October 2025 arXiv papers — page 139

Zhuotong Cai, Tianyi Zeng, Jiazhen Zhang, Eléonore V. Lieffrig

Head movement poses a significant challenge in brain positron emission tomography (PET) imaging, resulting in image artifacts and tracer uptake quantification inaccuracies. Effective head motion estimation and correction are crucial for precise quantitative image analysis and accurate diagnosis of neurological disorders. Hardware-based motion tracking (HMT)

Colin Davalo, Parker Evans

Let $S$ be a closed surface of genus $g \geq 2$. We construct locally homogeneous geometric structures on closed 5-manifolds fibering over $S$, modeled on the two partial flag manifolds $\mathrm{Ein}^{2,3}$ and $\mathrm{Pho}^\times$ of the split real form $\mathrm{G}_2'$ of the complex exceptional Lie group $\mathrm{G}_2^{\mathbb{C}}$. To this end, we co

Wojciech Kamiński, Qiaoyin Pan

In this paper, we introduce a general method to prove the non-degeneracy of the Hessian in the spinfoam vertex amplitude for quantum gravity and apply it to the spinfoam models with a cosmological constant ($\Lambda$-SF models). By reformulating the problem in terms of the transverse intersection of some submanifolds in the phase space of flat ${\rm SL}(2,\m

Diwakara Reddy, Christian Herglotz, André Kaup

In today's society, live video streaming and user generated content streamed from battery powered devices are ubiquitous. Live streaming requires real-time video encoding, and hardware video encoders are well suited for such an encoding task. In this paper, we introduce a high-level feature model using Gaussian process regression that can predict the encodin

Wenpu Li, Bangyan Liao, Yi Zhou, Qi Xu

The estimation of optical flow and 6-DoF ego-motion, two fundamental tasks in 3D vision, has typically been addressed independently. For neuromorphic vision (e.g., event cameras), however, the lack of robust data association makes solving the two problems separately an ill-posed challenge, especially in the absence of supervision via ground truth. Existing w

Siqi Li, Yasser Shoukry

Deep neural networks are highly susceptible to adversarial attacks, which pose significant risks to security- and safety-critical applications. We present KoALA (KL-L0 Adversarial detection via Label Agreement), a novel, semantics-free adversarial detector that requires no architectural changes or adversarial retraining. KoALA operates on a simple principle:

Junjie Wu, Benjamin B Risk, Taylor A James, Nicholas Seyfried

Background: Non-linear alterations in brain network connectivity may represent early neural signatures of Alzheimer's disease (AD) pathology in cognitively normal older adults. Understanding these changes and their cognitive relevance may help clarify early network vulnerability associated with AD pathology. Most prior studies recruited participants from mem

A. Alfarano, L. Venturoli, D. Negueruela del Castillo

Multimodal Large Language Models (MLLMs) have demonstrated significant capabilities in joint visual and linguistic tasks. However, existing Visual Question Answering (VQA) benchmarks often fail to evaluate deep semantic understanding, particularly in complex domains like visual art analysis. Confined to simple syntactic structures and surface-level attribute

Zhiliu Yang, Jinyu Dai, Jianyuan Zhang, Zhu Yang

The scene perception, understanding, and simulation are fundamental techniques for embodied-AI agents, while existing solutions are still prone to segmentation deficiency, dynamic objects' interference, sensor data sparsity, and view-limitation problems. This paper proposes a novel framework, named SPORTS, for holistic scene understanding via tightly integra

Yong-Kun Wang, Si Li, Shengyuan A. Yang

Altermagnets, a novel class of collinear magnetic materials, exhibit unique spin-split band structures, yet topological insulating states in intrinsic altermagnetic systems are rare. Here, we identify monolayer Fe$_2X_2$O ($X$ = Cl, Br, I) as a new family of 2D altermagnetic real Chern insulators. These materials display robust $d$-wave altermagnetic orderin

Junhao Zhuang, Shi Guo, Xin Cai, Xiaohui Li

Diffusion models have recently advanced video restoration, but applying them to real-world video super-resolution (VSR) remains challenging due to high latency, prohibitive computation, and poor generalization to ultra-high resolutions. Our goal in this work is to make diffusion-based VSR practical by achieving efficiency, scalability, and real-time performa

Young-Joon Song, Roser Valentí

Iron molybdate (Fe$_2$(MoO$_4$)$_3$) is a widely used commercial catalyst for oxidative dehydrogenation. Recently, the possibility that bulk oxygen atoms participate in catalytic reactions has been proposed based on the experimentally observed significant reduction in Raman intensity during the catalytic process, which implies the formation of oxygen defects

Mafal Ndiaye Diop, Abdou Bousso, Cheikh Khoule, Ameth Ndiaye

The objective of this paper is to deepen the study of vector fields on hyperbolic spaces $\mathbb{H}^n$ that transform them into a Ricci-Bourguignon soliton. Starting from a recent work in \cite{bousso2025ricci} which characterizes these fields as Killing fields of a specific shape, we propose a detailed geometric study of their structure and behavior in the

Micah Carroll, Adeline Foote, Kevin Feng, Marcus Williams

When users are dissatisfied with recommendations from a recommender system, they often lack fine-grained controls for changing them. Large language models (LLMs) offer a solution by allowing users to guide their recommendations through natural language requests (e.g., "I want to see respectful posts with a different perspective than mine"). We propose a meth

Adam Tupper, Christian Gagné

Foundation models open up new possibilities for the use of AI in healthcare. However, even when pre-trained on health data, they still need to be fine-tuned for specific downstream tasks. Furthermore, although foundation models reduce the amount of training data required to achieve good performance, obtaining sufficient data is still a challenge. This is due

Sanghee J. Kim, Kanishka Misra

Evaluating the naturalness of dialogue in language models (LMs) is not trivial: notions of 'naturalness' vary, and scalable quantitative metrics remain limited. This study leverages the linguistic notion of 'at-issueness' to assess dialogue naturalness and introduces a new method: Divide, Generate, Recombine, and Compare (DGRC). DGRC (i) divides a dialogue a

Mohanad Obeed, Ming Jian

Deep learning (DL) based methods for orthogonal frequency division multiplexing (OFDM) radio receivers demonstrated higher signal detection performance compared to the traditional receivers. However, the existing DL-based models, usually adapted from computer vision, aren't well suited for wireless communications. These models require high computational reso

Shoshannah Byrne-Mamahit, Sara L. Ellison, David R. Patton, Scott Wilkinson

Galaxy mergers are transformative events that can cause gaseous inflows capable of triggering active galactic nuclei (AGN). Previous studies of AGN in simulations have mainly focused on major interactions (i.e. between approximately equal mass galaxies), which produce the strongest inflows and, therefore, would be the most likely to trigger AGN activity. How

Pasquale Filice, Marco Schirò, Giacomo Mazza

We introduce time-dependent variational principles to study the non-unitary dynamics of open quantum many-body systems, including dynamics described by the full Lindblad master equation, the non-Hermitian dynamics corresponding to the no-click limit of the fully post-selected quantum trajectories, and the dynamics described by a hybrid Lindbladian with a con

Zijian Jin, Jaehyon Rhee

This paper builds upon ParamANN's novel approach (S. Pal & R. Saha 2024) of using ANNs to infer cosmological density parameters by determining optimal architecture for varying synthetic Hubble data SNRs in estimating the density parameters $\Omega_{m, 0}$ and $\Omega_{\Lambda, 0}$ across redshift values $z \in [0, 1]$. To generate the synthetic data, this st

Benjamin Friedman

Given a (possibly non-K\"ahler) Calabi--Yau threefold $(X,\Omega)$, we introduce the notion of a (perturbed) special Lagrangian (SL) submanifold of $(X,\omega,\Omega)$, where $\omega$ is a Hermitian metric on $X$. The equations defining this class of submanifolds reduce to the usual SL equations when $\omega$ is a K\"ahler metric. Using the Sard--Smale techn

Jon Donnelly, Srikar Katta, Emanuele Borgonovo, Cynthia Rudin

Variable importance (VI) methods are often used for hypothesis generation, feature selection, and scientific validation. In the standard VI pipeline, an analyst estimates VI for a single predictive model with only the observed features. However, the importance of a feature depends heavily on which other variables are included in the model, and essential vari

Hang Yu, Julian Jordan, Julian Schmidt, Silvan Lindner

Safe and interpretable motion planning in complex urban environments needs to reason about bidirectional multi-agent interactions. This reasoning requires to estimate the costs of potential ego driving maneuvers. Many existing planners generate initial trajectories with sampling-based methods and refine them by optimizing on learned predictions of future env

Myles Foley, Sergio Maffeis, Muhammad Fakhrur Rozi, Takeshi Takahashi

JavaScript engines are widely used in web browsers, PDF readers, and server-side applications. The rise in concern over their security has led to the development of several targeted fuzzing techniques. However, existing approaches use random selection to determine where to perform mutations in JavaScript code. We postulate that the problem of selecting bette

Tommaso Giovannini, Pablo Grobas Illobre, Piero Lafiosca, Luca Nicoli

We present the first public release of plasmonX, a novel open-source code for simulating the plasmonic response of complex nanostructures. The code supports both fully atomistic and implicit descriptions of nanomaterials. In particular, it employs the frequency-dependent fluctuating charges ($\omega$FQ) and dipoles ($\omega$FQF$\mu$) models to describe the r

Wei Song, Xiao-Yu Yan, Xin Yu, Desheng Wu

We investigate electron pairing in a super clean kagome superconductor CsV3Sb5 with a residual resistivity ratio (RRR) of 290. By using the dilution-refrigerator-based scanning tunneling microscopy (STM) at the Synergetic Extreme Condition User Facility (SECUF), we find that the pairing gap exhibits chiral 2x2 modulations, and their chirality can be controll

Roberto Garrone

We propose a two-level information-theoretic framework for characterizing the informational organization of Agent-Based Model (ABM) dynamics within the broader paradigm of Complex Adaptive Systems (CAS). At the macro level, a pooled $\varepsilon$-machine is reconstructed as a reference model summarizing the system-wide informational regime. At the micro leve

Minjae Lee, Minsuk Kahng

Large Language Models (LLMs) are increasingly embedded in applications, and people can shape model behavior by editing prompt instructions. Yet encoding subtle, domain-specific policies into prompts is challenging. Although this process often benefits from concrete test cases, test data and prompt instructions are typically developed as separate artifacts, r

Anas Abouaomar, Mohammed El hanjri, Abdellatif Kobbane, Anis Laouiti

In this paper, we presents a novel hierarchical federated learning architecture specifically designed for smart agricultural production systems and crop yield prediction. Our approach introduces a seasonal subscription mechanism where farms join crop-specific clusters at the beginning of each agricultural season. The proposed three-layer architecture consist

Juha Harviainen, Frank Sommer, Manuel Sorge

Algorithms for learning decision trees often include heuristic local-search operations such as (1) adjusting the threshold of a cut or (2) also exchanging the feature of that cut. We study minimizing the number of classification errors by performing a fixed number of a single type of these operations. Although we discover that the corresponding problems are

Daniel Cunha Oliveira, Grover Guzman, Nick Firoozye

Robust optimization provides a principled framework for decision-making under uncertainty, with broad applications in finance, engineering, and operations research. In portfolio optimization, uncertainty in expected returns and covariances demands methods that mitigate estimation error, parameter instability, and model misspecification. Traditional approache

Xin Fei, Zhixuan Xu, Huaicong Fang, Tianrui Zhang

Dexterous grasping remains a central challenge in robotics due to the complexity of its high-dimensional state and action space. We introduce T(R,O) Grasp, a diffusion-based framework that efficiently generates accurate and diverse grasps across multiple robotic hands. At its core is the T(R,O) Graph, a unified representation that models spatial transformati

Aditya Khanna

Many identities involving symmetric functions can be proved through bijective manipulations of tableaux. In this paper, we prove identities involving polysymmetric functions through bijections and sign-reversing involutions. In their paper titled "Polysymmetric functions and motivic measures of configuration spaces", Asvin G and Andrew O'Desky introduced the

Xi Yu, Yang Yang, Qun Liu, Yonghua Du

Cellular image segmentation is essential for quantitative biology yet remains difficult due to heterogeneous modalities, morphological variability, and limited annotations. We present GenCellAgent, a training-free multi-agent framework that orchestrates specialist segmenters and generalist vision-language models via a planner-executor-evaluator loop (choose

Nadine El-Naggar, Tatsuki Kuribayashi, Ted Briscoe

Whether language models (LMs) have inductive biases that favor typologically frequent grammatical properties over rare, implausible ones has been investigated, typically using artificial languages (ALs) (White and Cotterell, 2021; Kuribayashi et al., 2024). In this paper, we extend these works from two perspectives. First, we extend their context-free AL for

Dayin Gou, Sanghyun Byun, Nilesh Malpeddi, Gabrielle De Micheli

Large Language Models (LLMs) typically rely on a large number of parameters for token embedding, leading to substantial storage requirements and memory footprints. In particular, LLMs deployed on edge devices are memory-bound, and reducing the memory footprint by compressing the embedding layer not only frees up the memory bandwidth but also speeds up infere

Ziyang Ma, Ruiyang Xu, Zhenghao Xing, Yunfei Chu

Fine-grained perception of multimodal information is critical for advancing human-AI interaction. With recent progress in audio-visual technologies, Omni Language Models (OLMs), capable of processing audio and video signals in parallel, have emerged as a promising paradigm for achieving richer understanding and reasoning. However, their capacity to capture a

Matthew Adrian, Yunsie Chung, Kevin Boyd, Saee Paliwal

Chemical pretrained models, sometimes referred to as foundation models, are receiving considerable interest for drug discovery applications. The general chemical knowledge extracted from self-supervised training has the potential to improve predictions for critical drug discovery endpoints, including on-target potency and ADMET properties. Multi-task learnin

Poornendu Kumar, Jeet Sampat

We characterize the zero sets of functions in the Schur--Agler class over the unit polydisk as well as functions in the unit ball of the multiplier algebra of the Drury--Arveson space via operators associated with a unitary realization formula for these functions. To this end, new notions of `eigenvalues' for tuples of operators are introduced, where the eig

Se Hwan Jeon, Ho Jae Lee, Seungwoo Hong, Sangbae Kim

Model Predictive Control (MPC) provides interpretable, tunable locomotion controllers grounded in physical models, but its robustness depends on frequent replanning and is limited by model mismatch and real-time computational constraints. Reinforcement Learning (RL), by contrast, can produce highly robust behaviors through stochastic training but often lacks

Oli Jones, Alan Logan

We investigate fixed subgroups of automorphisms of generalised Baumslag-Solitar (GBS) groups. Our main results are for automorphisms leaving a Bass-Serre tree invariant, under the assumption that all edge stabilisers are strictly contained in the corresponding vertex stabilisers. We completely characterise which GBS groups admit such an automorphism with a f

Alejandro Lopez-Montes, Fereshteh Yousefirizi, Yizhou Chen, Yazdan Salimi

KEY WORDS: Artificial Intelligence (AI), Theranostics, Dosimetry, Radiopharmaceutical Therapy (RPT), Patient-friendly dosimetry KEY POINTS - The rapid evolution of radiopharmaceutical therapy (RPT) highlights the growing need for personalized and patient-centered dosimetry. - Artificial Intelligence (AI) offers solutions to the key limitations in current dos

Wissam Salhab, Darine Ameyed, Hamid Mcheick, Fehmi Jaafar

Robustness in AI systems refers to their ability to maintain reliable and accurate performance under various conditions, including out-of-distribution (OOD) samples, adversarial attacks, and environmental changes. This is crucial in safety-critical systems, such as autonomous vehicles, transportation, or healthcare, where malfunctions could have severe conse

Xingang Guo, Utkarsh Tyagi, Advait Gosai, Paula Vergara

Multimodal Large Language Models (MLLMs) are increasingly applied in real-world scenarios where user-provided images are often imperfect, requiring active image manipulations such as cropping, editing, or enhancement to uncover salient visual cues. Beyond static visual perception, MLLMs must also think with images: dynamically transforming visual content and

Baicheng Li, Dong Wu, Zike Yan, Xinchen Liu

Pre-trained Vision-Language-Action (VLA) models represent a major leap towards general-purpose robots, yet efficiently adapting them to novel, specific tasks in-situ remains a significant hurdle. While reinforcement learning (RL) is a promising avenue for such adaptation, the process often suffers from low efficiency, hindering rapid task mastery. We introdu

Lin Lin, Jiefeng Long, Zhihe Wan, Yuchi Wang

Multimodal embedding models aim to yield informative unified representations that empower diverse cross-modal tasks. Despite promising developments in the evolution from CLIP-based dual-tower architectures to large vision-language models, prior works still face unavoidable challenges in real-world applications and business scenarios, such as the limited moda

Imran Khan

Large Language Models (LLMs) are increasingly being deployed as the reasoning engines for agentic AI systems, yet they exhibit a critical flaw: a rigid adherence to explicit rules that leads to decisions misaligned with human common sense and intent. This "rule-rigidity" is a significant barrier to building trustworthy autonomous agents. While prior work has

Boyana Martinova

By adapting methods of Ein-Erman-Lazarsfeld, we prove an analogue of the Ein-Lazarsfeld result on asymptotic syzygies for Veronese embeddings, in the setting of weighted projective spaces of the form $\mathbb{P}(1^n,2)$.

Víctor Navarro-Fernández, David Villringer

We study the 3D magnetohydrodynamics (MHD) equations in an annular cylinder, perturbed around the explicit steady state given by the 3D Taylor-Couette velocity field and zero magnetic field. Combining a recent linear instability result for the magnetic field with the framework of Friedlander, Pavlovi\'c and Shvydkoy [Comm. Math. Phys. 264 (2006), no. 2, 335-

Robin Bartlett, Tomasz Przezdziecki, Lukas Tappeiner

We construct an analogue of Gerasimov-Kharchev-Lebedev-Oblezin (GKLO) representations for twisted Yangians of type $\mathsf{AI}$, using the recently found current presentation of these algebras due to Lu, Wang and Zhang. These new representations allow us to define interesting truncations of twisted Yangians, which, in the spirit of Ciccoli-Drinfeld-Gavarini

Shelley Zixin Shu, Haozhe Luo, Alexander Poellinger, Mauricio Reyes

Transformer-based deep learning models have demonstrated exceptional performance in medical imaging by leveraging attention mechanisms for feature representation and interpretability. However, these models are prone to learning spurious correlations, leading to biases and limited generalization. While human-AI attention alignment can mitigate these issues, i

Mattia Grasselli, Angelo Porrello, Carlo Augusto Grazia

Autonomous driving remains a challenging task, particularly due to safety concerns. Modern vehicles are typically equipped with expensive sensors such as LiDAR, cameras, and radars to reduce the risk of accidents. However, these sensors face inherent limitations: their field of view and line of sight can be obstructed by other vehicles, thereby reducing situ

Ansh Tiwari, Ayush Chauhan

We introduce Shielded RecRL, a reinforcement learning approach to generate personalized explanations for recommender systems without sacrificing the system's original ranking performance. Unlike prior RLHF-based recommender methods that directly optimize item rankings, our two-tower architecture keeps the recommender's ranking model intact while a language m

Cedric Richter, Heike Wehrheim

Automatic software verifiers have become increasingly effective at the task of checking software against (formal) specifications. Yet, their adoption in practice has been hampered by the lack of such specifications in real world code. Large Language Models (LLMs) have shown promise in inferring formal postconditions from natural language hints embedded in co

Jacob Mercer

We introduce and analyse a two-sided branching-selection particle system which generalises the well-known $N$-particle branching Brownian motion ($N$-BBM) model, which we call the $(N,p)$-BBM, where either the leftmost or rightmost particle is deleted at each branching event according to a parameter $p\in(0,1)$. We establish that, as $N\to\infty$, the empiri

Vicente Bosca, Tatum Rask, Sunia Tanweer, Andrew R. Tawfeek

This paper explores the topological signatures of ReLU neural network activation patterns. We consider feedforward neural networks with ReLU activation functions and analyze the polytope decomposition of the feature space induced by the network. Mainly, we investigate how the Fiedler partition of the dual graph and show that it appears to correlate with the

Sunny Yu, Ahmad Jabbar, Robert Hawkins, Dan Jurafsky

Different open-ended generation tasks require different degrees of output diversity. However, current LLMs are often miscalibrated. They collapse to overly homogeneous outputs for creative tasks and hallucinate diverse but incorrect responses for factual tasks. We argue that these two failure modes are unified by, and can both be addressed by, the notion of

Muhammad Mateen Yaqoob, Kulsoom Fatima, Shahab Shamshirband, Amir Mosavi

This paper presents a protocol for enhancement of life time of WBAN network as well other protocol related issues such as throughput, path loss, and residual energy. Bio-sensors are used for deployment on human body. Poisson distribution and equilibrium model techniques have been used for attaining the required results. Multi-hop network topology and random

Tianyu Hu, Zhen Tan, Song Wang, Huaizhi Qu

With advancements in reasoning capabilities, Large Language Models (LLMs) are increasingly employed for automated judgment tasks. While LLMs-as-Judges offer promise in automating evaluations, current approaches often rely on simplistic aggregation methods (e.g., majority voting), which can fail even when individual agents provide correct answers. To address

Lennart Dabelow

We consider quenches of a quantum system that is prepared in a canonical equilibrium state of one Hamiltonian and then evolves unitarily in time under a different Hamiltonian. Technically, our main result is a systematic expansion of the pre- and post-quench canonical ensembles in the quench strength. We first demonstrate how this can be used to predict the

R. N. Sahoo, M. Tessler, S. Halfon, Y. Kashiv

We report on experiments at the Soreq Applied Research Accelerator Facility - Liquid-Lithium Target (SARAF-LiLiT) laboratory dedicated to the study of s-process neutron capture reactions. The kW-power proton beam at 1.92 MeV (1-2 mA) from SARAF Phase I yields high-intensity 30 keV quasi-Maxwellian neutrons (3-5x10^10 n/s). The high neutron intensity enables

John Wimarsson, Fabio Ferrari, Martin Jutzi

Recent direct observations of atypically shaped rubble-pile satellites of sub-km asteroids in form of the spherically oblate Dimorphos and bilobate Selam challenge classical binary asteroid formation theories, which only explain the predominantly elongated population. This study further explores a rubble-pile satellite formation scenario for binary asteroid

Hanyang Chen, Mark Zhao, Rui Yang, Qinwei Ma

Recent advances in embodied AI highlight the potential of vision language models (VLMs) as agents capable of perception, reasoning, and interaction in complex environments. However, top-performing systems rely on large-scale models that are costly to deploy, while smaller VLMs lack the necessary knowledge and skills to succeed. To bridge this gap, we present

Sarina Xi, Orelia Pi, Miaomiao Zhang, Becca Xiong

There is growing interest in applying artificial intelligence (AI) to automate and support complex decision-making tasks. However, it remains unclear how algorithms compare to human judgment in contexts requiring semantic understanding and domain expertise. We examine this in the context of the judge assignment problem, matching submissions to suitably quali

Danial Hosseintabar, Fan Chen, Giannis Daras, Antonio Torralba

Diffusion models have emerged as powerful generative priors for high-dimensional inverse problems, yet learning them when only corrupted or noisy observations are available remains challenging. In this work, we propose a new method for training diffusion models with Expectation-Maximization (EM) from corrupted data. Our proposed method, DiffEM, utilizes cond

Carolina Crosta, Riccardo Nardin, Patrick Daoust, Stefano Achilli

Atomic-scale crystal defects in Si are quantum-light sources offering tantalizing integration with existing photonic technologies. Yet, the controlled creation of near-infrared color centers for long- haul quantum communication and information still remains a challenge. In this work, we utilize light ions, such as H+ and He+, to gently generate quantum emitt

Suyash Fulay, Jocelyn Zhu, Michiel Bakker

Large language models (LLMs) have shown promising accuracy in predicting survey responses and policy preferences, which has increased interest in their potential to represent human interests in various domains. Most existing research has focused on "behavioral cloning", effectively evaluating how well models reproduce individuals' expressed preferences. Draw

Fernand Pelletier, Patrick Cabau

The purpose of this paper is to propose a version of the notion of convenient Lie groupoid as a generalization of this concept in finite dimension. The authors point out which obstructions appear in the infinite dimensional context and how an adapted notion of "bi-algebroid " in finite dimension (cf. \cite{MaXu94}) can be, nevertheless, recovered. This paper

Kunyu Peng, Di Wen, Kailun Yang, Jia Fu

Open-Set Domain Generalization (OSDG) aims to enable deep learning models to recognize unseen categories in new domains, which is crucial for real-world applications. Label noise hinders open-set domain generalization by corrupting source-domain knowledge, making it harder to recognize known classes and reject unseen ones. While existing methods address OSDG

Muhammad Ayub Sabir, Junbiao Pang, Jiaqi Wu, Fatima Ashraf

Abnormal stop detection (ASD) in intercity coach transportation is critical for ensuring passenger safety, operational reliability, and regulatory compliance. However, two key challenges hinder ASD effectiveness: sparse GPS trajectories, which obscure short or unauthorized stops, and limited labeled data, which restricts supervised learning. Existing methods

Runyao Yu, Ruochen Wu, Yongsheng Han, Jochen L. Cremer

Accurate probabilistic forecasting of intraday electricity prices is critical for market participants to inform trading decisions. Existing studies rely on specific domain features, such as Volume-Weighted Average Price (VWAP) and the last price. However, the rich information in the orderbook remains underexplored. Furthermore, these approaches are often dev

Alvaro Belmonte-Baeza, Miguel Cazorla, Gabriel J. García, Carlos J. Pérez-Del-Pulgar

Robotics plays a pivotal role in planetary science and exploration, where autonomous and reliable systems are crucial due to the risks and challenges inherent to space environments. The establishment of permanent lunar bases demands robotic platforms capable of navigating and manipulating in the harsh lunar terrain. While wheeled rovers have been the mainsta

Nathan G. Drouillard, TJ Hammond

Although solids have been recently used in ultrafast experiments for spectral broadening due to their relatively high nonlinearity, their sensitivity to damage limits their long-term stability. Liquids are a possible alternative to solids as a nonlinear medium because of their comparable nonlinearity and resistance to permanent damage. We generate a supercon

J. Khatua, S. M. Kumawat, G. Senthil Murugan, C. -L. Huang

Quantum magnets offer a unique platform for exploring exotic quantum phases and quantum phase transitions through external magnetic fields. A prominent example is the field-induced Bose--Einstein condensation (BEC) of magnons near the saturation field. While this behavior has been observed in low-spin systems, its realization in high-spin, quasi-two-dimensio

Guo Qin, Zhi Chen, Yong Liu, Zhiyuan Shi

Time Series Foundation Models (TSFMs) have shown significant impact through their model capacity, scalability, and zero-shot generalization. However, due to the heterogeneity of inter-variate dependencies and the backbone scalability on large-scale multivariate datasets, most TSFMs are typically pre-trained on univariate time series. This limitation renders

Shouren Wang, Wang Yang, Xianxuan Long, Qifan Wang

Hybrid thinking enables LLMs to switch between reasoning and direct answering, offering a balance between efficiency and reasoning capability. Yet our experiments reveal that current hybrid thinking LLMs only achieve partial mode separation: reasoning behaviors often leak into the no-think mode. To understand and mitigate this, we analyze the factors influen

Zefu Lin, Wenbo Chen, Xiaojuan Jin, Yuran Yang

Unmanned Aerial Vehicle (UAV) swarm systems necessitate efficient collaborative perception mechanisms for diverse operational scenarios. Current Bird's Eye View (BEV)-based approaches exhibit two main limitations: bounding-box representations fail to capture complete semantic and geometric information of the scene, and their performance significantly degrade

Herbert Spohn

We consider the asymmetric version of the Popkov-Sch\"{u}tz two-lane lattice gas with general jump rates, subject to the stationary measure being of product form. This still leaves five free parameters. At density 1/2 the eigenvalues of the flux Jacobian are degenerate. We compute the second order expansion of the average fluxes at density 1/2 and thereby id

Marc-Antoine Roy, Thomas Pousset, Baptiste Royer

In order to achieve fault-tolerant quantum computing, we make use of quantum error correction schemes designed to protect the logical information of the system from decoherence. A promising way to preserve such information is to use the multimode Gottesman-Kitaev-Preskill (GKP) encoding, which encodes logical qubits into several harmonic oscillators. In this

Sergey Kiselev

Recent results on short-lived hadronic resonances obtained with the ALICE detector at LHC energies are presented. These results include masses, widths, transverse momentum spectra, yields, and ratios of resonance yields to longer-lived ground-state particles, and elliptic flows. The results are compared with model predictions and measurements at lower energi

Corey Jones, Emily McGovern

Discrete, unimodular inclusions of factors $(N\subseteq M, E)$ with $N$ of type $\rm{II}_{1}$ have a natural notion of standard invariant, generalizing the finite index case. When the unitary tensor category of $N$-$N$ bimodules generated by $_{N}L^{2}(M, \tau\circ E)_{N}$ is equivalent to the Temperley-Lieb-Jones category $\text{TLJ}(\delta)$, the associate

Noam Soker

I examine recent observations of the type Ia supernova remnant (SNR Ia) Tycho and conclude that Tycho is an SN Ia inside a planetary nebula (SNIP), strengthening such a previous suggestion from 1985. The observations reveal two opposite protrusions, termed ears, projected on the main shell of Tycho. The pair of ear structures resembles that of the SNRs Ia Ke

Ruben Belo, Marta Guimaraes, Claudia Soares

Large Language Models are susceptible to jailbreak attacks that bypass built-in safety guardrails (e.g., by tricking the model with adversarial prompts). We propose Concept Alignment and Concept Manipulation CALM, an inference-time method that suppresses harmful concepts by modifying latent representations of the last layer of the model, without retraining.

Julen Costa-Watanabe, Isabelle Wittmann, Benedikt Blumenstiel, Konrad Schindler

Earth observation (EO) satellites produce massive streams of multispectral image time series, posing pressing challenges for storage and transmission. Yet, learned EO compression remains fragmented and lacks publicly available, large-scale pretrained codecs. Moreover, prior work has largely focused on image compression, leaving temporal redundancy and EO vid

Kaiwen He, Petros Drineas, Rajiv Khanna

Spectral clustering is a fundamental method for graph partitioning, but its reliance on eigenvector computation limits scalability to massive graphs. Classical sparsification methods preserve spectral properties by sampling edges proportionally to their effective resistances, but require expensive preprocessing to estimate these resistances. We study whether

Minghao Tang, Shiyu Ni, Jingtong Wu, Zengxin Han

Context-grounded generation underpins many LLM applications, including long-document question answering (QA), conversational personalization, and retrieval-augmented generation (RAG). However, classic token-based context concatenation is costly for long inputs and can be lost in the middle at extreme context lengths. Recent work explores context parameteriza

Prasenjit K Mudi, Anshi Sachan, Dahlia Devapriya, Sheetal Kalyani

Whisper models have achieved remarkable progress in speech recognition; yet their large size remains a bottleneck for deployment on resource-constrained edge devices. This paper proposes a framework to design fine-tuned variants of Whisper which address the above problem. Structured sparsity is enforced via the Sparse Group LASSO penalty as a loss regularize

Thomas Rivasseau

Modern web and digital application password storage relies on password hashing for storage and security. Ad-hoc upgrade of password storage to keep up with hash algorithm norms may be used to save costs but can introduce unforeseen vulnerabilities. This is the case in the password storage scheme used by Meta Platforms which services several billion monthly u

Alexander Nazarov, Sergey Repin

The paper is concerned with a posteriori estimates for approximations of boundary value problems generated by the spectral fractional Laplace operator. The derivation is based upon the Stinga--Torrea extension, which generalizes the Caffarelli--Silvestre extension and transfers the corresponding nonlocal problem in a bounded domain to a local problem of high

Oz Gitelson, Satya Prakash Nayak, Ritam Raha, Anne-Kathrin Schmuck

We present a novel framework for human-robot \emph{logical} interaction that enables robots to reliably satisfy (infinite horizon) temporal logic tasks while effectively collaborating with humans who pursue independent and unknown tasks. The framework combines two key capabilities: (i) \emph{maximal adaptation} enables the robot to adjust its strategy \emph{

Roman Lavička, Paul Alois Bühler

The anomalous magnetic moment of the tau lepton ($a_{\tau}$) is a sensitive probe for the search for deviations from the Standard Model predictions and thus for new physics. This study investigates the feasibility of measuring $a_{\tau}$ using ultra-peripheral collisions (UPCs) at the LHC, where photon-photon interactions ($\gamma\gamma \to \tau^+ \tau^-$) p

Shuhei Tarashima, Yushan Wang, Norio Tagawa

In this work, we aim to develop simple and efficient models for human mesh recovery (HMR) and its predecessor task, human pose estimation (HPE). State-of-the-art HMR methods, such as HMR2.0 and its successors, rely on large, non-hierarchical vision transformers as encoders, which are inherited from the corresponding HPE models like ViTPose. To establish base

Anirudh Pradhan, A. Husain, M. Zeyauddin, S. H. Shekh

Late-time cosmic acceleration has motivated the exploration of various extensions of general relativity, among which $f(Q,\mathcal{T})$ gravity, based on the non-metricity scalar $Q$ and the trace of the energy--momentum tensor $\mathcal{T}$, has gained increasing attention. In this study, we explore the thermodynamic aspects of $f(Q,\mathcal{T})$ gravity by

Chih-Chuan Cheng, Yi-Ju Tseng

We propose SG-XDEAT (Sparsity-Guided Cross Dimensional and Cross-Encoding Attention with Target Aware Conditioning), a novel framework designed for supervised learning on tabular data. At its core, SG-XDEAT employs a dual-stream encoder that decomposes each input feature into two parallel representations: a raw value stream and a target-conditioned (label-aw

Surajit Mondal, Bin Chen, Sijie Yu, Xingyao Chen

Decades of solar coronal observations have provided substantial evidence for accelerated particles in the corona. In most cases, the location of particle acceleration can be roughly identified by combining high spatial and temporal resolution data from multiple instruments across a broad frequency range. In almost all cases, these nonthermal particles are as

Stefano Riolo, Edoardo Rizzi

We build a non-compact, orientable, hyperbolic four-manifold of finite volume that does not admit any spin structure.

Nischal Binod Gautam, Enrique P. Blair

Molecular quantum-dot Cellular Automata (QCA) may provide low-power, high-speed computational hardware for processing classical information. Simulation and modeling play an important role in the design of QCA circuits because fully-coherent models of QCA scale exponentially with the number of devices, and such models are severely limited in size. For larger

William Schertzer, Mohamed Al Otmi, Janani Sampath, Ryan P. Lively

The global transition to hydrogen-based energy infrastructures faces significant hurdles. Chief among these are the high costs and sustainability issues associated with acid-based proton exchange membrane fuel cells. Anion exchange membrane (AEM) fuel cells offer promising cost-effective alternatives, yet their widespread adoption is limited by rapid degrada

Jesús A. Toalá, Diego A. Vasquez-Torres

Symbiotic stars are binary systems where a white dwarf (WD) accretes material from the wind of an evolved, late-type companion. X-ray-emitting symbiotic systems are classified into $\alpha$, $\beta$, $\delta$, and $\beta/\delta$ types, attributed to distinct physical mechanisms such as thermonuclear burning, wind interactions, and accretion-driven boundary l

Shaofei Li, Xiao Han, Ziqi Zhang, Minyao Hua

As e-commerce platforms develop, fraudulent activities are increasingly emerging, posing significant threats to the security and stability of these platforms. Promotion abuse is one of the fastest-growing types of fraud in recent years and is characterized by users exploiting promotional activities to gain financial benefits from the platform. To investigate