of-service constraints. We propose a medium access control framework that multiplexes these tasks while optimizing radar scan patterns through an interference-aware scheduling algorithm. Specifically

Pedro Souza Fagundes, Thiago Castilho de Mello

The characterization of commutators in associative algebras is a classical problem in ring theory. In this paper, we address this problem for the natural class of generalized block-triangular algebras. To this end, we introduce a new invariant: the multitrace of an arbitrary element in an associative unital algebra, and prove that in a generalized block-tria

F. Lykou, J. Varga, F. Cruz-Saénz de Miera, P. Ábrahám

The mid-infrared (MIR) emitting regions of the individual protoplanetary disks in the binary system Z CMa are resolved by MATISSE/VLTI. The observations were obtained during a serendipitous large outburst of the HBe star that lasted more than 100 days, while the FUor companion is presumed to be in quiescence. The size of the MIR-emitting disk region of the m

Samuel Creedon, Volodymyr Mazorchuk

This paper discusses various aspects of the Hecke algebra combinatorics that are related to conditions appearing in K{\aa}hrstr{\"o}m's conjecture that addresses Kostant's problem for simple highest weight modules in the Bernstein-Gelfand-Gelfand category $\mathcal{O}$ for the complex Lie algebra $\mathfrak{sl}_n$. In particular, we study cyclic submodules o

Hayata Morisaki, Kaoru Sano, Seiseki Akibue

We propose two Clifford+$T$ synthesis algorithms that are optimal with respect to $T$-count. The first algorithm, called deterministic synthesis, approximates any single-qubit unitary by a single-qubit Clifford+$T$ circuit with the minimum $T$-count. The second algorithm, called probabilistic synthesis, approximates any single-qubit unitary by a probabilisti

Maria V. Sanchez-Vives, Arnau Manasanch, Andrea Pigorini, Alessandro Arena

The cerebral cortex spontaneously displays different patterns of activity that evolve over time according to the brain state. Sleep, wakefulness, resting states, and attention are examples of a wide spectrum of physiological states that can be sustained by the same structural network. Furthermore, additional states are generated by drugs (e.g., different lev

Yi-Hsin Li, Mårten Sjöström, Sebastian Knorr, Thomas Sikora

The Steered Mixture of Experts regression framework has demonstrated strong performance in image reconstruction, compression, denoising, and super-resolution. However, its high computational cost limits practical applications. This work introduces a rasterization-based optimization strategy that combines the efficiency of rasterized Gaussian kernel rendering

Boris Shoikhet

In this paper, we propose a method for constructing a colored $(d+1)$-operad $\mathbf{seq}_d$ in $\mathrm{Sets}$, in the sense of Batanin [Ba1,2], whose category of colors (=the category of unary operations) is the category $\Theta_d$, dual to the Joyal category of $d$-disks [J], [Be2,3]. For $d=1$ it is the Tamarkin $\Delta$-colored 2-operad $\mathbf{seq}$,

Anna Kasperczuk, Michał Ćwiąkała, Ernest Górka, Dariusz Baran

This study investigates the role of employee motivation as a critical factor in effective business management and explores how financial and non-financial motivators shape engagement and performance. Based on a quantitative survey of 102 employees, the research analyzes differences in motivation levels across gender, age, and work experience, as well as the

Chloé Boisson, Yannick Mogge, Aline Parreau, Théo Pierron

Generalized Tur\'an problems investigate the maximization of the number of certain structures (typically edges) under some constraints in a graph. We study a game version of these problems, the Constructor-Blocker game. We mainly focus on the case where Constructor tries to maximize the number of triangles in her graph, while forbidding her to claim short pa

Daye Lim, Tom Van Doorsselaere, Nancy Narang, Laura A. Hayes

Small scale extreme ultraviolet (EUV) transient brightenings are observationally abundant and critically important to investigate. Determining whether they share the same physical mechanisms as larger scale flares would have significant implications for the coronal heating problem. A recent study has revealed that quasi periodic pulsations (QPPs), a common f

Raghav Bongole, Amirreza Zamani, Tobias J. Oechtering, Mikael Skoglund

Minimax risk and regret focus on expectation, missing rare failures critical in safety-critical bandits and reinforcement learning. Minimax quantiles capture these tails. Three strands of prior work motivate this study: minimax-quantile bounds restricted to non-interactive estimation; unified interactive analyses that focus on expected risk rather than risk

Fabian Piper, Karl Wolf, Jonathan Heiss

Decentralized applications (dApps) in Decentralized Finance (DeFi) face a fundamental tension between regulatory compliance requirements like Know Your Customer (KYC) and maintaining decentralization and privacy. Existing permissioned DeFi solutions often fail to adequately protect private attributes of dApp users and introduce implicit trust assumptions, un

Argyrios Deligkas, Michelle Döring, Eduard Eiben, Tiger-Lily Goldsmith

We study the parameterized complexity of maximum temporal connected components (tccs) in temporal graphs, i.e., graphs that deterministically change over time. In a tcc, any pair of vertices must be able to reach each other via a time-respecting path. We consider both problems of maximum open tccs (openTCC), which allow temporal paths through vertices outsid

Pafue Christy Nganjimi, Andrew Soltan, Danielle Belgrave, Lei Clifton

Dataset condensation (DC) enables the creation of compact, privacy-preserving synthetic datasets that can match the utility of real patient records, supporting democratised access to highly regulated clinical data for developing downstream clinical models. State-of-the-art DC methods supervise synthetic data by aligning the training dynamics of models traine

John B. Etnyre

We survey what is known about various special types of submanifolds of contact manifolds and discuss their role in the development of contact geometry.

James Bailie, Ruobin Gong

The Five Safes is a framework used by national statistical offices (NSO) for assessing and managing the disclosure risk of data sharing. It can be understood as a specialization of a broader concept--contextual integrity--to the situation of statistical dissemination by an NSO. We demonstrate this by mapping the five parameters of contextual integrity onto t

Bowen Fu, Chenghan Hou, Jan Prüser

This paper proposes a structural multivariate unobserved components model with external instrument (SMUC-IV) to investigate the effects of monetary policy shocks on key U.S. macroeconomic "stars"-namely, the level of potential output, the growth rate of potential output, trend inflation, and the neutral interest rate. A key feature of our approach is the use

Josefa Caballero, Łukasz Płociniczak, Kishin Sadarangani

We study a class of nonlinear Volterra integral equations that generalize the classical capillary rise models, allowing for nonsmooth kernels and nonlinearities. To accommodate such generalities, we work in two families of function spaces: spaces with prescribed modulus of continuity and integral H\"older spaces. We establish existence results for solutions

Anne-Laure Boulesteix, Patrick Callahan, Luzia Hanssum, Vincent Gaertner

Simulations are valuable tools for empirically evaluating the properties of statistical methods and are primarily employed in methodological research to draw general conclusions about methods. In addition, they can often be useful to applied statisticians, who may rely on published simulation results to select an appropriate statistical method for their appl

Rikuto Kotoge, Yuichi Sasaki

Aligning text-to-speech (TTS) system outputs with human feedback through preference optimization has been shown to effectively improve the robustness and naturalness of language model-based TTS models. Current approaches primarily require paired desirable and undesirable samples at the utterance level. However, such pairs are often limited in TTS output data

Jacopo Bufalino, Mario Di Francesco, Agathe Blaise, Stefano Secci

Supply chain security is extremely important for modern applications running at scale in the cloud. In fact, they involve a large number of heterogeneous microservices that also include third-party software. As a result, security vulnerabilities are hard to identify and mitigate before they start being actively exploited by attackers. For this reason, govern

Mathias Thorsager, Israel Leyva-Mayorga, Petar Popovski

The traditional role of the network layer is to create an end-to-end route, through which the intermediate nodes replicate and forward the packets towards the destination. This role can be radically redefined by exploiting the power of Generative AI (GenAI) to pivot towards a prediction-based network layer, which addresses the problems of throughput limits a

Fernanda M. Baêta

A classification of upper semicontinuous, translation and dually epi-translation invariant valuations is established on the space of convex Lipschitz function on $\mathbb{R}$ with compact domain.

Rui-Heng Miao, Zhao-Di Liu, Chen-Xi Ning, Yu-Cong Hu

The energy-time uncertainty relation limits the maximum speed of quantum system evolution and is crucial for determining whether quantum tasks can be accelerated. However, multiparticle quantum speed limits have not been experimentally explored. In this work, we experimentally verify that both multiparticles and entanglement can accelerate the quantum speed

Ole Fredrik Brevig, Athanasios Kouroupis

We extend a classical theorem of Carlson on moments of Dirichlet series from $p=2$ to $1 \leq p < \infty$. When combined with the ergodic theorem for the Kronecker flow, a coherent approach to almost sure properties of vertical limit functions in $H^p$ spaces of Dirichlet series is obtained. This allows us to establish an almost sure analytic continuation of

L. De Paolis, E. Pace, C. Mazzanti, M. Morelli

Biophotons are non-thermal and non-bioluminescent ultraweak photon emissions, first hypothesised by Gurwitsch in 1924 as a regulatory mechanism in cell division, and then experimentally observed in living organisms. Today, two main hypotheses explain their origin: stochastic decay of excited molecules and coherent electromagnetic fields produced in biochemic

Daniel Marx, Ivan Gligonov, David Malsbenden, Dominik Wöll

Single fluorescent molecules, behaving as ideal electric dipole emitters, are powerful nanoscopic probes of complex optical fields. Here, this property is exploited to precisely map the polarization and vectorial structure of tightly focused laser beams, utilizing both linear and circular polarization states. The resulting three-dimensional fluorescence exci

Marc Joyeux

Bacterial genomes are partitioned into kilobases long domains that are topologically independent from each other, meaning that change of DNA superhelicity in one domain does not propagate to neighbours. This is made possible by proteins like the LacI repressor, which behave like topological barriers and block the diffusion of torsion along the DNA. Other pro

Jiawei Zheng, Gokcen Yilmaz, Ji Han, Saeema Ahmed-Kristensen

Many organisations pursue digital transformation to enhance operational efficiency, reduce manual efforts, and optimise processes by automation and digital tools. To achieve this, a comprehensive understanding of their unique needs is required. However, traditional methods, such as expert interviews, while effective, face several challenges, including schedu

Haohan Zeng, Zhenyu He, Tianxiang Zhang, Xiao Guo

Encrypted optical and acoustic meta-holograms only focus on the encrypted hologram in a single channel, viz. modulating spatial amplitude to project a holographic image. In this research, the unique concept of multi-channel amplitude-phase asymmetric-encrypted Janus acoustic meta-holograms is proposed, demonstrating remarkable capabilities of generating, enc

Nikita Pavlichenko, Iurii Nazarov, Ivan Dolgov, Ekaterina Garanina

We present the Mellum models family, open-weight code completion models designed for interactive use in JetBrains IDEs. Mellums have 4B parameters, adopt a Llama-style architecture, and are pre-trained on ~4T tokens of permissively licensed, multi-language code. Our studies show that (i) careful data curation and staged training significantly improve the mod

Panagiota Nikolaou, Freddy Gabbay, Jawad Haj-Yahya, Yiannakis Sazeides

This work aims to improve a data center's efficiency by optimizing the server upgrade plan: determine the optimal timing for replacing old servers with new ones. The opportunity presented by this approach is demonstrated through a study based on historical server data. The study establishes a significant opportunity to increase the QPS/(TCOxCO2) metric by fo

Patrick Forré, Abel Jansma

M\"obius inversion and Shapley values are two mathematical tools for characterizing and decomposing higher-order structure in complex systems. The former defines higher-order interactions as discrete derivatives over a partial order; the latter provides a principled way to attribute those interactions back to the `atomic' elements of the system. Both have fo

Henry Waldhausen, Christopher Griffin

Using techniques from information geometry, we construct a semi-Hamiltonian system modelling trader beliefs in a binary asset market and study the impact of inequality or asymmetry in beliefs, information, and power on price dynamics. We show that in a market with no inequality and $N$ completely symmetric traders, the resulting dynamics evolve on a $2N + 1$

Reinhard Wiesmayr, Lorenzo Maggi, Sebastian Cammerer, Jakob Hoydis

Adapting the modulation and coding scheme (MCS) to the wireless link quality is critical for maximizing spectral efficiency while ensuring reliability. We propose SALAD (self-adaptive link adaptation), an algorithm that exclusively leverages ACK/NACK feedback to reliably track the evolution of the signal-to-interference-plus-noise ratio (SINR), achieving hig

Anna Kasperczuk, Michał Ćwiąkała, Ernest Górka, Piotr Ręczajski

This study examines the role of work-life balance (WLB) as a strategic component of effective business management and its influence on employee motivation, job satisfaction, and organizational performance. Drawing on a quantitative survey of 102 economically active individuals, the research investigates the effectiveness of various WLB initiatives, including

I. M. De la Jara, C. Rodriguez-Opazo, D. Teney, D. Ranasinghe

Out-of-distribution (OOD) detection is essential for reliably deploying machine learning models in the wild. Yet, most methods treat large pre-trained models as monolithic encoders and rely solely on their final-layer representations for detection. We challenge this wisdom. We reveal the \textit{intermediate layers} of pre-trained models, shaped by residual

Runxi Cheng, Yuchen Guan, Yucheng Ding, Qingguo Hu

In this work, we first explore whether the parameters activated by the MoE layer remain highly sparse at inference. We perform a sparsification study on several representative MoE models. For each expert, we rank parameters by the magnitude of their activations from the gate projection and progressively prune the activated subset. Pruning up to 60% of parame

Andreas Christou, Andreas Sochopoulos, Elliot Lister, Sethu Vijayakumar

Wearable robots offer a promising solution for quantitatively monitoring gait and providing systematic, adaptive assistance to promote patient independence and improve gait. However, due to significant interpersonal and intrapersonal variability in walking patterns, it is important to design robot controllers that can adapt to the unique characteristics of e

Michael Kramer, Simon Johnston

A striking aspect of the radio profiles of many millisecond pulsars (MSPs) is that they consist of components separated from each other by regions lacking in emission. We devise a technique for determining "disjoint" from "contiguous" components and show that 35% of MSPs have disjoint components as opposed to only 3% of the slow pulsar population. We surmise

Shadi Rahimian, Mario Fritz

Single nucleotide polymorphism (SNP) datasets are fundamental to genetic studies but pose significant privacy risks when shared. The correlation of SNPs with each other makes strong adversarial attacks such as masked-value reconstruction, kin, and membership inference attacks possible. Existing privacy-preserving approaches either apply differential privacy

Selstø Sølve, Bendik Steinsvåg Dalen

In a recent publication, Dalen, and Selst{\o}, Phys. Rev. A {\bf 111}, 033116 (2025), it was demonstrated how converged photo electron spectra could be determined using a complex absorbing potential on a truncated numerical domain considerably smaller than the extension of the dynamical wave function. That approach required simulation until virtually all unb

Ovidio García-Oliva, Carsten Lemmen, Xiangyu Li, Kai Wirtz

Episodes of low dissolved oxygen concentration--hypoxia--threaten the functioning of and the services provided by aquatic ecosystems, particularly those of urban rivers. Here, we disentangle oxygen-related processes in the highly modified Elbe River flowing through the major German city of Hamburg, where low oxygen levels are frequently observed. We use a pr

Weichun Shi, Minghao Liu, Wanting Zhang, Langchen Shi

Constraint programming (CP) is a crucial technology for solving real-world constraint optimization problems (COPs), with the advantages of rich modeling semantics and high solving efficiency. Using large language models (LLMs) to generate formal modeling automatically for COPs is becoming a promising approach, which aims to build trustworthy neuro-symbolic A

Ulrich Haisch

The decay of the Higgs boson into two photons, $h \to \gamma \gamma$, is a loop-induced process within the Standard Model, predominantly mediated by loops of $W$ bosons and top quarks. While these leading contributions are well understood, the role of hadronic effects, which arise from non-perturbative QCD dynamics, has received less attention, with recent s

Carolina Carreira, Anu Aggarwal, Alejandro Cuevas, Maria José Ferreira

Understanding how cognitive biases influence adversarial decision-making is essential for developing effective cyber defenses. Capture-the-Flag (CTF) competitions provide an ecologically valid testbed to study attacker behavior at scale, simulating real-world intrusion scenarios under pressure. We analyze over 500,000 submission logs from picoCTF, a large ed

Jureeporn Yuennan, Farruh Atamurotov, Phongpichit Channuie

Recent measurements from the Atacama Cosmology Telescope (ACT), particularly when combined with DESI baryon acoustic oscillation data, have reported a scalar spectral index $n_s$ slightly higher than that inferred by {\it Planck}~2018, suggesting a mild tension with the predictions of standard inflationary attractor models. In this work, we revisit the quant

Jianbin Shen, Christy Jie Liang, Junyu Xuan

Abstractive text summarization is integral to the Big Data era, which demands advanced methods to turn voluminous and often long text data into concise but coherent and informative summaries for efficient human consumption. Despite significant progress, there is still room for improvement in various aspects. One such aspect is to improve informativeness. Hen

Robin Kimmel, Judith Michael, Andreas Wortmann, Jingxi Zhang

Digital twins promise a better understanding and use of complex systems. To this end, they represent these systems at their runtime and may interact with them to control their processes. Software engineering is a wicked challenge in which stakeholders from many domains collaborate to produce software artifacts together. In the presence of skilled software en

Peter Ochieng

We derive non-asymptotic spectral bands that bound the squared InfoNCE gradient norm via alignment, temperature, and batch spectrum, recovering the \(1/\tau^{2}\) law and closely tracking batch-mean gradients on synthetic data and ImageNet. Using effective rank \(R_{\mathrm{eff}}\) as an anisotropy proxy, we design spectrum-aware batch selection, including a

Yogesh Kumar, Susanta Samanta, Atul Gaur

MDS matrices play a critical role in the design of diffusion layers for block ciphers and hash functions due to their optimal branch number. Involutory and orthogonal MDS matrices offer additional benefits by allowing identical or nearly identical circuitry for both encryption and decryption, leading to equivalent implementation costs for both processes. The

Caucher Birkar

We construct relatively bounded toroidal and toric models of relatively bounded fibrations over curves.

Lang Qin, Zijian Gan, Xu Cao, Pengcheng Jiang

Computational drug repurposing for rare diseases is especially challenging when no prior associations exist between drugs and target diseases. Therefore, knowledge graph completion and message-passing GNNs have little reliable signal to learn and propagate, resulting in poor performance. We present RareAgent, a self-evolving multi-agent system that reframes

N. F. Allard, J. F. Kielkopf

Collision broadening by molecular hydrogen of sodium and potassium is one of the major broadening mechanisms in the atmospheres of brown dwarf stars and exoplanets at an effective temperature of about 1000K. The accurate computation of line profiles from collision broadening at high density requires use of a Fourier transform of the autocorrelation function

Kotha Kartheek, Shankar K. Ghosh, Megha Iyengar, Vinod Sharma

The impact of Radio link failure (RLF) has been largely ignored in designing handover algorithms, although RLF is a major contributor towards causing handover failure (HF). RLF can cause HF if it is detected during an ongoing handover. The objective of this work is to propose an efficient power control mechanism based on Deep Q-Network (DQN), considering han

Sedat Dogan, Nina Dethlefs, Debarati Chakraborty

Memes are a central part of online culture, yet their virality remains difficult to predict, especially in cross-lingual settings. We present a large-scale, time-series dataset of 46,578 Reddit memes collected from 25 meme-centric subreddits across eight language groups, with more than one million engagement tracking points. We propose a data-driven definiti

Gianmarco Perantoni, Lorenzo Bruzzone

Deep learning has gained broad interest in remote sensing image scene classification thanks to the effectiveness of deep neural networks in extracting the semantics from complex data. However, deep networks require large amounts of training samples to obtain good generalization capabilities and are sensitive to errors in the training labels. This is a proble

Zexin Zheng, Huangyu Dai, Lingtao Mao, Xinyu Sun

Traditional vision search, similar to search and recommendation systems, follows the multi-stage cascading architecture (MCA) paradigm to balance efficiency and conversion. Specifically, the query image undergoes feature extraction, recall, pre-ranking, and ranking stages, ultimately presenting the user with semantically similar products that meet their pref

Haoxun Li, Yu Liu, Yuqing Sun, Hanlei Shi

Recent LLM-based TTS systems achieve strong quality and zero-shot ability, but lack fine-grained emotional control due to their reliance on discrete speech tokens. Existing approaches either limit emotions to categorical labels or cannot generalize to LLM-based architectures. We propose EMORL-TTS (Fine-grained Emotion-controllable TTS with Reinforcement Lear

Jingqi Sun, Shulin He, Ruizhe Pang, Zhong-Qiu Wang

We address monaural multi-speaker-image separation in reverberant conditions, aiming at separating mixed speakers but preserving the reverberation of each speaker. A straightforward approach for this task is to directly train end-to-end DNN systems to predict the reverberant speech of each speaker based on the input mixture. Although effective, this approach

Aleksandr Lukoianov, Anssi Klapuri

Whereas chord transcription has received considerable attention during the past couple of decades, far less work has been devoted to transcribing and encoding the rhythmic patterns that occur in a song. The topic is especially relevant for instruments such as the rhythm guitar, which is typically played by strumming rhythmic patterns that repeat and vary ove

Jamal Daoudi, Chakir Tajani

This paper tackles the data completion problem related to the Helmholtz equation. The goal is to identify unknown boundary conditions on parts of the boundary that cannot be accessed directly, by making use of measurements collected from accessible regions. Such inverse problems are known to be ill-posed in the Hadamard sense, which makes finding stable and

Yuxuan Bai, Gauri Pradhan, Marlon Tobaben, Antti Honkela

With the emergence of powerful large-scale foundation models, the training paradigm is increasingly shifting from from-scratch training to transfer learning. This enables high utility training with small, domain-specific datasets typical in sensitive applications. Membership inference attacks (MIAs) provide an empirical estimate of the privacy leakage by mac

Yike Wu, Yiwei Wang, Yujun Cai

While Large Vision-Language Models (LVLMs) achieve strong performance in multimodal tasks, hallucinations continue to hinder their reliability. Among the three categories of hallucinations, which include object, attribute, and relation, relation hallucinations account for the largest proportion but have received the least attention. To address this issue, we

Yongxuan Lyu, Guangfeng Jiang, Hongsi Liu, Jun Liu

The manual annotation of outdoor LiDAR point clouds for instance segmentation is extremely costly and time-consuming. Current methods attempt to reduce this burden but still rely on some form of human labeling. To completely eliminate this dependency, we introduce ALISE, a novel framework that performs LiDAR instance segmentation without any annotations. The

Jeffrey N. Clark, Elena Fillola, Nawid Keshtmand, Raul Santos-Rodriguez

Monitoring greenhouse gas emissions and evaluating national inventories require efficient, scalable, and reliable inference methods. Top-down approaches, combined with recent advances in satellite observations, provide new opportunities to evaluate emissions at continental and global scales. However, transport models used in these methods remain a key source

Yu Liu, Hanlei Shi, Haoxun Li, Yuqing Sun

Emotion Recognition in Conversations (ERC) is hard because discriminative evidence is sparse, localized, and often asynchronous across modalities. We center ERC on emotion hotspots and present a unified model that detects per-utterance hotspots in text, audio, and video, fuses them with global features via Hotspot-Gated Fusion, and aligns modalities using a

Haoxun Li, Yuqing Sun, Hanlei Shi, Yu Liu

Continuous dimensional speech emotion recognition captures affective variation along valence, arousal, and dominance, providing finer-grained representations than categorical approaches. Yet most multimodal methods rely solely on global transcripts, leading to two limitations: (1) all words are treated equally, overlooking that emphasis on different parts of

Nouar Aldahoul, Yasir Zaki

The rapid spread of misinformation on digital platforms threatens public discourse, emotional stability, and decision-making. While prior work has explored various adversarial attacks in misinformation detection, the specific transformations examined in this paper have not been systematically studied. In particular, we investigate language-switching across E

Hachem Madmoun, Salem Lahlou

Eliciting cooperation in multi-agent LLM systems is critical for AI alignment. We investigate two approaches: direct communication and curriculum learning. In a 4-player Stag Hunt, a one-word "cheap talk" channel increases cooperation from 0% to 96.7%, demonstrating communication as a robust coordination mechanism. In contrast, we find that curriculum learni

Jiahao Ma, Hongzong Li, Ye-Fan Hu, Jian-Dong Huang

Physicochemically informed biological sequence generation has the potential to accelerate computer-aided cellular therapy, yet current models fail to \emph{jointly} ensure novelty, diversity, and biophysical plausibility when designing variable regions of T-cell receptors (TCRs). We present \textbf{PhysicoGPTCR}, a large generative protein Transformer that i

Elena Litvinova

Nuclear resonances provide a rich and versatile testbed for exploring fundamental aspects of physics, particularly within the domain of strongly correlated many-body systems. The overarching goal of the theory is to develop a consistent and predictive framework that is (i) capable of a spectroscopically accurate description and (ii) sufficiently general to b

Liza Fretel, Baptiste Cecconi, Laura Debisschop

This ongoing work focuses on the development of a methodology for generating a multi-source mapping of astronomical observation facilities. To compare two entities, we compute scores with adaptable criteria and Natural Language Processing (NLP) techniques (Bag-of-Words approaches, sequential approaches, and surface approaches) to map entities extracted from

Yanwei Huang, Wesley Hanwen Deng, Sijia Xiao, Motahhare Eslami

Despite their increasing capabilities, text-to-image generative AI systems are known to produce biased, offensive, and otherwise problematic outputs. While recent advancements have supported testing and auditing of generative AI, existing auditing methods still face challenges in supporting effectively explore the vast space of AI-generated outputs in a stru

Vittorio Latorre

We examine the duality theory for a class of non-convex functions obtained by composing a convex function with a continuous one. Using Fenchel duality, we derive a dual problem that satisfies weak duality under general assumptions. To better understand this duality, we compare it with classical Lagrange duality by analyzing a related, yet more complex, const

Amirtaha Amanzadi, Zahra Dehghanian, Hamid Beigy, Hamid R. Rabiee

The rapid development of generative models has made it increasingly crucial to develop detectors that can reliably detect synthetic images. Although most of the work has now focused on cross-generator generalization, we argue that this viewpoint is too limited. Detecting synthetic images involves another equally important challenge: generalization across vis

Jiechen Zhang

We establish explicit, universal, and distribution-free bounds for the $n$-th cumulant, $\kappa_n(X)$, of a scalar random variable, controlled solely by an $n$-th order absolute moment functional $M_n(X)$. The bounds take the form $\lvert\kappa_n(X)\rvert \le C_n M_n(X)$. Our principal contribution is the derivation of coefficients satisfying $C_n \sim (n-1)

Ritesh Chandra, Sonali Agarwal, Navjot Singh, Sadhana Tiwari

Exponential growth in heterogeneous healthcare data arising from electronic health records (EHRs), medical imaging, wearable sensors, and biomedical research has accelerated the adoption of data lakes and centralized architectures capable of handling the Volume, Variety, and Velocity of Big Data for advanced analytics. However, without effective governance,

Adi Zilka

We study the zeros of modular forms in the Miller basis, a natural basis for the space of modular forms. We show that the zeros of their Faber polynomials have linear moments. By analyzing the moments we can extend the known range of the forms in the Miller basis for which at least one of the zeros is not on the arc - the circular part of the boundary of the

Abhay Mehta, Dan Parsons, Tim Lukas Holch, David Berge

The identification of $\gamma$-rays from the predominant hadronic-background is a key aspect in their ground-based detection using Imaging Atmospheric Cherenkov Telescopes (IACTs). While current methods are limited in their ability to exploit correlations in complex data, deep learning-based models offer a promising alternative by directly leveraging image-l

Manizheh Botshekananfard, Elif Büşra Güraksın, Gizem Şengör

We identify raising and lowering operators of the de Sitter algebra with focus on their action on states in particular in 4 spacetime dimensions. There isn't a unique solution to the question of how the de Sitter ladder operators act on states. By fixing the action of certain generators one can conclude on the action of the rest. Our main aim is to be able t

Romain Veyron, Clément Métayer, Jean-Baptiste Gérent, Ruiyang Huang

In this paper, we present an all-optical method to produce shell-shaped traps for ultracold atoms in microgravity. Our scheme exploits optical double dressing of the ground state to create a short range strongly repulsive central potential barrier. Combined with a long range attractive central potential, this barrier forms the shell trap. We demonstrate that

Zijun Jia, Shuang Liang, Jinsong Yu

Industrial fault diagnosis faces the dual challenges of data scarcity and the difficulty of deploying large AI models in resource-constrained environments. This paper introduces Syn-Diag, a novel cloud-edge synergistic framework that leverages Large Language Models to overcome these limitations in few-shot fault diagnosis. Syn-Diag is built on a three-tiered

Gonzalo Cao-Labora, Antonio J. Fernández

We show the existence of a family of nontrivial smooth contractible domains on the sphere that admit Neumann eigenfunctions of the Laplacian which are constant on the boundary. These domains are contained on the half-sphere, in stark contrast with the rigidity literature for Serrin-type problems. The proof relies on a local bifurcation argument around the fa

Rémi Delaunay, Christoph Hennersperger, Stefan Wörz

Ultrafast ultrasound imaging enables visualization of rapid physiological dynamics by acquiring data at exceptionally high frame rates. However, this speed often comes at the cost of spatial resolution and image quality due to unfocused wave transmissions and associated artifacts. In this work, we propose a novel modulated Implicit Neural Representation (INR

Aritra Lahiri, Juan Carlos Cuevas, Björn Trauzettel

The Higgs mode, originally proposed in the context of superconductivity, corresponds to oscillations of the amplitude of the superconducting order parameter. Recent THz-domain optical studies have found signatures consistent with the Higgs mode, but its unambiguous detection is still challenging. We predict that the existence of the Higgs mode can be unambig

Marc Kaufeld, Johannes Betz

This research introduces two efficient methods to estimate the collision risk of planned trajectories in autonomous driving under uncertain driving conditions. Deterministic collision checks of planned trajectories are often inaccurate or overly conservative, as noisy perception, localization errors, and uncertain predictions of other traffic participants in

Satori Tsuzuki

Predicting when a three-dimensional turbulent flow reaches its dissipation peak is essential for both theory and adaptive algorithms in simulations and experiments. Using direct numerical simulations (DNSs) of the Taylor--Green vortex (TGV) at resolutions of $256^3$--$1024^3$, we introduce and test a small-scale weighted diagnostic: the spectrum of $|\nabla

L. Foffano, C. Arcaro, A. Arbet-Engels, F. D'Ammando

Extremely high-peaked BL Lac objects (or extreme blazars) are unique extragalactic laboratories where particle acceleration processes are pushed at their physical limits. In these blazars, synchrotron emission peaking above keV energies is reprocessed to very-high-energy (VHE, energies > 100 GeV) gamma rays, often resulting in very hard TeV spectra. Over the

S. P. Maydanyuk, Ju-Jun Xie, V. S. Vasilevsky, K. A. Shaulskyi

Purpose To investigate synthesis of nuclei in pycnonuclear reactions in dense medium of neutron stars on the basis of understanding, how the compound nucleus is formed during collision of two nuclei. To implement microscopic formulation of nuclear interactions and fusion in pycnonuclear reactions in dense medium. Methods (1) Nuclei synthesis in pycnonuclear

Chunsan Hong, Seonho An, Min-Soo Kim, Jong Chul Ye

Masked diffusion models (MDMs) have recently emerged as a novel framework for language modeling. MDMs generate sentences by iteratively denoising masked sequences, filling in [MASK] tokens step by step. Although MDMs support any-order sampling, performance is highly sensitive to the choice of which position to unmask next. Prior work typically relies on rule

Tung H. Nguyen

We obtain some $d\ge2$ such that every graph $G$ with no induced copy of the five-vertex path $P_5$ has at most $\alpha(G)\omega(G)^d$ vertices. This ``off-diagonal Ramsey'' statement implies that every such graph $G$ has fractional chromatic number at most $\omega(G)^d$, and is another step towards the polynomial Gy\'arf\'as-Sumner conjecture for $P_5$. The

J. A. Gracey

Using the recent six loop renormalization group functions for Lee-Yang and percolation theory constructed by Schnetz from a scalar cubic Lagrangian, we deduce the $\epsilon$ expansion of the critical exponents for both cases. Estimates for the exponents in three, four and five dimensions are extracted using two-sided Pad\'{e} approximants and shown to be com

Jiaojiao Ye, Jiaxing Zhong, Qian Xie, Yuzhou Zhou

Generating enough and diverse data through augmentation offers an efficient solution to the time-consuming and labour-intensive process of collecting and annotating pixel-wise images. Traditional data augmentation techniques often face challenges in manipulating high-level semantic attributes, such as materials and textures. In contrast, diffusion models off

Raffaele Mura, Giorgio Piras, Kamilė Lukošiūtė, Maura Pintor

Jailbreaks are adversarial attacks designed to bypass the built-in safety mechanisms of large language models. Automated jailbreaks typically optimize an adversarial suffix or adapt long prompt templates by forcing the model to generate the initial part of a restricted or harmful response. In this work, we show that existing jailbreak attacks that leverage s

Derek Marsh

Progress of the COVID-19 pandemic was quantified, in the first instance, using the daily number of positive cases recorded by the national public health authorities. Averaged over a seven-day window, the daily incidence of COVID-19 in Germany reveals clear sections of exponential growth or decay in propagation of infection. Comparing with incidence profiles