May 2025 arXiv papers — page 86

F. Marin

Herbig Ae/Be stars are young, pre-main-sequence stars that provide critical insights into the processes of stellar formation, early stellar evolution and protoplanetary disks.Two of the key features of such stars are their circumstellar dusty disk and bipolar ionized outflows, which are key components for understanding planet formation processes and energy/m

Gennie Nguyen, Lei Wang, Yangxueqing Jiang, Tom Gedeon

Misinformation poses significant risks to public opinion, health, and security. While most fake news detection methods rely on text analysis, little is known about how people physically respond to false information or repeated exposure to the same statements. This study investigates whether wearable sensors can detect belief in a statement or prior exposure

Jie Cao

This paper is devoted to study the equilibrium states for almost-additive potentials defined over topologically mixing countable Markov shifts (that is a non-compact space) without the big images and preimages (BIP) property. Let $\F$ be an almost-additive and summable potential with bounded variation potential. We prove that there exists an unique equilibri

Andrea Richaud, Pietro Massignan

We present a method for measuring the superfluid fraction of a Bose-Einstein condensate (BEC) without relying on external perturbations or imposed optical lattices. Our approach leverages the intrinsic rotation of vortex necklaces in one component of a binary superfluid mixture, where the vortex cores act as effective potential wells for the second component

Chenglong Yu, Zhiwei Zheng, Yiming Zhong

In this paper, we study moduli spaces of sextic curves with simple singularities. Through period maps of K3 surfaces with ADE singularities, we prove that such moduli spaces admit algebraic open embeddings into arithmetic quotients of type IV domains. For all cases, we prove the identifications of GIT compactifications and Looijenga compactifications. We als

Lucia Coto-Elena, J. E. Maese, L. Merino, F. Caballero

This paper presents a new approach for 6DoF Direct LiDAR-Inertial Odometry (D-LIO) based on the simultaneous mapping of truncated distance fields on CPU. Such continuous representation (in the vicinity of the points) enables working with raw 3D LiDAR data online, avoiding the need of LiDAR feature selection and tracking, simplifying the odometry pipeline and

Phillip Mueller, Jannik Wiese, Sebastian Mueller, Lars Mikelsons

Datasets in engineering domains are often small, sparsely labeled, and contain numerical as well as categorical conditions. Additionally. computational resources are typically limited in practical applications which hinders the adoption of generative models for engineering tasks. We introduce a novel masked-conditioning approach, that enables generative mode

Konstantinos Barmpas, Na Lee, Yannis Panagakis, Dimitrios A. Adamos

Recent advances in large-scale pre-trained Electroencephalogram (EEG) models have shown great promise, driving progress in Brain-Computer Interfaces (BCIs) and healthcare applications. However, despite their success, many existing pre-trained models have struggled to fully capture the rich information content of neural oscillations, a limitation that fundame

Thibaud Gloaguen, Robin Staab, Nikola Jovanović, Martin Vechev

Most LLM fingerprinting methods teach the model to respond to a few fixed queries with predefined atypical responses (keys). This memorization often does not survive common deployment steps such as finetuning or quantization, and such keys can be easily detected and filtered from LLM responses, ultimately breaking the fingerprint. To overcome these limitatio

Himanshu Beniwal, Youngwoo Kim, Maarten Sap, Soham Dan

As large language models (LLMs) become increasingly prevalent in global applications, ensuring that they are toxicity-free across diverse linguistic contexts remains a critical challenge. We explore "Cross-lingual Detoxification", a cross-lingual paradigm that mitigates toxicity, enabling detoxification capabilities to transfer between high and low-resource

Giuseppe La Scala

We consider a generic, suitable class of optimal control problems under a constraint given by a finite-dimensional SDE-ODE system, describing a system of two interacting species of particles: the herd, described by SDEs, and the herders, described by ODEs with the addition of a control function. In particular, we firstly show that for a low number of herders

Magnús M. Halldórsson, Nicolaos Matsakis, Pavel Veselý

We improve the space bound for streaming approximation of Diameter but also of Farthest Neighbor queries, Minimum Enclosing Ball and its Coreset, in high-dimensional Euclidean spaces. In particular, our deterministic streaming algorithms store $\mathcal{O}(\varepsilon^{-2}\log(\frac{1}{\varepsilon}))$ points. This improves by a factor of $\varepsilon^{-1}$ t

Olcay Coşkun, Deniz Yılmaz

We introduce and study the category of $p$-bifree biset functors for a fixed prime $p$, defined via bisets whose left and right stabilizers are $p'$-groups. This category naturally lies between the classical biset functors and the diagonal $p$-permutation functors, serving as a bridge between them. Every biset functor and every diagonal $p$-permutation funct

Paul Barry

In this note, we show how to define certain Riordan arrays, that we call the Fuss-Catalan-Riordan arrays, by means of a special family of $d$-orthogonal polynomials. We relate the Fuss-Catalan Riordan arrays to the Fuss Catalan numbers, and to certain lattice paths. We emphasise the role of the production matrices of the Riordan arrays that we encounter in o

T. I. Khalilov, V. S. Beskin, V. I. Pariev

The observed power-law spectra of relativistic jets from active galactic nuclei clearly indicate a synchrotron mechanism of radiation by particles that similarly possess a power-law energy spectrum. However, the issue of their angular anisotropy has not been given sufficient attention until recently, although the example of the solar wind (where a strongly m

Moritz Stargalla, Christoph Hertrich, Daniel Reichman

An established measure of the expressive power of a given ReLU neural network is the number of linear regions into which it partitions the input space. There exist many different, non-equivalent definitions of what a linear region actually is. We systematically assess which papers use which definitions and discuss how they relate to each other. We then analy

Kean Chen, Qisheng Wang, Zhan Yu, Zhicheng Zhang

We consider a fundamental task in quantum information theory, estimating the values of $\operatorname{tr}(O\rho)$, $\operatorname{tr}(O\rho^2)$, ..., $\operatorname{tr}(O\rho^k)$ for an observable $O$ and a quantum state $\rho$. We show that $\widetilde\Theta(k)$ samples of $\rho$ are sufficient and necessary to simultaneously estimate all the $k$ values. Th

M. Jelínek, S. A. Grebenev, P. Yu. Minaev, C. C. Thöne

Long gamma-ray bursts (GRBs) are characterized by a brief gamma-ray flash followed by a longer-lasting multiwavelength afterglow. The basic mechanism is largely understood, and the early afterglow evolution often shows complex features that provide crucial insights into the transition between prompt and afterglow phases. We present a detailed analysis of GRB

M. A. Matveev, A. T. Sitnikov, A. V. Sarantsev

We construct a formalism which describes the resonances decaying into four pseudoscalar meson final states. This method is fully covariant and can be directly applied for the partial-wave analysis of high statistical data. Two topologies of the process are considered: two intermediate resonances each decaying into two final mesons and cascade decay via three

Wenhao Li, Yuxin Zhang, Gen Luo, Daohai Yu

While long-context large language models (LLMs) exhibit remarkable document processing capabilities, their prohibitively high training costs often hinder customized applications. To mitigate this issue, we propose \textit{Sequential Chunk-wise Optimization} (SeCO), a memory-efficient training paradigm that partitions lengthy inputs into manageable chunks. Ea

Kai Hsiang Hsieh, Monyneath Yim, Jui Chiu Chiang

To encode point clouds containing both geometry and attributes, most learning-based compression schemes treat geometry and attribute coding separately, employing distinct encoders and decoders. This not only increases computational complexity but also fails to fully exploit shared features between geometry and attributes. To address this limitation, we propo

Jianfeng Deng, Qingfeng Chen, Debo Cheng, Jiuyong Li

Accurately predicting counterfactual user feedback is essential for building effective recommender systems. However, latent confounding bias can obscure the true causal relationship between user feedback and item exposure, ultimately degrading recommendation performance. Existing causal debiasing approaches often rely on strong assumptions-such as the availa

Yongliang Wu, Zonghui Li, Xinting Hu, Xinyu Ye

Recent advances in multi-modal generative models have enabled significant progress in instruction-based image editing. However, while these models produce visually plausible outputs, their capacity for knowledge-based reasoning editing tasks remains under-explored. In this paper, we introduce KRIS-Bench (Knowledge-based Reasoning in Image-editing Systems Ben

Lia Vas

The Graded Classification Conjecture (GCC) states that the pointed $K_0^{\operatorname{gr}}$-group is a complete invariant of the Leavitt path algebras of finite graphs when these algebras are considered with their natural grading by $\mathbb Z.$ The conjecture has previously been shown to hold in some special cases. The main result of the paper shows that t

Seonghwan Park, Jueun Mun, Donghyun Oh, Namhoon Lee

Concept bottleneck models (CBMs) ensure interpretability by decomposing predictions into human interpretable concepts. Yet the annotations used for training CBMs that enable this transparency are often noisy, and the impact of such corruption is not well understood. In this study, we present the first systematic study of noise in CBMs and show that even mode

Joás Venâncio, Herondy Mota, Azadeh Mohammadi

This study explores the finite temperature Casimir effect for a massive spinor field in cosmic dispiration spacetime, formed by the combination of a cosmic string and a screw dislocation using the generalized zeta function regularization method. First, we examine the cosmic string spacetime with a quasi-antiperiodic boundary condition, where the Casimir ener

Zeping Yu, Sophia Ananiadou

Although multimodal large language models (MLLMs) have achieved impressive performance, the multimodal instruction tuning stage often causes catastrophic forgetting of the base LLM's language ability, even in strong models like Llama3. To address this, we propose Locate-then-Merge, a training-free parameter fusion framework that first locates important param

Gennie Nguyen, Lei Wang, Yangxueqing Jiang, Tom Gedeon

Understanding how individuals physiologically respond to false information is crucial for advancing misinformation detection systems. This study explores the potential of using physiological signals, specifically electrodermal activity (EDA) and photoplethysmography (PPG), to classify both the veracity of information and its interaction with user belief. In

Ngo P. N. Ngoc, Gunter M. Schütz

We first survey some open questions concerning stochastic interacting particle systems with open boundaries. Then an asymmetric exclusion process with open boundaries that generalizes the lattice gas model of Katz, Lebowitz, and Spohn (KLS) is introduced and invariance of the one-dimensional Ising measure is proved. The stationary current is computed in expl

Xuanqi Gao, Siyi Xie, Juan Zhai, Shiqing Ma

As Large Language Models (LLMs) evolve from passive text generators to active reasoning agents capable of interacting with external tools, the Model Context Protocol (MCP) has emerged as a key standardized framework for dynamic tool discovery and orchestration. Despite its widespread industry adoption, existing evaluation methods do not adequately assess too

Bikash Kumar Acharya, Indra Kumar Banerjee, Ujjal Kumar Dey

We study in detail the effect of quantum decoherence in neutrino oscillations. We adopt a phenomenological approach that allows us to parametrize the energy dependence of the decoherence effects resulting from the modification of the neutrino dispersion relation. Using the open quantum system framework we derive decoherence parameters, which are usually conn

Xue-Ping Ren, Yue Hu, Long-Ye Lu, Xin-Ran Ma

The bulk-boundary correspondence plays a crucial role in topological quantum systems, however,this principle is broken in non-Hermitian systems. The breakdown of the bulk-boundary correspondence indicates that the global phase diagrams under open boundary conditions are significantly different from those under periodic boundary conditions. In this paper, we

Larissa Schmid, Tobias Hey, Martin Armbruster, Sophie Corallo

Large Language Models (LLMs) are used for many different software engineering tasks. In software architecture, they have been applied to tasks such as classification of design decisions, detection of design patterns, and generation of software architecture design from requirements. However, there is little overview on how well they work, what challenges exis

Preetam V. Tanikella, Laryssa Abdala, Karin Leiderman, Annie Green Howard

Cardiac digital twins (CDT) are emerging as a potentially transformative tool in cardiology. A critical yet understudied determinant of CDT accuracy is the His-Purkinje system (HPS), which influences ventricular depolarization and shapes the QRS complex of the electrocardiogram (ECG). Here, we quantify how structural variations in the HPS alter QRS morpholog

Liu Jiapeng, Liu Fucheng, Wu Jinpeng, Ren Chunpin

With the rapid growth of renewable energy being integrated, transmitted, and utilized in various forms, power systems are evolving from traditional metal-based infrastructures to advanced semiconductor-based technologies. Central to this transformation, power semiconductor devices now face the crucial demands of ultrahigh switching ability, presenting a long

Gouki Minegishi, Hiroki Furuta, Shohei Taniguchi, Yusuke Iwasawa

Transformer-based language models exhibit In-Context Learning (ICL), where predictions are made adaptively based on context. While prior work links induction heads to ICL through a sudden jump in accuracy, this can only account for ICL when the answer is included within the context. However, an important property of practical ICL in large language models is

Muhammad Zeeshan Mumtaz, Mohammadali Mohammadi, Hien Quoc Ngo, Michail Matthaiou

This paper investigates a discrete energy state transition model for energy harvesting (EH) in cell-free massive multiple-input-multiple-output (CF-mMIMO) networks. A Markov chain-based stochastic process is conceived to characterize the temporal evolution of the user equipment (UE) energy level by leveraging state transition probabilities (STP) based on the

Gabriella Gilli, Francisco González-Galindo, Jean-Yves Chaufray, Ehouarn Millour

It is still unknown how much water has escaped from Mars during its history. Hydrogen escape from Mars's atmosphere probably played a major role in drying the planet, but present-day Hloss rates (about 3x10^26 atoms per second on average) cannot explain the geological evidence for the large volumes of liquid water on ancient Mars. Here we used the three-dime

Advait Joglekar, Divyanshu Singh, Rooshil Rohit Bhatia, S. Umesh

Voice Conversion research in recent times has increasingly focused on improving the zero-shot capabilities of existing methods. Despite remarkable advancements, current architectures still tend to struggle in zero-shot cross-lingual settings. They are also often unable to generalize for speakers of unseen languages and accents. In this paper, we adopt a simp

Beier Luo, Shuoyuan Wang, Sharon Li, Hongxin Wei

Post-training of large language models is essential for adapting pre-trained language models (PLMs) to align with human preferences and downstream tasks. While PLMs typically exhibit well-calibrated confidence, post-trained language models (PoLMs) often suffer from over-confidence, assigning high confidence to both correct and incorrect outputs, which can un

Jean-Philippe Burelle, Mohamed Moussadek Maiza, Maxence Mayrand

We introduce a notion of deformations of quasi-Hamiltonian $G$-spaces to Hamiltonian $G$-spaces and provide several examples. In particular, we show that the double $G \times G$ of a Lie group, viewed as a quasi-Hamiltonian $G \times G$-space, deforms smoothly to the cotangent bundle $T^*G$. Likewise, any conjugacy class of $G$ sufficiently close to the iden

Hakar Raji, Oliver C. Schnürer

There exist rotationally symmetric translating solutions to mean curvature flow that can be written as a graph over Euclidean space. This result is well-known. Its proof uses the symmetry and techniques from partial differential equations. However, the result can also be formulated as an existence result for a singular ordinary differential equation. Here, w

Naifu Xue, Zhaoyang Jia, Jiahao Li, Bin Li

While recent diffusion-based generative image codecs have shown impressive performance, their iterative sampling process introduces unpleasing latency. In this work, we revisit the design of a diffusion-based codec and argue that multi-step sampling is not necessary for generative compression. Based on this insight, we propose OneDC, a One-step Diffusion-bas

Souradip Nath

Multimodal Large Language Models (MLLMs) like GPT-4V are capable of reasoning across text and image modalities, showing promise in a variety of complex vision-language tasks. In this preliminary study, we investigate the out-of-the-box capabilities of GPT-4V in the domain of image forensics, specifically, in detecting image splicing manipulations. Without an

Lars Benedikt Kaesberg, Jan Philip Wahle, Terry Ruas, Bela Gipp

Existing reasoning datasets saturate and fail to test abstract, multi-step problems, especially pathfinding and complex rule constraint satisfaction. We introduce SPaRC (Spatial Pathfinding Reasoning Challenge), a dataset of 1,000 2D grid pathfinding puzzles to evaluate spatial and symbolic reasoning, requiring step-by-step planning with arithmetic and geome

Corentin Dufourg, Charlotte Pelletier, Stéphane May, Sébastien Lefèvre

The Earth's surface is subject to complex and dynamic processes, ranging from large-scale phenomena such as tectonic plate movements to localized changes associated with ecosystems, agriculture, or human activity. Satellite images enable global monitoring of these processes with extensive spatial and temporal coverage, offering advantages over in-situ method

B. Petrović, T. Sato, S. Birner, J. Zanon

We investigate triple V-shaped type-II quantum wells designed to emit at 6-9 {\mu}m wavelength range, consisting of three InAs(Sb) electron quantum wells and two Ga0.6In0.4Sb hole quantum wells. The wells' composition and thicknesses are optimized in terms of wavefunction overlap and valence intersubband absorption (VISA) by using kp calculation. The triple

Stefan Haar, Juri Kolčák

The analysis of biological networks has benefited from the richness of Boolean networks (BNs) and the associated theory. These results have been further fortified in recent years by the emergence of Most Permissive (MP) semantics, combining efficient analysis methods with a greater capacity of explaining pathways to states hitherto thought unreachable, owing

Giovanni Pollo, Mohamed Amine Hamdi, Matteo Risso, Lorenzo Ruotolo

Virtual Platforms (VPs) enable early software validation of autonomous systems' electronics, reducing costs and time-to-market. While many VPs support both functional and non-functional simulation (e.g., timing, power), they lack the capability of simulating the environment in which the system operates. In contrast, robotics simulators lack accurate timing a

Qi Jun Zong, Haolin Wang, Qi Zhang, Xinle Cheng

Electrons can form an ordered solid crystal phase ascribed to the interplay between Coulomb repulsion and kinetic energy. Tuning these energy scales can drive a phase transition from electron solid to liquid, i.e. melting of Wigner crystal. Generalized Wigner crystals (GWCs) pinned to moire superlattices have been reported by optical and scanning-probe-based

Spiros Baxavanakis, Manos Schinas, Symeon Papadopoulos

Recent advancements in DeepFake generation, along with the proliferation of open-source tools, have significantly lowered the barrier for creating synthetic media. This trend poses a serious threat to the integrity and authenticity of online information, undermining public trust in institutions and media. State-of-the-art research on DeepFake detection has p

Filip Thor, Carl Nettelblad

Sequencing a genome to determine an individual's DNA produces an enormous number of short nucleotide subsequences known as reads, which must be reassembled to reconstruct the full genome. We present a method for analyzing this type of data using contrastive learning, in which an encoder model is trained to produce embeddings that cluster together sequences f

Jordan Dotzel, Tony Montes, Mohamed S. Abdelfattah, Zhiru Zhang

Traditional methods for 3D object compression operate only on structural information within the object vertices, polygons, and textures. These methods are effective at compression rates up to 10x for standard object sizes but quickly deteriorate at higher compression rates with texture artifacts, low-polygon counts, and mesh gaps. In contrast, semantic compr

Jimin Bai, Peter Keim, Matteo Baggioli

Long-wavelength collective shear dynamics are profoundly different in solids and liquids. According to the theoretical framework developed by Maxwell and Frenkel, collective shear waves vanish upon melting by acquiring a characteristic wave-vector gap, known as the $k$-gap. While this prediction has been supported by numerous simulations, experimental valida

Habib Ammari, Silvio Barandun, Bryn Davies, Erik Orvehed Hiltunen

We consider the spectral properties of aperiodic block subwavelength resonator systems in one dimension, with a primary focus on the density of states. We prove that for random block configurations, as the number of blocks $M\to \infty$, the integrated density of states converges to a non-random, continuous function. We show both analytically and numerically

Chen-Yu Liu

Variational quantum learning faces practical challenges in the noisy intermediate-scale quantum (NISQ) era. Parameterized quantum circuit (PQC) models suffer from statistical uncertainty due to finite-shot measurements and are highly sensitive to quantum noise, while purely classical approximations like neural quantum states (NQS) lack access to genuine quan

Wenwen Qiang, Jingyao Wang, Zeen Song, Jiangmeng Li

In this paper, we focus on the out-of-distribution (OOD) generalization of self-supervised learning (SSL). By analyzing the mini-batch construction during the SSL training phase, we first give one plausible explanation for SSL having OOD generalization. Then, from the perspective of data generation and causal inference, we analyze and conclude that SSL learn

Marcella Astrid, Abdelrahman Shabayek, Djamila Aouada

Batteries are essential for various applications, including electric vehicles and renewable energy storage, making safety and efficiency critical concerns. Anomaly detection in battery thermal images helps identify failures early, but traditional deep learning methods require extensive labeled data, which is difficult to obtain, especially for anomalies due

Huanjin Yao, Qixiang Yin, Jingyi Zhang, Min Yang

In this work, we aim to incentivize the reasoning ability of Multimodal Large Language Models (MLLMs) via reinforcement learning (RL) and develop an effective approach that mitigates the sparse reward and advantage vanishing issues during RL. To this end, we propose Share-GRPO, a novel RL approach that tackle these issues by exploring and sharing diverse rea

Elisa Varani

We study spinors in the framework of general relativity, starting from the Dirac field Lagrangian in the approximation of weak gravity. We focus on how fermions couple to gravity through the spin connection, and we analyze these couplings by analogy with the Ginzburg-Landau model and the Yukawa interaction known from the Higgs mechanism. By solving the field

Yun-Cheng Tsai, Samuel Yen-Chi Chen

Blockchain transaction data exhibits high dimensionality, noise, and intricate feature entanglement, presenting significant challenges for traditional clustering algorithms. In this study, we conduct a comparative analysis of three clustering approaches: (1) Classical K-Means Clustering, applied to pre-processed feature representations; (2) Hybrid Clustering

Lino Benedetto

This paper is dedicated to the spectral analysis of the semiclassical purely magnetic Laplacian on the plane in the situation where the magnetic field $B$ vanishes nondegenerately on an open smooth curve $\Gamma$. We prove the existence of a discrete spectrum for energy windows of the scale $h^{4/3}$ and give complete asymptotics in the semiclassical paramat

Xiaobei Yan, Yiming Li, Hao Wang, Han Qiu

Large language models (LLMs) are widely deployed, but their substantial compute demands make them vulnerable to inference cost attacks that aim to deliberately maximize the output length. In this work, we investigate a distinct attack surface: maximizing inference cost by tampering with the model parameters instead of inputs. This approach leverages the esta

Yee Li, Fan, Bernardo Palacios Muniz, Nayoung Kim

We experimentally investigate the impact of a flat, horizontal disk onto a boiling liquid, i.e., a liquid in thermal equilibrium with its vapor phase. We observe exceptionally high impact pressures deviating strongly from the inertial scaling found for impact in a non-condensable environment, coinciding with the rapid collapse of the vapor pocket entrapped b

Xinwei Yang, Zhaofeng Liu, Chen Huang, Jiashuai Zhang

While recent research increasingly emphasizes the value of human-LLM collaboration in competitive programming and proposes numerous empirical methods, a comprehensive understanding remains elusive due to the fragmented nature of existing studies and their use of diverse, application-specific human feedback. Thus, our work serves a three-fold purpose: First,

Sofia Zinzani, Robert M. Jones, Mirko Vanzan, Francesca Baletto

Hybrid metallic nanoalloys combining plasmonic and catalytic metals are essential for developing advanced photocatalysts. A promising design called core-satellites comprises a spherical nanogold dotted with smaller transition-metal clusters. While these nanoalloys' catalytic activity and hot-carriers generation have been extensively studied, their morphologi

Jinfeng Xu, Zheyu Chen, Jinze Li, Shuo Yang

The data sparsity problem significantly hinders the performance of recommender systems, as traditional models rely on limited historical interactions to learn user preferences and item properties. While incorporating multimodal information can explicitly represent these preferences and properties, existing works often use it only as side information, failing

Rebecca Maria Kuntz, Heinrich von Campe, Björn Malte Schäfer

We derive the information geometry induced by the statistical R\'enyi divergence, namely its metric tensor, its dual parametrized connections, as well as its dual Laplacians. Based on these results, we demonstrate that the R\'enyi-geometry, though closely related, differs in structure from Amari's well-known $\alpha$-geometry. Subsequently, we derive the can

Khoa Tran, Tri Le, Bao Huynh, Hung-Cuong Trinh

Accurate prediction of the Remaining Useful Life (RUL) is essential for enabling timely maintenance of lithium-ion batteries, impacting the operational efficiency of electric applications that rely on them. This paper proposes a RUL prediction approach that leverages data from recent charge-discharge cycles to estimate the number of remaining usable cycles.

Haihong Hao, Mingfei Han, Changlin Li, Zhihui Li

Embodied navigation demands comprehensive scene understanding and precise spatial reasoning. While image-text models excel at interpreting pixel-level color and lighting cues, 3D-text models capture volumetric structure and spatial relationships. However, unified fusion approaches that jointly fuse 2D images, 3D point clouds, and textual instructions face ch

Chuan Huang, Gustaf Hendeby, Isaac Skog

This paper presents a new method for jointly calibrating a magnetometer and inertial measurement unit (IMU), focusing on balancing calibration accuracy and computational efficiency. The proposed method is based on a maximum a posteriori estimation framework, treating both the calibration parameters and orientation trajectory of the sensors as unknowns. This

Shinnosuke Ono, Issey Sukeda, Takuro Fujii, Kosei Buma

We present a Japanese domain-specific language model for the pharmaceutical field, developed through continual pretraining on 2 billion Japanese pharmaceutical tokens and 8 billion English biomedical tokens. To enable rigorous evaluation, we introduce three new benchmarks: YakugakuQA, based on national pharmacist licensing exams; NayoseQA, which tests cross-

Chang Liu, Dongbo Wang, Liu liu, Zhixiao Zhao

This study addresses the challenges in intelligent processing of Chinese ancient mathematical classics by constructing Guji_MATH, a benchmark for evaluating classical texts based on Suanjing Shishu. It systematically assesses the mathematical problem-solving capabilities of mainstream reasoning models under the unique linguistic constraints of classical Chin

Kaiyu Guo, Tan Pan, Chen Jiang, Zijian Wang

Medical anomaly detection (AD) is crucial for early clinical intervention, yet it faces challenges due to limited access to high-quality medical imaging data, caused by privacy concerns and data silos. Few-shot learning has emerged as a promising approach to alleviate these limitations by leveraging the large-scale prior knowledge embedded in vision-language

Giuseppe Guarino, Matteo Ciotola, Gemine Vivone, Giovanni Poggi

Hyperspectral pansharpening has received much attention in recent years due to technological and methodological advances that open the door to new application scenarios. However, research on this topic is only now gaining momentum. The most popular methods are still borrowed from the more mature field of multispectral pansharpening and often overlook the uni

Peijin Li, Saminathan Ponnusamy

The main aim of this paper is to obtain the sharp upper and lower bounds for the growth and distortion of the analytic part $h$ of sense-preserving convex $K$-quasiconformal harmonic mappings.

Mohammad Azam, Tatiana Zajarniuk, Konrad Kwatek, Paolo Mele

We optimize the superconducting properties of Sm-based oxypnictide (Sm1111: SmFeAsO0.80F0.20) by using the Spark Plasma Sintering (SPS) technique under various synthesis conditions, including heating temperatures ranging from 600 to 1000 {\deg}C for durations of 5 to 30 minutes at the applied pressure of 45 MPa. All prepared bulks are characterized by struct

Afshin Akhshani, Małgorzata Białous, Leszek Sirko

We studied the statistical properties of a quantum system in the pseudo-integrable regime through the gap ratios between consecutive energy levels of the scattering spectra. A two-dimensional quantum billiard containing a point-like (zero-range) perturbation was experimentally simulated by a flat rectangular resonator with wire antennas. We show that the sys

Martin Tautenhahn, Ivan Veselic

We consider elliptic second order partial differential operators with Lipschitz continuous leading order coefficients on finite cubes and the whole Euclidean space. We prove quantitative sampling and equidistribution theorems for eigenfunctions. The estimates are scale-free, in the sense that for a sequence of growing cubes we obtain uniform estimates. These

Dóra Gréta Petróczy

In a weighted majority voting game, the players' weights are determined based on the constitutional planner's intentions. The weights are challenging to change in numerous cases, as they represent some desired disparity. However, the voting weights and the actual voting power do not necessarily coincide. Changing a decision threshold would offer some remedy.

Jaron Kent-Dobias

We study the typical behavior of random walkers on the microcanonical configuration space of mean-field disordered systems. Passive walks have an ergodicity-breaking transition at precisely the energy density associated with the dynamical glass transition, but persistent walks remain ergodic at lower energies. In models where the energy landscape is thorough

Feilong Tang, Chengzhi Liu, Zhongxing Xu, Ming Hu

Recent advancements in multimodal large language models (MLLMs) have significantly improved performance in visual question answering. However, they often suffer from hallucinations. In this work, hallucinations are categorized into two main types: initial hallucinations and snowball hallucinations. We argue that adequate contextual information can be extract

Alexander Shapiro, Yan Li

The aim of this paper is to investigate risk-averse and distributionally robust modeling of Stochastic Optimal Control (SOC) and Markov Decision Process (MDP). We discuss construction of conditional nested risk functionals, a particular attention is given to the Value-at-Risk measure. Necessary and sufficient conditions for existence of non-randomized optima

Michael Neri, Sara Baldoni

Due to the recent increase in the number of connected devices, the need to promptly detect security issues is emerging. Moreover, the high number of communication flows creates the necessity of processing huge amounts of data. Furthermore, the connected devices are heterogeneous in nature, having different computational capacities. For this reason, in this w

Zhichao Zhu, Yang Qi, Hengyuan Ma, Wenlian Lu

The Forward-Forward (FF) learning algorithm provides a bottom-up alternative to backpropagation (BP) for training neural networks, relying on a layer-wise "goodness" function with well-designed negative samples for contrastive learning. Existing goodness functions are typically defined as the sum of squared postsynaptic activations, neglecting correlated var

Kexin Shang, Chia-Hsuan Chang, Christopher C. Yang

Empowered by vast internal knowledge reservoir, the new generation of large language models (LLMs) demonstrate untapped potential to tackle medical tasks. However, there is insufficient effort made towards summoning up a synergic effect from multiple LLMs' expertise and background. In this study, we propose a multi-LLM collaboration framework tailored on a m

Sushant Gautam, Michael A. Riegler, Pål Halvorsen

We investigate fine-tuning Vision-Language Models (VLMs) for multi-task medical image understanding, focusing on detection, localization, and counting of findings in medical images. Our objective is to evaluate whether instruction-tuned VLMs can simultaneously improve these tasks, with the goal of enhancing diagnostic accuracy and efficiency. Using MedMultiP

Yujie Hou, Mei Wang, Yaoyao Zhong, Ting Zhang

Large Language Models (LLMs) have achieved remarkable performance across a wide range of mathematical benchmarks. However, concerns remain as to whether these successes reflect genuine reasoning or superficial pattern recognition. Existing evaluation methods, which typically focus either on the final answer or on the intermediate reasoning steps, reduce math

Vladimir I Korobov

The finite operators are derived for the nonrecoil (leading order in the $(m/M)$ expansion) relativistic corrections in hydrogen-like atoms and ions at orders $m\alpha^6$ in the two- and three-body formalism beyond the adiabatic approximation. Singular operators that appear in the derivation are analysed, and various regularization methods are discussed.

Stephen Y. Zhang, Michael P H Stumpf

We consider the Schr\"odinger bridge problem which, given ensemble measurements of the initial and final configurations of a stochastic dynamical system and some prior knowledge on the dynamics, aims to reconstruct the "most likely" evolution of the system compatible with the data. Most existing literature assume Brownian reference dynamics, and are implicit

Yiwei Sun, Peiqi Jiang, Chuanbin Liu, Luohao Lin

While the safety risks of image-based large language models (Image LLMs) have been extensively studied, their video-based counterparts (Video LLMs) remain critically under-examined. To systematically study this problem, we introduce VideoSafetyEval - a large-scale, real-world benchmark for Video LLM safety, which comprises 11.4k video-query pairs and spans 1

Arkadiusz Bochniak, Ludwik Dąbrowski, Andrzej Sitarz, Paweł Zalecki

We investigate spectral functionals associated with Dirac and Laplace-type differential operators on manifolds, defined via the Wodzicki residue, extending classical results for Dirac operators derived from the Levi-Civita connection to geometries with torsion. The local densities of these functionals recover fundamental geometric tensors, including the volu

Mathew Thomas Arun, Anuja Bandu Khadse

Baryon number violation in the visible sector induced by anti-baryonic dark matter provides a viable mechanism for low-scale baryogenesis. Two of the most sensitive probes of this scenario are neutron decay processes such as $n \to \bar{\nu} + \text{invisible}$ and $n \to \pi^0 + \text{invisible}$. In this work, we discuss the possible spontaneous breaking o

Xueyang Zhou, Guiyao Tie, Guowen Zhang, Hechang Wang

Vision-Language-Action (VLA) models have advanced robotic control by enabling end-to-end decision-making directly from multimodal inputs. However, their tightly coupled architectures expose novel security vulnerabilities. Unlike traditional adversarial perturbations, backdoor attacks represent a stealthier, persistent, and practically significant threat-part

Christoph Bartsch, Karol Kampf, David Podivín

We present a simple derivation of MHV amplitudes in massless spinor and scalar electrodynamics. Working with permutationally invariant amplitudes, we show that they are fully determined by their soft photon behavior and admit a simple factorized form in terms of soft factors and lower-point amplitudes. We prove these formulae using recursion relations. Final

Wenjie Yang, Mao Zheng, Mingyang Song, Zheng Li

Large language models (LLMs) have recently demonstrated remarkable capabilities in machine translation (MT). However, most advanced MT-specific LLMs heavily rely on external supervision signals during training, such as human-annotated reference data or trained reward models (RMs), which are often expensive to obtain and challenging to scale. To overcome this

Victor Dheur, Souhaib Ben Taieb

Reliably characterizing the full conditional distribution of a multivariate response variable given a set of covariates is crucial for trustworthy decision-making. However, misspecified or miscalibrated multivariate models may yield a poor approximation of the joint distribution of the response variables, leading to unreliable predictions and suboptimal deci

Zhaomin Wu, Ziyang Wang, Bingsheng He

Relational databases are often fragmented across organizations, creating data silos that hinder distributed data management and mining. Collaborative learning (CL) -- techniques that enable multiple parties to train models jointly without sharing raw data -- offers a principled approach to this challenge. However, existing CL frameworks (e.g., federated and

Melik Karapetyan, Ruben Manvelyan

We propose a new constructive approach to the solutions of the conservation condition for the three-point conformal correlation function in the Osborn-Petkou formulation generalized by the authors for higher spins. We propose for correlation functions of the same spin conformal currents the general hypothesis that the Osborn-Petkou structural tensor of highe

Muhammed Rizwan, Lars Carlsson, Mohammad Loni

The introduction of Large Language Models (LLMs) has significantly transformed Natural Language Processing (NLP) applications by enabling more advanced analysis of customer personas. At Volvo Construction Equipment (VCE), customer personas have traditionally been developed through qualitative methods, which are time-consuming and lack scalability. The main o