November 2025 arXiv papers — page 52

Christian Döding, Patrick Henning

In this review article, we provide an overview of recent advances in the numerical approximation of minimizers of the Ginzburg-Landau energy in multiscale spaces. Such minimizers represent the most stable states of type-II superconductors and, for large material parameters $\kappa$, capture the formation of lattices of quantized vortices. As the vortex cores

Manuel Valle Torre, Marcus Specht, Catharine Oertel

This study uses a Design-Based Research (DBR) cycle to refine the integration of Large Language Models (LLMs) in high school programming education. The initial problem was identified in an Intervention Group where, in an unguided setting, a higher proportion of executive, solution-seeking queries correlated strongly and negatively with exam performance. A co

Francesco G. Celiberto, André V. Giannini, Victor P. Gonçalves, Yuri N. Lima

In this paper, we investigate the production of a fully charmed tetraquark state $T_{4c}$ in $pp$ collisions at forward rapidities through the fragmentation mechanism considering the Color Glass Condensate (CGC) formalism and the solution of the running coupling Balitsky-Kovchegov (BK) equation. The contributions of gluon - and charm - initiated processes ar

Ching-Yi Lai, Chih-Yu Jian, Pei-Cheng Chuang, Chia-Ming Lee

In deepfake detection, the varying degrees of compression employed by social media platforms pose significant challenges for model generalization and reliability. Although existing methods have progressed from single-modal to multimodal approaches, they face critical limitations: single-modal methods struggle with feature degradation under data compression i

Raz Kupferman, Cy Maor, David Padilla-Garza

We study isometric immersions of a Riemannian surface $(\Omega,\frak{g})$, where $\Omega \subset \mathbb{R}^2$, into $\mathbb{R}^3$. We consider their bending energy, i.e., the square of the $L^2$-norm of their second fundamental form, which is equivalent to the Willmore functional. We obtain two new lower bounds for this energy, one in terms of the Gaussian

L. N. A. van Haastere, J. Varga, M. R. Hogerheijde, C. Dominik

The inner regions of planet-forming disks hold invaluable insights for our understanding of planet formation. The disk around the Herbig star HD 100453 presents one such environment, with an inner disk that is significantly misaligned with respect to the outer disk. This paper expands the existing H-band (PIONIER) and K-band (GRAVITY) interferometric studies

S. Rodionov, A. Burguete-Lopez, M. Makarenko, Q. Wang

Foundation models (FM) are transforming artificial intelligence by enabling generalizable, data-efficient solutions across different domains for a broad range of applications. However, the lack of large and diverse datasets limits the development of FM in nanophotonics. This work presents MOCLIP (Metasurface Optics Contrastive Learning Pretrained), a nanopho

Can Lei, Hayat Rajani, Nuno Gracias, Rafael Garcia

Side-scan sonar (SSS) imagery is widely used for seafloor mapping and underwater remote sensing, yet the measured intensity is strongly influenced by seabed reflectivity, terrain elevation, and acoustic path loss. This entanglement makes the imagery highly view-dependent and reduces the robustness of downstream analysis. In this letter, we present PhysDNet,

H. R. Paz

Engineering programmes, particularly in Latin America, are often governed by rigid curricula and strict regularity rules that are claimed to create a Regularity Trap for capable students. This study tests that causal hypothesis using the CAPIRE framework, a leakage-aware pipeline that integrates curriculum topology and causal estimation. Using longitudinal d

Santiago Moreno, Pablo Meseguer, Rocío del Amor, Valery Naranjo

Accurate annotation of cutaneous neoplasm biopsies represents a major challenge due to their wide morphological variability, overlapping histological patterns, and the subtle distinctions between benign and malignant lesions. Vision-language foundation models (VLMs), pre-trained on paired image-text corpora, learn joint representations that bridge visual fea

Xin Yuan, Siqi Li, Jiateng Wei, Chengrui Zhu

Pruning is an effective method for compressing Large Language Models, but finding an optimal, non-uniform layer-wise sparsity allocation remains a key challenge. While heuristic methods are fast but yield suboptimal performance, more powerful search-based approaches like Reinforcement Learning are often hindered by prohibitive computational costs on large-sc

Huaming Ling, Ying Wang, Si Chen, Junfeng Fan

We address two fundamental challenges in adapting general deep CNNs for FHE-based inference: approximating non-linear activations such as ReLU with low-degree polynomials while minimizing accuracy degradation, and overcoming the ciphertext capacity barrier that constrains high-resolution image processing on FHE inference. Our contributions are twofold: (1) a

Claudia M. Raiteri

With their jet pointing towards us, blazars are ideal tools to study the physics and structure of extragalactic jets. Their powerful jets are cosmic particle accelerators and are alleged to be one of the production sites of the high-energy neutrinos detected by the IceCube Observatory. Doppler beaming of the jet nonthermal radiation increases blazar brightne

Gesa-R. Siemann, Davide Curcio, Anders S. Mortensen, Charlotte E. Sanders

Optical control offers a compelling route for tailoring material properties on an ultrafast time scale. Ordered states such as charge density waves (CDWs) can be transiently melted by an ultrafast light excitation. This is also the case for the rare-earth tritelluride LaTe$_3$, a prototypical CDW compound. For this material it has recently been reported that

Michal Molnár, Zbyněk Šír, Jana Vráblíková

In this paper, we develop a new and efficient approach to the computation of envelope surfaces. We interpret one-parameter systems of surfaces as curves in the homogeneous spaces of suitable Lie groups. Using the formalism of Lie groups and Lie algebras, we rigorously capture the inherent symmetry and linearity in the computation of envelopes. In particular,

V. I. Bondarenko, M. H. Urin

The Landau-Migdal parameter and quenching factor for the Gamow-Teller strength distributions in the 48Ca, 90Zr, 132Sn and 208Pb parent nuclei are deduced from a detailed comparison of experimental distributions (obtained for the \b{eta}^((-)) -channel) with respective strength functions evaluated within the semi-microscopic Particle-Hole Dispersive Optical M

Tommaso Ruggeri, Ferdinand Thein, Qinghua Xiao

We consider self-similar, radially symmetric solutions of the relativistic Euler equations with constitutive relations from relativistic kinetic theory, based on Synge energies for monatomic and its extension to diatomic gases. For the corresponding initial--boundary value problem, including the spherical piston problem, we prove existence and uniqueness of

Avijit Maity, Vikram Tripathi, Andriy H. Nevidomskyy

We extend non-invertible duality concepts from one-dimensional systems to two spatial dimensions by constructing a web of non-invertible dualities for lattice models with subsystem symmetries. For the $\mathbb{Z}_2 \times \mathbb{Z}_2$ subsystem symmetry on the square lattice, we build two complementary dualities: a map that sends spontaneous subsystem symme

Remi Petitpierre

This thesis presents methods and datasets to investigate cartographic heritage on a large scale and from a cultural perspective. Heritage institutions worldwide have digitized more than one million maps, and automated techniques now enable large-scale recognition and extraction of map content. Yet these methods have engaged little with the history of cartogr

Dmitry Kochetkov

The integration of generative artificial intelligence (GenAI) and large language models (LLMs) into scientific research and higher education presents a paradigm shift, offering revolutionizing opportunities while simultaneously raising profound ethical, legal, and regulatory questions. This study examines the complex intersection of AI and science, with a sp

Bhuvan Sachdeva, Sneha Kumari, Rudransh Agarwal, Shalaka Kumaraswamy

Cataract surgery is one of the most commonly performed surgeries worldwide, yet intraoperative complications such as iris prolapse, posterior capsule rupture (PCR), and vitreous loss remain major causes of adverse outcomes. Automated detection of such events could enable early warning systems and objective training feedback. In this work, we propose Cataract

Dan Dai, Chenhao Lu

In this paper, we prove an optimal global rigidity estimate for the eigenvalues of the Jacobi unitary ensemble. Our approach begins by constructing a random measure defined through the eigenvalue counting function. We then prove its convergence to a Gaussian multiplicative chaos measure, which leads to the desired rigidity result. To establish this convergen

Muhammad Usman Shahid, Chuadhry Mujeeb Ahmed, Rajiv Ranjan

The security of code generated by large language models (LLMs) is a significant concern, as studies indicate that such code often contains vulnerabilities and lacks essential defensive programming constructs. This work focuses on examining and evaluating the security of LLM-generated code, particularly in the context of C/C++. We categorized known vulnerabil

Susanne Hindennach, Mayar Elfares, Céline Gressel, Andreas Bulling

The proliferation of eye tracking in high-stakes domains - such as healthcare, marketing and surveillance - underscores the need for researchers to be ethically aware when employing this technology. Although privacy and ethical guidelines have emerged in recent years, empirical research on how scholars reflect on their own work remains scarce. To address thi

Antonia Wüst, Wolfgang Stammer, Hikaru Shindo, Lukas Helff

Vision-Language models (VLMs) achieve strong performance on multimodal tasks but often fail at systematic visual reasoning tasks, leading to inconsistent or illogical outputs. Neuro-symbolic methods promise to address this by inducing interpretable logical rules, though they exploit rigid, domain-specific perception modules. We propose Vision-Language Progra

Adéla Hladká, Patrícia Martinková

This study introduces a novel nonparametric approach for detecting Differential Item Functioning (DIF) in binary items through direct comparison of Item Response Curves (IRCs). Building on prior work on nonparametric comparison of regression curves, we extend the methodology to accommodate binary response data, which is typical in psychometric applications.

Muhao Guo, Yang Weng

The rapid expansion of distributed photovoltaic (PV) systems poses challenges for power grid management, as many installations remain undocumented. While satellite imagery provides global coverage, traditional computer vision (CV) models such as CNNs and U-Nets require extensive labeled data and fail to generalize across regions. This study investigates the

Sid Assawaworrarit, Alex Song, Shanhui Fan

Control of thermal emission is important in a number of applications from thermal energy harvesting and management and sensing of gas and chemical to thermal camouflage. Semiconductor-based devices can be engineered to enable electrical control of thermal emission, offering high modulation speed and ease of voltage control. Existing device designs for modula

Jana Vráblíková, Bert Jüttler

Computing the envelope of deforming planar domains is a significant and challenging problem with a wide range of potential applications. We approximate the envelope using circular arc splines, curves that balance geometric flexibility and computational simplicity. Our approach combines two concepts to achieve these benefits. First, we represent a planar doma

Qisen Chai, Yansong Wang, Junjie Huang, Tao Jia

As graph-structured data grow increasingly large, evaluating their robustness under adversarial attacks becomes computationally expensive and difficult to scale. To address this challenge, we propose to compress graphs into compact representations that preserve both topological structure and robustness profile, enabling efficient and reliable evaluation. We

Jianhao Zeng, Yancheng Bai, Ruidong Chen, Xuanpu Zhang

Video virtual try-on technology provides a cost-effective solution for creating marketing videos in fashion e-commerce. However, its practical adoption is hindered by two critical limitations. First, the reliance on a single garment image as input in current virtual try-on datasets limits the accurate capture of realistic texture details. Second, most existi

Louis Richard, Anton V. Artemyev, Cecilia Norgren, Xin An

In collisionless plasmas, relaxation of the deviations of ion velocity distribution functions (VDFs) from local thermodynamic equilibrium occurs through particle interactions with electromagnetic fields. In particular, in the Earth's magnetotail, the deviations of the ion VDFs, typically consisting of multiple components, from the equilibrium must be limited

Yixin Wu, Rui Wen, Chi Cui, Michael Backes

Inference attacks have been widely studied and offer a systematic risk assessment of ML services; however, their implementation and the attack parameters for optimal estimation are challenging for non-experts. The emergence of advanced large language models presents a promising yet largely unexplored opportunity to develop autonomous agents as inference atta

Wouter W. L. Nuijten, Mykola Lukashchuk

Automated decision-making under uncertainty requires balancing exploitation and exploration. Classical methods treat these separately using heuristics, while Active Inference unifies them through Expected Free Energy (EFE) minimization. However, EFE minimization is computationally expensive, limiting scalability. We build on recent theory recasting EFE minim

Atef Lechiheb

This paper establishes a comprehensive theory of geometric rough paths for mixed fractional Brownian motion (MFBM) and its generalized multi-component extensions. We prove that for a generalized MFBM of the form $M_t^H(a) = \sum_{k=1}^N a_k B_t^{H_k}$ with $\min\{H_k\} > \frac{1}{4}$, there exists a canonical geometric rough path obtained as the limit of smo

Vasudevarao Allu, Raju Biswas, Rajib Mandal

In this paper, we investigates the Bohr phenomenon for holomorphic mappings $F$ from the unit ball $\mathbb{B}_X$ of a complex Banach space $X$ into the closure of the unit polydisc $\mathbb{D}^m$ within the space $\mathbb{C}^m$. First, we prove an improved Bohr inequality involving the squared norms of the mapping and its homogeneous expansions. Second, we

Hui Gao

The celebrated Nash-Williams and Tutte's theorem states that a graph $G=(V, E)$ contains $k$ edge disjoint spanning trees if and only if $\nu_{f}(G) \geq k$, where $$\nu_{f}(G):=\min_{|\mathcal{\mathcal{P}}|>1, \text{$\mathcal{P}$ is a partition of $V(G)$}}\frac{|E( \mathcal{P})|}{|\mathcal{P}|-1}.$$ Inspired by the NDT theorem as structural explanations for

Pablo Tuero, César González-Ruano, Igor Žutić, Yuan Lu

Over the recent years, crossroads of magnetism and superconductivity led to the emerging field of superconducting spintronics. A cornerstone of this venture is the generation of equal-spin triplet Cooper pairs in superconductor-ferromagnet hybrids, enabling long-range spin-polarized supercurrents and magnetic control over superconducting quantum states for t

Juntao Gao, Feiyang Ye, Jing Zhang, Wenjing Qian

Vision-Language-Action (VLA) models have emerged as a powerful paradigm in Embodied AI. However, the significant computational overhead of processing redundant visual tokens remains a critical bottleneck for real-time robotic deployment. While standard token pruning techniques can alleviate this, these task-agnostic methods struggle to preserve task-critical

Tsubasa Masumura, Masato Taki

Associative memory models based on Hopfield networks and self-attention based on key-value mechanisms have been popular approaches in the study of memory mechanisms in deep learning. It has been pointed out that the state update rule of the modern Hopfield network (MHN) in the adiabatic approximation is in agreement with the self-attention layer of Transform

Alejandro Claros

In this note, we study a quantitative extension of the John-Nirenberg inequality for the Hardy-Littlewood maximal function of a $\operatorname{BMO}$ function. More precisely, for every nonconstant locally integrable function $f$ such that $Mf$ is not identically infinite, we prove the inequality \begin{equation*} \left( \frac{1}{w(Q)}\int_Q \left( \frac{Mf(x

Ruilin Bai, Bo Chen

X-chromosome association study has specific model uncertainty challenges, such as unknown X-chromosome inactivation status and baseline allele, and considering nonadditive and gene-sex interaction effects in the analysis or not. Although these challenges have been answered for single-locus X-chromosome variants, it remains unclear how to properly perform mul

Raphaël Maire, Andrea Plati, Frank Smallenburg, Giuseppe Foffi

While equilibrium interfaces display universal large-scale statistics, interfaces in phase-separated active and driven systems are predicted to belong to distinct non-equilibrium universality classes. Yet, such behavior has proven difficult to observe, with most systems exhibiting equilibrium-like fluctuations despite their strongly non-equilibrium microscop

José Teixeira, Pascal Klöckner, Diana Montezuma, Melis Erdal Cesur

In addition to evaluating tumor morphology using H&E staining, immunohistochemistry is used to assess the presence of specific proteins within the tissue. However, this is a costly and labor-intensive technique, for which virtual staining, as an image-to-image translation task, offers a promising alternative. Although recent, this is an emerging field of res

Ilán Carretero, Roshni Mahtani, Silvia Perez-Deben, José Francisco González-Muñoz

Accurate diagnosis of spitzoid tumors (ST) is critical to ensure a favorable prognosis and to avoid both under- and over-treatment. Epigenetic data, particularly DNA methylation, provide a valuable source of information for this task. However, prior studies assume complete data, an unrealistic setting as methylation profiles frequently contain missing entrie

Juan A. Crespo, Armajac Raventós-Pujol

We revisit Esteban and Ray's (1994) seminal model of polarization. Their main result (unnecessarily) relies on the assumption that individuals are infinitely divisible, which imposes strong restrictions on admissible polarization indices. We show that relaxing this assumption yields a broader family of indices consistent with the original axioms. The resulti

Paola Pia Foligno, Daniele Boffi, Fabio Credali, Riccardo Vescovini

Standard Virtual Element Methods (VEM) are based on polynomial projections and require a stabilization term to evaluate the contribution of the non-polynomial component of the discrete space. However, the stabilization term is not uniquely defined by the underlying variational formulation and is typically introduced in an ad hoc manner, potentially affecting

Zong-Wei Hong, Jing-lun Li, Lin-Ze Li, Shen Zhang

Flow Matching (FM) has recently emerged as a principled and efficient alternative to diffusion models. Standard FM encourages the learned velocity field to follow a target direction; however, it may accumulate errors along the trajectory and drive samples off the data manifold, leading to perceptual degradation, especially in lightweight or low-step configur

Saboura Yousefi, Azam Kaheni, Farangis Johari

In this paper, we investigate nilpotent Lie algebras $ L $ of nilpotency class $3 $ and provide a complete classification of those satisfying $ \dim L^2 = 3 $ and $Z(L) = L^3 \cong A(2). $ Furthermore, we explicitly characterize the structure of such Lie algebras in the case when $ \dim L = 7. $

Jean-Louis Le Mouël, Vincent Courtillot, Vladimir Kossobokov, Dominique Gibert

Whether planetary motions influence the solar magnetic cycle has remained an open question due to the lack of a rigorous physical mechanism. Here we develop a Lagrangian framework based on the Virial theorem to show that planetary orbital angular momentum modulates the Sun's rotation through barycentric dynamics, consistent with angular momentum conservation

Francois Vandenhende, Anna Georgiou, Michalis Georgiou, Theodoros Psaras

We present a graphical, knowledge-based method for reviewing treatment-emergent adverse events (AEs) in clinical trials. The approach enhances MedDRA by adding a hidden medical knowledge layer (Safeterm) that captures semantic relationships between terms in a 2-D map. Using this layer, AE Preferred Terms can be regrouped automatically into similarity cluster

Santhosh G S, Saurav Prakash, Balaraman Ravindran

Large Language Models (LLMs) face a significant bottleneck during autoregressive inference due to the massive memory footprint of the Key-Value (KV) cache. Existing compression techniques like token eviction, quantization, or other low-rank methods often risk information loss, have fixed limits, or introduce significant computational overhead from explicit d

Peng-Cheng Lu, Zong-Guo Si, Han Zhang, Xin-Yi Zhang

We study the $\gamma\gamma \to \tau^+\tau^-$ process including full $\tau^+ \tau^-$ spin information in Pb--Pb ultraperipheral collision and at lepton colliders. We present the predictions for the corresponding cross sections and spin correlations at NLO electroweak precision, and find that the NLO electroweak contributions are numerically small for the obse

Yuchen Ji, Bo Xu, Jie Shi, Jiaqing Liang

The task of translating natural language questions into query languages has long been a central focus in semantic parsing. Recent advancements in Large Language Models (LLMs) have significantly accelerated progress in this field. However, existing studies typically focus on a single query language, resulting in methods with limited generalizability across di

Ryan Wong, Hosea David Yu Fei Ng, Dhananjai Sharma, Glenn Jun Jie Ng

Large Language Models (LLMs) remain susceptible to jailbreak exploits that bypass safety filters and induce harmful or unethical behavior. This work presents a systematic taxonomy of existing jailbreak defenses across prompt-level, model-level, and training-time interventions, followed by three proposed defense strategies. First, a Prompt-Level Defense Frame

Henning Bostelmann, Daniela Cadamuro, Leonardo Sangaletti

In thermal quantum field theory, the global Liouvillian (the generator of time translations) is passive. How is this reflected in the properties of its local density, a quantum field? We propose that the locally averaged density is bounded below, but not above, with respect to the noncommutative $L^4$ norm. This is analogous to the known quantum energy inequ

Sahil Kale

Modern large language models increasingly integrate internal web-based retrieval to provide real-time answers, yet it remains unclear how effectively these systems identify information need, trigger retrieval, and use retrieved evidence. To understand these parameters better, we evaluate the necessity and effectiveness of internal web search through an exter

Salah Eddine Choutri, Prajwal Chauhan, Othmane Mazhar, Saif Eddin Jabari

The Monte Carlo-type Neural Operator (MCNO) introduces a lightweight architecture for learning solution operators for parametric PDEs by directly approximating the kernel integral using a Monte Carlo approach. Unlike Fourier Neural Operators, MCNO makes no spectral or translation-invariance assumptions. The kernel is represented as a learnable tensor over a

Zijian Song, Xiaoxin Lin, Tao Pu, Zhenlong Yuan

Recent progress in robotics and embodied AI is largely driven by Large Multimodal Models (LMMs). However, a key challenge remains underexplored: how can we advance LMMs to discover tasks that assist humans in open-future scenarios, where human intentions are highly concurrent and dynamic. In this work, we formalize the problem of Human-centric Open-future Ta

Szilvia Homolya, Jenő Szigeti

An nxn matrix A over an arbitrary unitary ring R satisfies invariant left and right Cayley-Hamilton identities with matrix coefficients C(i), D(i) having commutator sum entries. If R has a grading similar to the case of Grassmann algebras, then we prove that C(i)-D(i)-AC(i+1)+D(i+1)A=-2p(i+1)A1 for all i, where A1 and p(i+1) are the odd components of A and o

Keming Shen, Bizhu Wu, Junliang Chen, Xiaoqin Wang

Recent works have sought to enhance the controllability and precision of text-driven motion generation. Some approaches leverage large language models (LLMs) to produce more detailed texts, while others incorporate global 3D coordinate sequences as additional control signals. However, the former often introduces misaligned details and lacks explicit temporal

Haifeng Jing, Yujie Hou, Junfei Liu, Rui Xie

With the rapid development of Large Language Models, dialogue systems are shifting from information tools to emotional companions, heralding the era of Emotional Companionship Dialogue Systems (ECDs) that provide personalized emotional support for users. However, the field lacks clear definitions and systematic evaluation standards for ECDs. To address this,

Yash Mali, Evan Shelhamer

Test-time adaptation (TTA) updates models during inference to reduce error on distribution shifts. While entropy minimization over the output distribution has proven effective as a TTA loss, we study using the intermediate distributions computed by transformers in the attention mechanism. We propose LookSharp, which minimizes the entropy of CLS-to-patch atte

Arina Kharlamova, Jiawen Liu, Tianyi Zhang, Xinrui Yang

Linux kernel evolution breaks drivers through API/ABI changes, semantic shifts, and security-hardening updates. We introduce DRIVEBENCH, an executable corpus of kernel$\rightarrow$driver co-evolution cases, and AUTODRIVER, a closed-loop, LLM-driven system for automating driver maintenance. The system integrates prompt engineering, multi-agent collaboration,

Marco Cirant, Nicolò De Bernardi

We study the local stability properties of solutions to ergodic and discounted mean field games systems, as the time horizon $T \to +\infty$, around stationary equilibria, when the Hamiltonian is quadratic. We replace the usual monotonicity of the coupling term with a weaker, local assumption on the stationary equilibrium (that need not be unique), stemming

Zhenxing Mi, Yuxin Wang, Dan Xu

We present One4D, a unified framework for 4D generation and reconstruction that produces dynamic 4D content as synchronized RGB frames and pointmaps. By consistently handling varying sparsities of conditioning frames through a Unified Masked Conditioning (UMC) mechanism, One4D can seamlessly transition between 4D generation from a single image, 4D reconstruc

Juncheng Li, Yige Li, Hanxun Huang, Yunhao Chen

Backdoor attacks undermine the reliability and trustworthiness of machine learning systems by injecting hidden behaviors that can be maliciously activated at inference time. While such threats have been extensively studied in unimodal settings, their impact on multimodal foundation models, particularly vision-language models (VLMs), remains largely underexpl

Wenhao Xu, Xin Dong, Yue Li, Haoyuan Shi

Video large language models have demonstrated strong video understanding capabilities but suffer from high inference costs due to the massive number of tokens in long videos. Inspired by event-based vision, we propose an event-guided, training-free framework for efficient spatio-temporal understanding, named EventSTU. In the temporal domain, we design a coar

Ruiying Liu, Yuanzhi Liang, Haibin Huang, Tianshu Yu

Group Relative Policy Optimization (GRPO) has emerged as an effective and lightweight framework for post-training visual generative models. However, its performance is fundamentally limited by the ambiguity of textual visual correspondence: a single prompt may validly describe diverse visual outputs, and a single image or video may support multiple equally c

Qingchao Shen, Zan Wang, Haoyang Ma, Yongqiang Tian

Deep Learning (DL) compilers have been widely utilized to optimize DL models for efficient deployment across various hardware. Due to their vital role in the DL ecosystem, ensuring their reliability and security is critical. However, existing approaches have limitations in testing optimization stages, which is the core functionality of DL compilers, due to t

Qianshuo Liu, Chang Feng, Filipe B. Abdalla

Ultralight axions (ULAs) are hypothetical particles which can behave like dark matter (DM) or dark energy (DE) depending on masses generated at the symmetry-breaking scale. It remains a mystery whether the ULAs can make up a fraction of DM or DE. Although theoretical predictions indicate that the ULAs may leave distinct imprints on cosmological signals, thes

Binyu Pang, Zihao Bo, Wei Chen, Xun Chen

The landmark detection of neutrinos from SN1987A marked the dawn of neutrino astrophysics. The neutrino burst provided essential insights into fundamental properties of neutrinos, and served as key probes of stellar evolution and supernova dynamics. The recent advancement in coherent elastic neutrino-nucleus scattering enables the detection of core-collapse

Ke Xu, Gang Wang, Ting Liang, Yang Xiao

Self-heating in next-generation, high-power-density field-effect transistor limits performance and complicates fabrication. Here, we introduce NEP-FET, a machine-learned framework for device-scale heat transport simulations of field-effect transistors. Built upon the neuroevolution potential, the model extends a subset of the OMat24 dataset through an active

Liu-Lin Wang, Xing-Meng Zhao, Xiao-Hai Liu, Mao-Jun Yan

We systematically investigate the $S$-wave interactions between Nambu-Goldstone bosons (NGBs) and charmed mesons in the $(S,I)=(1,0)$ sector using the chiral unitary approach. The scattering amplitudes incorporate both the Weinberg-Tomozawa term and additional contributions from $s$- and $u$-channel exchanges of $c\bar{s}$ states predicted by the constituent

Deepan Patra, Devojyoti Kansabanik, Divya Oberoi, Yûki Kubo

Some of the most interesting insights into solar physics and space weather come from studying radio emissions associated with solar activity, which remain inherently unpredictable. Hence, a real-time triggering system is needed for solar observations with the versatile new-generation radio telescopes to efficiently capture these episodes of solar activity wi

Marcel Kokorsch, Guido Dietl

The following paper presents a holistic approach to the processing of entangled links within entanglement based quantum key distribution protocols, whose security relies on the Bell inequality. We investigate the interactions, and the collective impact, of the whole processing chain on the final secure key rate. This includes the quantum mechanical preproces

Orphée Collin

We study the expansion of the limiting free energy density of the random field Ising chain with centered IID disorder and homogeneous coupling parameter, when the latter goes to infinity. We extend the first order result of [Collin, 2025] to the general case of finite second moment. Furthermore, we identify the value of the unknown constant appearing in the

Jingtian Liu, Xiongwei Yang, Yi Wei, Jianjun Yu

Free-space optical (FSO) transmission enables fast, secure, and efficient next-generation communications with abundant spectrum resources. However, atmospheric turbulence, pointing errors, path loss, and atmospheric loss induce random attenuation, challenging link reliability. Adaptive rate control technology enhances spectrum utilization and reliability. We

Samuel Cerezo, Seong Hun Lee, Javier Civera

In this letter, we present a closed-form initialization method that recovers the full visual-inertial state without nonlinear optimization. Unlike previous approaches that rely on iterative solvers, our formulation yields analytical, easy-to-implement, and numerically stable solutions for reliable start-up. Our method builds on small-rotation and constant-ve

Atila Poro, Raul Michel, Francisco Javier Tamayo, Mahya Hedayatjoo

This study broadens our comprehensive investigation of total-eclipse W Ursae Majoris-type contact binaries by analyzing eight additional systems, continuing our previous research. Multiband $BVR_cI_c$ photometric data were obtained at an observatory in Mexico, from which new times of minima were determined. All target systems also had available space-based T

Ivan Ivashkin, Eduard Kim, Emin Nugaev, Yakov Shnir

We obtain localized field configurations with finite energy in a ($2+1$)-dimensional model with Maxwell and Chern-Simons gauge terms coupled to a massive complex scalar field. These non-topological solitons are characterized by the $U(1)$ frequency and a winding number. Thus, the solutions possess Noether charge and non-trivial angular momentum, which is not

Daikang Wei, Christoph Bode, Kohei Yamamoto, Yongho Lee

We experimentally demonstrate a novel interferometric architecture for next-generation gravity missions, featuring a laser ranging interferometer (LRI) that enables monoaxial transmission and reception of laser beams between two optical benches with a heterodyne frequency of 7.3 MHz. Active beam steering loops, utilizing differential wavefront sensing (DWS)

Haobin Mao, Lipeng Zhu, Wenyan Ma, Zhenyu Xiao

Movable antennas (MAs) have emerged as a promising technology to improve wireless communication and sensing performance towards sixth-generation (6G) networks through flexible antenna movement. In this paper, we propose a novel wireless sensing system based on MA arrays to enhance multi-target spatial angle estimation performance. We begin by characterizing

Marco Zambianco, Lorenzo Fasol, Roberto Doriguzzi-Corin

The explosive growth of AI applications has created unprecedented demand for GPU resources. Cloud providers meet this demand through GPU-as-a-Service platforms that offer rentable GPU resources for running AI workloads. In this context, the sharing of GPU resources between different tenants is essential to maximize the number of scheduled workloads. Among th

Dandan Chen

Amdeberhan and Merca recently studied arithmetic properties of the sequence $a(n)$, the reciprocal of the crank parity function, which counts the number of integer partitions of weight $n$ whose even parts are monochromatic and whose odd parts may appear in one of three colors (OEIS A298311). A key result of their work was the congruence $a(7n + 2) \equiv 0

Parada T. P. Hutauruk

In this review paper, I present a study of the structure of the hadrons computed in the covariant Nambu-Jona-Lasinio model as the chiral effective quark theory of QCD. I describe how the NJL model is treated to imitate the spontaneous chiral symmetry breaking and confinement QCD properties. The consistency for the parton distribution functions and electromag

Zengjie Hu, Jiantao Qiu, Tianyi Bai, Haojin Yang

Group-based policy optimization methods like GRPO and GSPO have become standard for training multimodal models, leveraging group-wise rollouts and relative advantage estimation. However, they suffer from a critical \emph{gradient vanishing} problem when all responses within a group receive identical rewards, causing advantage estimates to collapse and traini

Boquan Chen, Matthew D. A. Orkney, Yuan-Sen Ting, Michael R. Hayden

The earliest phase of the Milky Way's evolution involved a transition from a dispersion-supported proto-galaxy to a rotationally supported disk. A key chemical signature of this transition is the moderate rise in [$\alpha$/Fe] near $\mathrm{[Fe/H]}\approx-1.3$, which we previously interpreted as evidence for $\alpha$-enhanced gas accretion fueling early disk

Xiuchao Wu, Pengfei Zhu, Jiangjing Lyu, Xinguo Liu

Applying diffusion models to physically-based material estimation and generation has recently gained prominence. In this paper, we propose \ttt, a novel material reconstruction framework for 3D objects, offering the following advantages. First, \ttt\ adopts a two-stage reconstruction, starting with accurate material prediction from inputs and followed by pri

Sergei Barakin, Kirill Gubarev, Edvard T. Musaev

We investigate behavior of brane backgrounds under poly-vector deformations in Type IIB and D=11 supergravities. We find that the standard bi-vector deformations add dissolved F1 string charge to a Dp-brane background, quadri-vector deformations add D3-brane charge, tri- and six-vector deformations in D=11 add M2- and M5-brane charges respectively. We discus

Thomas Lahens, Alain Sommier, Marie-Marthe Groz, Jean-Christophe Batsale

Estimation of thermal contact resistances between cylinders can be achieved using heating on the cross sections by a laser spot and measurement of the temperature response by IR thermography. This type of measurement makes it possible to characterise contacts in clusters of cylinders to simulate clusters of grains, or the properties of fibrous media in the t

Ziqi Wei, Yuanjian Wan, Yuhu Cheng, Xiao Yu

Optical computing represents a groundbreaking technology that leverages the unique properties of photons, with innate parallelism standing as its most compelling advantage. Parallel optical computing like cascaded Mach-Zehnder interferometers (MZIs) based offers powerful computational capabilities but also introduces new challenges, particularly concerning d

G. Zietek, N. Pillet, M. Anguiano, P. Carpentier

Energy Density Functionals are of major interest for the study of the atomic nucleus as, coupled with mean-field and beyond N-body approaches, they are applicable to the whole nuclear chart, including superheavy elements. On the one hand, the growing need for nuclear data and, on the other hand, the large amount of experimental data on exotic nuclei explain

Bruno Franchi, Pierre Pansu

There are three approaches to currents tuned to the anisotropic geometry of Heisenberg groups: Ambrosio and Kirchheim's approach valid for general metric spaces; distributions dual to horizontal differential forms; distributions dual to Rumin's complex. It is shown that, in dimensions less than half the ambient dimension, these three theories coincide. On th

Matthew W. Cotton, Alain Goriely, David Klenerman, Georg Meisl

Neurodegenerative diseases are driven by the accumulation of protein aggregates in the brain of affected individuals. The aggregation behaviour in vitro is well understood and driven by the equilibration of a super-saturated protein solution to its aggregated equilibrium state. However, the situation is altered fundamentally in living systems where active pr

Thomas Buchholtzer, Michel de Lara

Energy systems are changing rapidly. More and more, energy production is becoming decentralized, highly variable and intermittent (solar, wind), while demand is diversifying (electric vehicles). As a result, balancing supply and demand is becoming more complex, making the adjustment of demand an interesting tool. Demand response is a typical leader-follower

Qiaoyan Peng, Qingqing Wu, Wen Chen, Guangji Chen

The performance of the sensing system is limited by the signal attenuation and the number of receiving components. In this letter, we investigate the sensor position selection in a semi-passive intelligent reflecting surface (IRS) enabled non-line-of-sight (NLoS) sensing system. The IRS consists of passive elements and active sensors, where the sensors can r

Adam Rychert, Gasper Spagnolo, Evgenii Posashkov

In this reproducibility study, we revisit the LLAMBO framework of Daxberger et al. (2024), a prompting-based Bayesian optimization (BO) method that uses large language models as discriminative surrogates and acquisition optimizers via text-only interactions. We replicate the core Bayesmark and HPOBench experiments under the original evaluation protocol, but

Yonggan Fu, Xin Dong, Shizhe Diao, Matthijs Van keirsbilck

Efficient deployment of small language models (SLMs) is essential for numerous real-world applications with stringent latency constraints. While previous work on SLM design has primarily focused on reducing the number of parameters to achieve parameter-optimal SLMs, parameter efficiency does not necessarily translate into proportional real-device speed-ups.

Athanase Papadopoulos, Vladimir Turaev

We review Euler's work on spherical geometry. After an introduction concerning the general place that trigonometric formulae occupy in geometry, we start by the two memoirs of Euler on spherical trigonometry, in which he establishes the trigonometric formulae using different methods, namely, the calculus of variations in the first memoir, and classical metho