October 2025 arXiv papers — page 79

Kazutaka Takahashi, Pratik Nandy, Adolfo del Campo

The quantum dynamics of a complex system can be efficiently described in Krylov space, the minimal subspace in which the dynamics unfolds. We apply the Krylov subspace method for Hamiltonian deformations, which provides a systematic way of constructing solvable models from known instances. In doing so, we relate the evolution of deformed and undeformed theor

Gakusei Sato, Hiroya Nakao, Riccardo Muolo

Timbre allows us to distinguish between sounds even when they share the same pitch and loudness, playing an important role in music, instrument recognition, and speech. Traditional approaches, such as frequency analysis or machine learning, often overlook subtle characteristics of sound. Topological Data Analysis (TDA) can capture complex patterns, but its a

C. R. García

In this thesis, we introduce novel methods for analyzing pulsar populations using a variety of mathematical techniques. These tools-particularly graph theory-have been thoroughly validated in advanced mathematics, enabling us to overcome some of the constraints (even dimensional) inherent in conventional visualization approaches. This exploration benefits fr

Nika Mlinarič Hribar, Matjaž Depolli, Gregor Kosec

This study investigates the impact of wind velocity averaging on Dynamic Thermal Rating (DTR) calculations. It is based on a high-temporal-resolution (1 second) wind measurements obtained from a transmission line in Slovenia, Europe. Wind speed and direction variability are analysed, and two averaging methods, namely vector averaging, where velocity is avera

Yuki Mori, Kazuma Kano, Yusuke Asai, Shin Katayama

With the spread of e-commerce, the logistics market is growing around the world. Therefore, improving the efficiency of warehouse operations is essential. To achieve this, various approaches have been explored, and among them, the use of digital twins is gaining attention. To make this approach possible, it is necessary to accurately collect the positions of

Jonas Cassel, Fabio Schlindwein, Peter Albers, Christoph Schnörr

Gauge symmetric methods for data representation and analysis utilize tools from the differential geometry of vector bundles in order to achieve consistent data processing architectures with respect to local symmetry and equivariance. In this work, we elaborate concepts of geometric gauge theory for data science. Motivated by lattice gauge theory, we focus on

GigaBrain Team, Angen Ye, Boyuan Wang, Chaojun Ni

Training Vision-Language-Action (VLA) models for generalist robots typically requires large-scale real-world robot data, which is expensive and time-consuming to collect. The inefficiency of physical data collection severely limits the scalability, and generalization capacity of current VLA systems. To address this challenge, we introduce GigaBrain-0, a nove

Wonje Choi, Jooyoung Kim, Honguk Woo

We address the challenge of adopting language models (LMs) for embodied tasks in dynamic environments, where online access to large-scale inference engines or symbolic planners is constrained due to latency, connectivity, and resource limitations. To this end, we present NeSyPr, a novel embodied reasoning framework that compiles knowledge via neurosymbolic p

Xinliang Lyu, Naoki Kawashima

Tensor-network renormalization group (TNRG) is an efficient real-space renormalization group method for studying the criticality in both classical and quantum lattice systems. Exploiting symmetries of a system in a TNRG algorithm can simplify the implementation of the algorithm and can help produce correct tensor RG flows. Although a general framework for co

M. Klabunde, L. Caspari, F. Lemmerich

The modified universality hypothesis proposed by Jones et al. (2022) suggests that adversarially robust models trained for a given task are highly similar. We revisit the hypothesis and test its generality. While we verify Jones' main claim of high representational similarity in specific settings, results are not consistent across different datasets. We also

Clément de Chaisemartin, Diego Ciccia, Felix Knau, Mélitine Malézieux

The command did_multiplegt_dyn can be used to estimate event-study effects in complex designs with a potentially non-binary and/or non-absorbing treatment. This paper starts by providing an overview of the estimators computed by the command. Then, simulations based on three real datasets are used to demonstrate the estimators' properties. Finally, the comman

Insu Jeon, Youngjin Park, Gunhee Kim

Learning to infer the conditional posterior model is a key step for robust meta-learning. This paper presents a new Bayesian meta-learning approach called Neural Variational Dropout Processes (NVDPs). NVDPs model the conditional posterior distribution based on a task-specific dropout; a low-rank product of Bernoulli experts meta-model is utilized for a memor

Juan Carlos Sampedro

This note investigates the hidden relationship between the concept of algebraic multiplicity of an eigenvalue and the local intersection index of algebraic varieties.

Jia-Kai Dong, I-Wei Huang, Chun-Tin Wu, Yi-Tien Tsai

We introduce ETOM, a five-level benchmark for evaluating multi-hop, end-to-end tool orchestration by LLM agents within a hierarchical Model-Context Protocol (MCP) ecosystem. Existing benchmarks often assess tools in isolation, overlooking challenges such as functional overlap and cross-server orchestration, which can lead to overly optimistic evaluations. ET

Kemou Li, Qizhou Wang, Yue Wang, Fengpeng Li

Large language models trained on vast corpora inherently risk memorizing sensitive or harmful content, which may later resurface in their outputs. Prevailing unlearning methods generally rely on gradient ascent and its variants to lower the probability of specific target responses. However, we find that this strategy induces a critical side effect: probabili

Songqi Zhou, Zeyuan Liu, Benben Jiang

Ensuring fairness in machine learning models is a critical challenge. Existing debiasing methods often compromise performance, rely on static correction strategies, and struggle with data sparsity, particularly within minority groups. Furthermore, their utilization of sensitive attributes is often suboptimal, either depending excessively on complete attribut

J Rosser, José Luis Redondo García, Gustavo Penha, Konstantina Palla

As Large Language Models (LLMs) scale to million-token contexts, traditional Mechanistic Interpretability techniques for analyzing attention scale quadratically with context length, demanding terabytes of memory beyond 100,000 tokens. We introduce Sparse Tracing, a novel technique that leverages dynamic sparse attention to efficiently analyze long context at

Yuan Gao, Suchir Salhan, Andrew Caines, Paula Buttery

To bridge the gap between performance-oriented benchmarks and the evaluation of cognitively inspired models, we introduce BLiSS 1.0, a Benchmark of Learner Interlingual Syntactic Structure. Our benchmark operationalizes a new paradigm of selective tolerance, testing whether a model finds a naturalistic learner error more plausible than a matched, artificial

Mete Harun Akcay, Buse Gul Atli, Siddharth Prakash Rao, Alexandros Bakas

As the volume of stored data continues to grow, identifying and protecting sensitive information within large repositories becomes increasingly challenging, especially when shared with multiple users with different roles and permissions. This work presents a system architecture for trusted data sharing with policy-driven access control, enabling selective pr

Noy Soffer Aranov, Steven Robertson

Let $p$ be a prime. In 2017, Kemarsky, Paulin, and Shapira (KPS) conjectured that any Laurent series over $\mathbb{F}_p$ exhibits full escape of mass with respect to any irreducible polynomial $P(t)\in\mathbb{F}_p[t]$. In 2025, this was shown to be false in the case $p=2$ and $P(t)=t$ by Nesharim, Shapira and the first named author. This work shows that for

Abdelrahman Sayed Sayed

The growing interest in ocean discovery imposes a need for inspection and intervention in confined and demanding environments. Eely's slender shape, in addition to its ability to change its body configurations, makes articulated underwater robots an adequate option for such environments. However, operation of Eely in such environments imposes demanding requi

Tong Zhang, Yihuan Huang, Yanzhen Ren

The growing prevalence of speech deepfakes has raised serious concerns, particularly in real-world scenarios such as telephone fraud and identity theft. While many anti-spoofing systems have demonstrated promising performance on lab-generated synthetic speech, they often fail when confronted with physical replay attacks-a common and low-cost form of attack u

Yasser Hamidullah, Josef van Genabith, Cristina España-Bonet

This paper describes the DFKI-MLT submission to the WMT-SLT 2022 sign language translation (SLT) task from Swiss German Sign Language (video) into German (text). State-of-the-art techniques for SLT use a generic seq2seq architecture with customized input embeddings. Instead of word embeddings as used in textual machine translation, SLT systems use features e

Mathieu Dedenon

In soft matter, the phase of nematic liquid crystals can be made from anisotropic molecules in single component materials, or as a suspension of mesoscopic nematogens. The later offers more versatility in the experimental design of complex shapes, in particular thin curved shells, and is often found in biological systems at multiple scales from cells to tiss

Davide Mattiolo, Giuseppe Mazzuoccolo, Jozef Rajník, Gloria Tabarelli

A $d$-dimensional nowhere-zero $r$-flow on a graph $G$, an $(r,d)$-NZF from now on, is a flow where the value on each edge is an element of $\mathbb{R}^d$ whose (Euclidean) norm lies in the interval $[1, r-1]$. Such a notion is a natural generalization of the well-known concept of a circular nowhere-zero $r$-flow (i.e.\ $d = 1$). The minimum of the real numb

Victor Morand, Nadi Tomeh, Josiane Mothe, Benjamin Piwowarski

Identifying which text spans refer to entities - mention detection - is both foundational for information extraction and a known performance bottleneck. We introduce ToMMeR, a lightweight model (<300K parameters) probing mention detection capabilities from early LLM layers. Across 13 NER benchmarks, ToMMeR achieves 93% recall zero-shot, with an estimated 90%

Rafael S. de Souza, Ana L. Chies-Santos

We present a Bayesian latent model to describe the scaling relation between globular cluster populations and their host galaxies, updating the framework proposed in de Souza 2015. GC counts are drawn from a negative-binomial (NB) process linked to host stellar mass, augmented with a newly introduced Gaussian observation layer that enables efficient propagati

Anna Qi, Jianfeng Huang, Lihua Yang, Chao Huang

In this paper, we consider a class of single-ratio fractional minimization problems, where both the numerator and denominator of the objective are convex functions satisfying positive homogeneity. Many nonsmooth optimization problems on the sphere that are commonly encountered in application scenarios across different scientific fields can be converted into

Vanshika Datta, C. Nahak

A sensor has the ability to probe its surroundings. However, uncertainties in its exact location can significantly compromise its sensing performance. The radius of robust feasibility defines the maximum range within which robust feasibility is ensured. This work introduces a novel approach integrating it with the directional sensor networks to enhance cover

Yicheng Xiong, Yanmei Cai, Tianxing Ma

The pairing symmetry of the recently discovered bilayer nickelate superconductor La$_3$Ni$_2$O$_7$ is a subject of intense debate in condensed matter physics, with the two leading theoretical candidates being a sign-reversing $s_{\pm}$-wave and a $d$-wave state. To investigate its ground-state properties in the intermediate coupling regime which is critical

Qian Huang, Nachamma Sockalingam, Thijs Willems, King Wang Poon

This study explores how graduate students in an urban planning program transitioned from passive users of generative AI to active creators of custom GPT-based knowledge tools. Drawing on Self-Determination Theory (SDT), which emphasizes the psychological needs of autonomy, competence, and relatedness as foundations for intrinsic motivation, the research inve

K. Somesh, T. Thebault, V. Taufour, D. Aoki

We present electrical-resistivity measurements on the prototypical heavy-fermion ferromagnet UGe$_2$ under pulsed magnetic field up to 60~T. An anisotropic field-induced suppression of the electronic correlations is revealed. The electrical resistivity strongly decreases when a magnetic field $\mathbf{H}$ is applied along the easy magnetic axis $\mathbf{a}$,

Sumati Surya

We construct a family of closeness functions on the space of finite volume Lorentzian geometries using the abundance of discrete intervals in the underlying random causal sets. Although strictly weaker than a Lorentzian Gromov-Hausdorff distance function, it has the advantage of being numerically calculable for large causal sets. It thus provides a concrete

Kaifeng Bao, Tao Zhou, Chaoyi Li, Liu Liu

Channel measurements are the prerequisite for applying emerging transmission technologies and designing communication systems. Conventional time or frequency domain channel measurement methods cannot directly obtain Doppler information induced by high-mobility scenarios. The channel spreading function (CSF) simultaneously captures delay and Doppler informati

Tao Zhou, Liying Geng, Kaifeng Bao, Tianyun Feng

This paper investigates narrow-beam channel characterization and performance evaluation for 5G for railway (5G-R) systems based on ray-tracing (RT) simulation. Three representative high-speed railway (HSR) scenarios including viaduct, cutting, and station are established, and RT-based dynamic narrow-beam channel simulations are conducted using a designed bea

Abbi Abdel-Rehim, Emma Tate, Larisa N. Soldatova, Ross D. King

Early identification of sensitive cancer cell lines is essential for accelerating biomarker discovery and elucidating drug mechanism of action. Given the efficiency and low cost of small-scale drug screens relative to extensive omics profiling, we compared drug-response panel (DRP) descriptors against omics features for predictive capacity using gradient boo

Zhiyuan Feng, Zhaolu Kang, Qijie Wang, Zhiying Du

Vision-language models (VLMs) are essential to Embodied AI, enabling robots to perceive, reason, and act in complex environments. They also serve as the foundation for the recent Vision-Language-Action (VLA) models. Yet most evaluations of VLMs focus on single-view settings, leaving their ability to integrate multi-view information underexplored. At the same

Yasser Hamidullah, Shakib Yazdani, Cennet Oguz, Josef van Genabith

Sign language translation (SLT) is typically trained with text in a single spoken language, which limits scalability and cross-language generalization. Earlier approaches have replaced gloss supervision with text-based sentence embeddings, but up to now, these remain tied to a specific language and modality. In contrast, here we employ language-agnostic, mul

Javier Álvarez-Vizoso, David Barral

The decomposition of arbitrary unitary transformations into sequences of simpler, physically realizable operations is a foundational problem in quantum information science, quantum control, and linear optics. We establish a 1D Quantum Field Theory model for justifying the universality of a broad class of such factorizations. We consider parametrizations of t

Hira Asif, Taner Tarik Aytas, Ramazan Sahin

Active control of the radiative properties of quantum emitters through engineered light-matter interactions is a key challenge in nanophotonics and quantum optics. In this work, we demonstrate dynamic modulation of dipole's decay rate by exploiting the tunable plexcitonic modes (graphene plasmons and QD-excitons) in the strong coupling regime. By integrating

Alexander Mrokon, Till Wachweger, Dongsung Shin, Karsten Buse

Adiabatic frequency conversion enables fast and efficient tuning of laser light by coupling it into an optical resonator whose eigenfrequency is varied on a timescale shorter than its photon lifetime. In this regime, the optical frequency follows the cavity resonance, allowing frequency shifts of several hundred gigahertz within sub-microsecond time - indepe

Stefan Schott, Serena Elisa Ponta, Wolfram Fischer, Jonas Klauke

On average, 71% of the code in typical Java projects comes from open-source software (OSS) dependencies, making OSS dependencies the dominant component of modern software code bases. This high degree of OSS reliance comes with a considerable security risk of adding known security vulnerabilities to a code base. To remedy this risk, researchers and companies

Zixu Feng, Lunxu Liu, Qinglin Tang

The gradient flow with semi-implicit discretization (GFSI) is the most widely used algorithm for computing the ground state of Gross-Pitaevskii energy functional. Numerous numerical experiments have shown that the energy dissipation holds when calculating the ground states of multicomponent Bose-Einstein condensates (MBECs) with GFSI, while rigorous proof re

Guanping Xu, Zirui Zhao, Muqing Su, Hai-Feng Li

Multiferroic materials, particularly rare-earth orthochromates (RECrO$_3$), have garnered significant interest due to their unique magnetic and electric-polar properties, making them promising candidates for multifunctional devices. Although extensive research has been conducted on their antiferromagnetic (AFM) transition temperature (N$\acute{\textrm{e}}$el

Max O. Al-Hasso, Marko von der Leyen

The closest vector problem (CVP) is a fundamental optimization problem in lattice-based cryptography and its conjectured hardness underpins the security of lattice-based cryptosystems. Furthermore, Schnorr's lattice-based factoring algorithm reduces integer factoring (the foundation of current cryptosystems, including RSA) to the CVP. Recent work has investi

Lin Xv, Jingsheng Gao, Xian Gao, Ting Liu

In the field of large language model (LLM) compression, singular value decomposition (SVD) is a widely studied and adopted low-rank decomposition technique. Since SVD operates exclusively on linear modules, and these modules in LLMs are separated by nonlinear components, SVD can only be applied independently to each linear module. Under a global compression

M. A. Krishnakumar, Bhal Chandra Joshi, P. K. Manoharan

We report wideband scatter-broadening estimates of 14 pulsars towards the Gum nebula region using the Band-3 of the upgraded GMRT. This work increases the measurements of frequency scaling index of scatter-broadening ($α$) across the nebula by more than 3 times. A strong correlation between the distance and the scattering strength is observed for pulsars beh

Phat Tran, Enbai Kuang, Fred Xu

Intracranial hemorrhage (ICH) secondary to Traumatic Brain Injury (TBI) represents a critical diagnostic challenge, with approximately 64,000 TBI-related deaths annually in the United States. Current diagnostic modalities including Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) have significant limitations: high cost, limited availability, and

Akiteru Takahashi, Kaichi Teranishi, Shonosuke Takaichi, Taishi Nishihara

Owing to their small binding energies, excitons in bulk semiconductors typically exhibit a sharp optical peak at low temperatures only. This limitation can be overcome by single-walled carbon nanotubes (SWCNTs) and other low-dimensional semiconductors with highly enhanced exciton binding energies. Exciton thermal radiation, which can potentially be exploited

Ning Li, Qiqiang Lin, Zheng Wu, Xiaoyun Mo

With the advancements in hardware, software, and large language model technologies, the interaction between humans and operating systems has evolved from the command-line interface to the rapidly emerging AI agent interactions. Building an operating system (OS) agent capable of executing user instructions and faithfully following user desires is becoming a r

Lin Xv, Xian Gao, Ting Li, Yuzhuo Fu

Low-rank decomposition, particularly Singular Value Decomposition (SVD), is a pivotal technique for mitigating the storage and computational demands of Large Language Models (LLMs). However, prevalent SVD-based approaches overlook the critical phenomenon that decomposition errors exhibit significant disparity across different components of the parameter matr

Yuhang Liu, Minglai Shao, Zengyi Wo, Yunlong Chu

Pre-training Graph Foundation Models (GFMs) on text-attributed graphs (TAGs) is central to web-scale applications such as search, recommendation, and knowledge discovery. However, existing CLIP-style graph-text aligners face two key limitations: they assume strict one-to-one correspondences between nodes and texts, overlooking the inherent many-to-many relat

Guus Toussaint, Arno Knobbe

Many systems in our world age, degrade or otherwise move slowly but steadily in a certain direction. When monitoring such systems by means of sensors, one often assumes that some form of `age' is latently present in the data, but perhaps the available sensors do not readily provide this useful information. The task that we study in this paper is to extract p

Mark Dukes, Andrew Mullins

Banach's matchbox problem considers the setting of two matchboxes that each initially contain the same number of matches. Boxes are chosen with equal probability and a match removed each time. The problem concerns the law of the number of matches remaining in one box once the other box empties. Knuth considered a generalization of this problem whereby `big-c

Riccardo Messina, Philippe Ben-Abdallah

We develop a general theory of radiative heat exchange between dipoles with time-modulated optical properties. This framework extends fluctuational electrodynamics beyond equilibrium by incorporating nonstationary correlations and memory effects induced by temporal modulation. Closed-form expressions for the heat currents in modulated many-body systems are o

Fabian Schaipp

The training of diffusion models is often absent in the evaluation of new optimization techniques. In this work, we benchmark recent optimization algorithms for training a diffusion model for denoising flow trajectories. We observe that Muon and SOAP are highly efficient alternatives to AdamW (18% lower final loss). We also revisit several recent phenomena r

Gaven Martin, Cong Yao

Here we advance the study of boundary the value problem for extremal functions of mean distortion and the associated Teichm\"uller spaces interpolating between the classical examples of extremal quasiconformal mappings, and the more recent approach through harmonic mappings (of extreme Dirichlet energy). In this paper we focus on the Alhfors-Hopf differentia

Basavasagar Patil, Sydney Belt, Jayjun Lee, Nima Fazeli

Increasingly large datasets of robot actions and sensory observations are being collected to train ever-larger neural networks. These datasets are collected based on tasks and while these tasks may be distinct in their descriptions, many involve very similar physical action sequences (e.g., 'pick up an apple' versus 'pick up an orange'). As a result, many da

Corentin Pla, Hugo Richard, Marc Abeille, Nadav Merlis

We study reinforcement learning (RL) with transition look-ahead, where the agent may observe which states would be visited upon playing any sequence of $\ell$ actions before deciding its course of action. While such predictive information can drastically improve the achievable performance, we show that using this information optimally comes at a potentially

Woo Jae Kim, Kyu Beom Han, Yoonki Cho, Youngju Na

As Neural Radiance Fields (NeRFs) have emerged as a powerful tool for 3D scene representation and novel view synthesis, protecting their intellectual property (IP) from unauthorized use is becoming increasingly crucial. In this work, we aim to protect the IP of NeRFs by injecting adversarial perturbations that disrupt their unauthorized applications. However

Lorenzo Luperi Baglini, Alessandro Vegnuti

We introduce the notion of asymptotic partition regularity for Diophantine equations. We show how this notion is at the core of almost all known negative results in the Ramsey theory of equations, and we use it to produce new ones, as in the case of Fermat-Catalan equations. The methods we use here are based on translating asymptotic partition regularity int

Jorge Vallejo Bustamante, Viet-Hung Nguyen, Liam S. Farrar, Kenji Watanabe

When monolayer graphene is crystallographically aligned to hexagonal boron nitride (BN), a moir\'e superlattice is formed, producing characteristic satellite Dirac peaks in the electronic band structure. Aligning a second BN layer to graphene creates two coexisting moir\'e patterns, which can interfere to produce periodic, quasi-periodic or non-periodic supe

Weichuang Shao, Iman Yi Liao, Tomas Henrique Bode Maul, Tissa Chandesa

Recent foundational models, SSAST, EAT, HuBERT, Qwen-Audio, and Audio Flamingo, achieve top-tier results across standard audio benchmarks but are limited by fixed input rates and durations, hindering their reusability. This paper introduces the Augmentation-driven Multiview Audio Transformer (AMAuT), a training-from-scratch framework that eliminates the depe

Yasser Hamidullah, Josef van Genabith, Cristina España-Bonet

State-of-the-art sign language translation (SLT) systems facilitate the learning process through gloss annotations, either in an end2end manner or by involving an intermediate step. Unfortunately, gloss labelled sign language data is usually not available at scale and, when available, gloss annotations widely differ from dataset to dataset. We present a nove

Xinfeng Xia, Xiaofeng Hou, Jiacheng Liu, Wenfeng Wang

Mixture-of-Experts (MoE) scales model capacity through sparse activation, and is becoming an important architecture for large language models (LLMs). However, existing MoE serving systems typically execute all requests under a fixed routing configuration, limiting their ability to exploit heterogeneous computation requirements across requests. Routing top-$k

Umar Butler, Abdur-Rahman Butler, Adrian Lucas Malec

We present the Massive Legal Embedding Benchmark (MLEB), the largest, most diverse, and most comprehensive open-source benchmark for legal information retrieval to date. MLEB consists of ten expert-annotated datasets spanning multiple jurisdictions (the US, UK, EU, Australia, Ireland, and Singapore), document types (cases, legislation, regulatory guidance, c

Golnaz Raja, Ruslan Agishev, Miloš Prágr, Joni Pajarinen

Uncertainty-aware robot motion prediction is crucial for downstream traversability estimation and safe autonomous navigation in unstructured, off-road environments, where terrain is heterogeneous and perceptual uncertainty is high. Most existing methods assume deterministic or spatially independent terrain uncertainties, ignoring the inherent local correlati

Siyuan Wang, Gaokai Zhang, Li Lyna Zhang, Ning Shang

Reasoning over long contexts is essential for large language models. While reinforcement learning (RL) enhances short-context reasoning by inducing "Aha" moments in chain-of-thought, the advanced thinking patterns required for long-context reasoning remain largely unexplored, and high-difficulty RL data are scarce. In this paper, we introduce LoongRL, a data

Tim Ehret, Vyacheslav Shatokhin, Andreas Buchleitner

We formulate a Floquet-Markov Lindblad master equation for translationally cold two-level atoms driven by a strong monochromatic wave and coupled to a common electromagnetic bath. The resulting dipole-dipole interaction reproduces the anisotropic Heisenberg model.

Xianyang Liu, Yilin Liu, Shuai Wang, Hao Cheng

The creation of high-quality datasets to improve Large Language Model (LLM) reasoning remains a significant challenge, as current methods often suffer from generating low-quality/incorrect answers and limited information richness from available data sources. To address this, we propose AgenticMath, a novel agentic method for generating high-quality mathemati

Junfeng Gong, Zhiyi Wei, Junying Chen, Cheng Liu

Despite significant evolution of CUDA programming and domain-specific libraries, effectively utilizing GPUs with massively parallel engines remains difficult. Large language models (LLMs) show strong potential in generating optimized CUDA code from sequential code. However, using LLMs in practice faces two major challenges: cloud-based APIs pose risks of cod

Ngoc-Bao-Quang Nguyen, Tuan-Minh Do, Cong-Tam Phan, Thi-Thu-Hong Phan

Automated image-based garbage classification is a critical component of global waste management; however, systematic benchmarks that integrate Machine Learning (ML), Deep Learning (DL), and efficient hybrid solutions remain underdeveloped. This study provides a comprehensive comparison of three paradigms: (1) machine learning algorithms using handcrafted fea

Dongwon Kim, Jiwan Seo, Joonhyuk Kang

The integration of artificial intelligence (AI) with the Internet of Things (IoT) enables task-oriented communication for multi-edge cooperative inference system, where edge devices transmit extracted features of local sensory data to an edge server to perform AI-driven tasks. However, the privacy concerns and limited communication bandwidth pose fundamental

Jia Deng, Jin Li, Zhenhua Zhao, Shaowei Wang

Vision-Language Models (VLMs), such as CLIP, have demonstrated remarkable zero-shot generalizability across diverse downstream tasks. However, recent studies have revealed that VLMs, including CLIP, are highly vulnerable to adversarial attacks, particularly on their visual modality. Traditional methods for improving adversarial robustness, such as adversaria

Yue Chu, Chen-Hao Wu, Ya-Peng Hu

By employing Duan's topological method, we classify critical points by their topological charge Q = +/-1 or 0. Previous work (Wei et al., Phys. Rev. D 105, 104003, 2022) investigated two typical anti-de Sitter (AdS) black holes: the Reissner-Nordstroem (RN) case (with only one critical point Q = -1) and the Born-Infeld (BI) case (with two critical points Q =

Yejin Kwon, Taewoo Kang, Hyunsoo Yoon, Changouk Kim

We present M3-SLU, a new multimodal large language model (MLLM) benchmark for evaluating multi-speaker, multi-turn spoken language understanding. While recent models show strong performance in speech and text comprehension, they still struggle with speaker-attributed reasoning, the ability to understand who said what and when in natural conversations. M3-SLU

Andrey Pudovikov, Alexandra Khirianova, Ekaterina Solodneva, Aleksandr Katrutsa

Advertisement auctions play a crucial role in revenue generation for e-commerce companies. To make the bidding procedure scalable to thousands of auctions, the automatic bidding (autobidding) algorithms are actively developed in the industry. Therefore, the fair and reproducible evaluation of autobidding algorithms is an important problem. We present a stand

Yu Fang, Xinyu Wang, Xuehe Zhang, Wanli Xue

The wide application of flow-matching methods has greatly promoted the development of robot imitation learning. However, these methods all face the problem of high inference time. To address this issue, researchers have proposed distillation methods and consistency methods, but the performance of these methods still struggles to compete with that of the orig

Alessio Caminata, Francesco Zerman

We extend the theory of $p$-fractals of Monsky and Teixeira by introducing the notion of weak $p$-fractal. We prove that for a hypersurface $f$ having rational Hilbert-Kunz series is equivalent to the weak $p$-fractality of the associated function $\phi_{f,p}$ and having rational F-signature series is equivalent to the weak $p$-fractality of the reflection $

Eylon Zohar, Israel Nelken, Boaz Rafaely

Classical auditory-periphery models, exemplified by Bruce et al., 2018, provide high-fidelity simulations but are stochastic and computationally demanding, limiting large-scale experimentation and low-latency use. Prior neural encoders approximate aspects of the periphery; however, few are explicitly trained to reproduce the deterministic, rate-domain neurog

Guoying Sun, Weiyu Guo, Tong Shao, Yang Yang

Brain decoding is a key neuroscience field that reconstructs the visual stimuli from brain activity with fMRI, which helps illuminate how the brain represents the world. fMRI-to-image reconstruction has achieved impressive progress by leveraging diffusion models. However, brain signals infused with prior knowledge and associations exhibit a significant infor

Ahsan Raza Siyal, Markus Haltmeier, Ruth Steiger, Malik Galijasevic

Deformable medical image registration is a fundamental task in medical image analysis. While deep learning-based methods have demonstrated superior accuracy and computational efficiency compared to traditional techniques, they often overlook the critical role of regularization in ensuring robustness and anatomical plausibility. We propose DARE (Deformable Ad

Omer Tariq, Muhammad Bilal, Muneeb Ul Hassan, Dongsoo Han

Data-driven inertial sequence learning has revolutionized navigation in GPS-denied environments, offering superior odometric resolution compared to traditional Bayesian methods. However, deep learning-based inertial tracking systems remain vulnerable to privacy breaches that can expose sensitive training data. \hl{Existing differential privacy solutions ofte

Nilesh Ramgolam, Gustavo Carneiro, Hsiang-Ting Chen

This paper addresses the critical data scarcity that hinders the practical deployment of learning to defer (L2D) systems to the population. We introduce a context-aware, semi-supervised framework that uses meta-learning to generate expert-specific embeddings from only a few demonstrations. We demonstrate the efficacy of a dual-purpose mechanism, where these

Varsha Suresh, M. Hamza Mughal, Christian Theobalt, Vera Demberg

In conversation, humans use multimodal cues, such as speech, gestures, and gaze, to manage turn-taking. While linguistic and acoustic features are informative, gestures provide complementary cues for modeling these transitions. To study this, we introduce DnD Gesture++, an extension of the multi-party DnD Gesture corpus enriched with 2,663 semantic gesture a

Evgenia Shustova, Marina Sheshukova, Sergey Samsonov, Evgeny Frolov

In this paper, we introduce PSI-LinUCB, a scalable variant of LinUCB that enables efficient training, inference, and memory usage by representing the inverse regularized design matrix as a sum of a diagonal matrix and low-rank correction. We derive numerically stable rank-1 and batched updates that maintain the inverse without explicitly forming the matrix.

Paul Strang, Zacharie Alès, Côme Bissuel, Olivier Juan

Mixed-Integer Linear Programming (MILP) is a powerful framework used to address a wide range of NP-hard combinatorial optimization problems, often solved by Branch and Bound (B&B). A key factor influencing the performance of B&B solvers is the variable selection heuristic governing branching decisions. Recent contributions have sought to adapt reinforcement

Xingyang Nie, Caoliang Zhang, Su Pan, Biao Wang

Most machine learning models are vulnerable to adversarial examples, which poses security concerns on these models. Adversarial examples are crafted by applying subtle but intentionally worst-case modifications to examples from the dataset, leading the model to output a different answer from the original example. In this paper, adversarial examples are forme

Prakrithi Shivaprakash, Lekhansh Shukla, Animesh Mukherjee, Prabhat Chand

Removing Personally Identifiable Information (PII) from clinical notes in Electronic Health Records (EHRs) is essential for research and AI development. While Large Language Models (LLMs) are powerful, their high computational costs and the data privacy risks of API-based services limit their use, especially in low-resource settings. To address this, we deve

Alvaro Perez-Diaz, James C. Loach, Danielle E. Toutoungi, Lee Middleton

Time-series foundation models (TSFMs) achieve strong forecast accuracy, yet accuracy alone does not determine practical value. The form of a forecast -- point, quantile, parametric, or trajectory ensemble -- fundamentally constrains which operational tasks it can support. We survey recent TSFMs and find that two-thirds produce only point or parametric foreca

Gwyn Bellamy

The seminal paper "J.T. Stafford, Module structure of Weyl algebras, J. London Math. Soc. (2) 18 (1978), no. 3, 429--442" was a major step forward in our understanding of Weyl algebras. Beginning with Serre's Theorem on free summands of projective modules and Bass' Stable Range Theorem in commutative algebra, we attempt to trace the origins of this work and

Zhen Zhu, Shan Gao, Jing Zhang, Xuxin Kang

Elucidating the catalytic descriptor that accurately characterizes the structure-activity relationships of typical catalysts for various important heterogeneous catalytic reactions is pivotal for designing high-efficient catalytic systems. Here, an interpretable machine learning technique was employed to identify the key determinants governing the nitrate re

Thijs Willems, Sumbul Khan, Qian Huang, Bradley Camburn

This pilot study traces students' reflections on the use of AI in a 13-week foundational design course enrolling over 500 first-year engineering and architecture students at the Singapore University of Technology and Design. The course was an AI-enhanced design course, with several interventions to equip students with AI based design skills. Students were re

Francisco J. Aragón-Artacho, Rubén Campoy, Pedro Pérez-Aros, David Torregrosa-Belén

In this paper we present a nonmonotone line search subgradient algorithm tailored to upper-$\mathcal{C}^2$ functions. This is a family of nonsmooth and nonconvex functions that satisfies a nonsmooth and local version of the descent lemma, making them suitable for line searches. We prove subsequential convergence of the proposed algorithm to a stationary poin

L. Caspari, M. Dinzinger, K. Ghosh Dastidar, C. Fellicious

Dense retrieval systems have proven to be effective across various benchmarks, but require substantial memory to store large search indices. Recent advances in embedding compression show that index sizes can be greatly reduced with minimal loss in ranking quality. However, existing studies often overlook the role of corpus complexity -- a critical factor, as

Martin Hairer, Harprit Singh

We introduce an approach to study homogenisation of a large class of singular SPDEs of the form $$ \partial_t u_\varepsilon - \nabla\cdot {A}(x/\varepsilon,t/\varepsilon^2) \nabla u_\varepsilon = F(x/\varepsilon , t/\varepsilon^2, u_\varepsilon , \nabla u_\varepsilon , \xi ) $$ which is based on the idea of importing (classical) homogenisation results into t

Ling Team, Bin Han, Caizhi Tang, Chen Liang

In this technical report, we present the Ring-linear model series, specifically including Ring-mini-linear-2.0 and Ring-flash-linear-2.0. Ring-mini-linear-2.0 comprises 16B parameters and 957M activations, while Ring-flash-linear-2.0 contains 104B parameters and 6.1B activations. Both models adopt a hybrid architecture that effectively integrates linear atte

Kai Shi, Jun Yang, Ni Yang, Binqiang Pan

Mobile Phone Agents (MPAs) have emerged as a promising research direction due to their broad applicability across diverse scenarios. While Multimodal Large Language Models (MLLMs) serve as the foundation for MPAs, their effectiveness in handling multiple mobile phone tasks simultaneously remains limited. Although multitask supervised fine-tuning (SFT) is wid

Jian Cui, Guo-Wang Du, Xin-Zhong Er, Chu-Xiang Li

Accurate identification of unobservable regions in nighttime is essential for autonomous scheduling and data quality control in observations.Traditional methods-such as infrared sensing or photometric extinction-provide only coarse,non-spatial estimates of sky clarity,making them insufficient for real-time decision-making.This not only wastes observing time

Cuize Han, Sesh Jalagam

The advent of Large Language Models has revolutionized tasks across domains, including the automation of legal document analysis, a critical component of modern contract management systems. This paper presents a comprehensive implementation of LLM-enhanced metadata extraction for contract review, focusing on the automatic detection and annotation of salient