October 2025 arXiv papers — page 103

Yanling Pan, Qi Li, Gongping Zheng, Yongping Zhang

We propose the realization of a spin-2 Floquet spinor Bose-Einstein condensate via Floquet engineering of the quadratic Zeeman energy. In the Floquet system, the coupling strengths of all angular-momentum-conserving spin-flip processes are renormalized by driving-parameter-dependent Bessel functions. Such Floquet-engineered interactions significantly enriche

Michael Kapovich

We prove an estimate on intersection pairing of homology classes in hyperbolic 4-manifolds in terms of Thurston norms of these classes.

Roberto Massi De Oliveira, M^onica Cristina Garbin, Rodolfo Azevedo

Computational Thinking (CT) has emerged as a critical component in modern education, essential to equip students with the skills necessary to thrive in a technology-driven world. This survey provides a comprehensive analysis of the presence and integration of CT in school curricula across various countries. In addition, this study categorizes digital tools i

Elena Masciadri, Alessio Turchi, Camilo Weinberger, Marlene De Sepibus

Forecasting optical turbulence in the Earth's atmosphere has been an ambitious challenge for the astronomical scientific community for several decades. While earlier research primarily focused on whether it was possible to predict optical turbulence and its vertical distribution, current efforts are more concentrated on the accuracy achievable at different t

Jiajie Jin, Yuyao Zhang, Yimeng Xu, Hongjin Qian

Generating professional financial reports is a labor-intensive and intellectually demanding process that current AI systems struggle to fully automate. To address this challenge, we introduce FinSight (Financial InSight), a novel multi agent framework for producing high-quality, multimodal financial reports. The foundation of FinSight is the Code Agent with

Ilaria Maccari, Aline Ramires

Time-reversal symmetry-breaking (TRSB) superconductivity has been reported in a growing number of materials. In some cases, TRSB arises naturally from chiral superconductivity, but in many low-symmetry systems this explanation is not viable. In these latter cases, TRSB is often attributed to phase frustration among multiple superconducting gaps on different

Michael Bowler

Variations in the polar angle of the precessing jets in SS 433 are negatively correlated with variations in the speed of the jets. This was established some 20 years ago, based on the analysis of archival data accumulated from 40 years ago. This curious correlation has never been explained. Here I consider a promising mechanism involving the effect of flares

Wenxi Chen, Xinsheng Wang, Ruiqi Yan, Yushen Chen

Speech codecs that convert continuous speech signals into discrete tokens have become essential for speech language models. However, existing codecs struggle to balance high-quality reconstruction with semantically rich representations, limiting their effectiveness in both generative and understanding tasks. In this work, we propose SAC, a neural speech code

N. N. Chugai

I explore the origin of the circumstellar (CS) shell of the unusual SN Ia 2020aeuh based on the light curve model abd observational constraints. I estimate the $^{56}$Ni mass ($1.1 M\odot$), CS shell mass ($0.04-0.2 M\odot$), radius ($2\times10^{16}$ cm), and expansion velocity $\lesssim 200$ km/s. Large $^{56}$Ni mass and properties of the CS shell are cons

Jan Hladký, Petr Savický

The theory of graphons has proven to be a powerful tool in many areas of graph theory. In this paper, we introduce several foundational aspects of the theory of digraphons -- asymmetric two-variable functions that arise as limits of sequences of directed graphs (digraphs). Our results address their decomposition into strongly connected components, periodicit

Yu Guo, Jinfeng Liao, Shuzhe Shi

The chromo-magnetic monopoles (CMM), emergent topological excitations of non-Abelian gauge fields carrying chromo-magnetic charge, have long been postulated to play an important role in the vacuum confinement of quantum chromodynamics (QCD), the deconfinement transition at temperature $T_c\approx 160\rm MeV$, as well as the strongly coupled nature of quark-g

Haofan Ren, Qingsong Yan, Ming Lu, Rongfeng Lu

Recent advancements in 3D Gaussian Splatting (3DGS) have greatly influenced neural fields, as it enables high-fidelity rendering with impressive visual quality. However, 3DGS has difficulty accurately representing surfaces. In contrast, 2DGS transforms the 3D volume into a collection of 2D planar Gaussian disks. Despite advancements in geometric fidelity, re

Lin Shang, Shuai Geng, Xingli Li, Jiasen Jin

We investigate the steady-state phases of the one-dimensional quantum contact process model. We present the Liouvillian gap in the thermodynamic limit and uncover the metastability of the system. Exploiting the mean-field approximations with a novel self-consistent condition based on the effective field, we capture the avoid the interference of the metastabl

Hongpeng Bai, Minhong Dong, Yao Zhang, Shunzhe Zhao

The rapidly evolving Android malware ecosystem demands high-quality, real-time datasets as a foundation for effective detection and defense. With the widespread adoption of mobile devices across industrial systems, they have become a critical yet often overlooked attack surface in industrial cybersecurity. However, mainstream datasets widely used in academia

Xiangyu Mu, Dongliang Zhou, Jie Hou, Haijun Zhang

Mannequin-based clothing displays offer a cost-effective alternative to real-model showcases for online fashion presentation, but lack realism and expressive detail. To overcome this limitation, we introduce a new task called mannequin-to-human (M2H) video generation, which aims to synthesize identity-controllable, photorealistic human videos from footage of

Abdur Rahman, Mohammad Marufuzzaman, Jason Street, Haifeng Wang

Accurate and quick prediction of wood chip moisture content is critical for optimizing biofuel production and ensuring energy efficiency. The current widely used direct method (oven drying) is limited by its longer processing time and sample destructiveness. On the other hand, existing indirect methods, including near-infrared spectroscopy-based, electrical

Alain Goriely

One of the oldest and most enduring myths in human history is the belief that the Parthenon was cleverly designed with various curved structures and sizes in order to correct optical illusions, and therefore appear straight and regular. The myth has its origin in the writings of Vitruvius more than 2,000 years ago and was renewed in the nineteenth century wh

Hasan Akgul, Daniel Borg, Arta Berisha, Amina Rahimova

Large language models are often adapted through parameter efficient fine tuning, but current release practices provide weak assurances about what data were used and how updates were computed. We present Verifiable Fine Tuning, a protocol and system that produces succinct zero knowledge proofs that a released model was obtained from a public initialization un

Navreet Kaur, Hoda Ayad, Hayoung Jung, Shravika Mittal

Language model users often embed personal and social context in their questions. The asker's role -- implicit in how the question is framed -- creates specific needs for an appropriate response. However, most evaluations, while capturing the model's capability to respond, often ignore who is asking. This gap is especially critical in stigmatized domains such

Yoshiki Matsuoka

Recent report by CMS Collaboration on the excess of top and anti-top pair production is studied, under the hypothesis of the coexistence of a toponium $(\eta_t)$ and an additional elementary field $(\Psi)$. We examine the scenario where toponium and an additional field are mixed, and consider the plausible scenarios in that case. Two scenarios are examined:

Kangkang Deng, Rui Wang, Zhenyuan Zhu, Junyu Zhang

Large-scale constrained optimization is pivotal in modern scientific, engineering, and industrial computation, often involving complex systems with numerous variables and constraints. This paper provides a unified and comprehensive perspective on constructing augmented Lagrangian functions (based on Hestenes-Powell-Rockafellar augmented Lagrangian) for vario

Qinxiu Sun, Shuangjian Guo

In this paper, we introduce the notion of Leibniz-dendriform bialgebras and establish their equivalence with phase spaces and matched pairs of Leibniz algebras. The study of the coboundary case leads naturally to the Leibniz-dendriform Yang-Baxter equation (LD-YBE). We prove that skew-symmetric solutions of the LD-YBE give rise to coboundary Leibniz-dendrifo

Louis H. Rowen, Uzi Vishne

Suppose $F$ is an infinite field and let $f \in F\{X_1, \dots,X_m\}$ be a noncommutative polynomial. Partially answering a query of Makar-Limanov, we show that there are numbers $d$ and $m'$ such that, if $F$ is closed under taking $d$th roots, for any $n \ge m'$ there are matrices $A_1,\dots,A_m$ in~$M_n(F)$ such that $f(A_1,\dots,A_m)$ is upper triangular

Yingxu Wang, Kunyu Zhang, Jiaxin Huang, Nan Yin

Multimodal molecular representation learning, which jointly models molecular graphs and their textual descriptions, enhances predictive accuracy and interpretability by enabling more robust and reliable predictions of drug toxicity, bioactivity, and physicochemical properties through the integration of structural and semantic information. However, existing m

Yue Liu, Zhenchang Xing, Shidong Pan, Chakkrit Tantithamthavorn

In recent years, the AI wave has grown rapidly in software development. Even novice developers can now design and generate complex framework-constrained software systems based on their high-level requirements with the help of Large Language Models (LLMs). However, when LLMs gradually "take the wheel" of software development, developers may only check whether

Abdulwahab Felemban, Yahia Battach, Faizan Farooq Khan, Yuqian Fu

Coral reefs are rapidly declining under anthropogenic pressures (e.g., climate change), creating an urgent need for scalable and automated monitoring. Progress in data-driven coral analysis, however, is constrained by the scarcity of large-scale datasets with fine-grained labels that are taxonomically consistent across sites and studies. To address this gap,

Robert Plato, Bernd Hofmann

In this work, we consider a class of linear ill-posed problems with operators that map from the sequence space $ \ell_r $ ($r \ge 1$) into a Banach space and in addition satisfy a conditional stability estimate in the scale of sequence spaces $ \ell_q, \, q \ge 0 $. For the regularization of such problems in the presence of deterministic noise, we consider v

Thomas Dooms, Ward Gauderis

Sparse autoencoders are a standard tool for uncovering interpretable latent representations in neural networks. Yet, their interpretation depends on the inputs, making their isolated study incomplete. Polynomials offer a solution; they serve as algebraic primitives that can be analysed without reference to input and can describe structures ranging from linea

Shantanu Agarwal, Joel Barry, Steven Fincke, Scott Miller

Authorship attribution (AA) is the task of identifying the most likely author of a query document from a predefined set of candidate authors. We introduce a two-stage retrieve-and-rerank framework that finetunes LLMs for cross-genre AA. Unlike the field of information retrieval (IR), where retrieve-and-rerank is a de facto strategy, cross-genre AA systems mu

Mengwei Xu, Yu-Hong Dai, Xin-Wei Liu, Meiqi Ma

The value function formulation captures the hierarchical nature of bilevel optimization through the optimal value function of the lower level problem, yet its implicit and nonsmooth characteristics pose significant analytical and computational difficulties. We introduce a surrogate value function formulation that replaces the intractable value function with

Doyoon Kim, Junbin Song

We propose an enhanced physics-informed neural network (PINN), the Trace Regularity Physics-Informed Neural Network (TRPINN), which enforces the boundary loss in the Sobolev-Slobodeckij norm $H^{1/2}(\partial \Omega)$, the correct trace space associated with $H^1(\Omega)$. We reduce computational cost by computing only the theoretically essential portion of

Ming Zhong, Zhenya Yan

Neural operators offer a powerful data-driven framework for learning mappings between function spaces, in which the transformer-based neural operator architecture faces a fundamental scalability-accuracy trade-off: softmax attention provides excellent fidelity but incurs quadratic complexity $\mathcal{O}(N^2 d)$ in the number of mesh points $N$ and hidden di

Hans Hergen Lehmann, Jae Hee Lee, Steven Schockaert, Stefan Wermter

Large Language Models (LLMs) are increasingly used for knowledge-based reasoning tasks, yet understanding when they rely on genuine knowledge versus superficial heuristics remains challenging. We investigate this question through entity comparison tasks by asking models to compare entities along numerical attributes (e.g., ``Which river is longer, the Danube

Germain Pastén, Carla Silva Oliveira, João Domingos G. da Silva Junior, Claudia M. Justel

Let $G$ be a graph with adjacency matrix $A(G)$ and Laplacian matrix $L(G)$. In 2024, Samanta \textit{et} \textit{al.} defined the convex linear combination of $A(G)$ and $L(G)$ as $B_\alpha(G) = \alpha A(G) + (1-\alpha)L(G)$, for $\alpha \in [0,1]$. This paper presents some results on the eigenvalues of $B_{\alpha}(G)$ and their multiplicity when some sets

Mohammad Shahverdikondori, Jalal Etesami, Negar Kiyavash

We study regret minimization in causal bandits under causal sufficiency where the underlying causal structure is not known to the agent. Previous work has focused on identifying the reward's parents and then applying classic bandit methods to them, or jointly learning the parents while minimizing regret. We investigate whether such strategies are optimal. So

Amirkia Rafiei Oskooei, Kaan Baturalp Cosdan, Husamettin Isiktas, Mehmet S. Aktas

Large Language Models (LLMs) with vast context windows offer new avenues for in-context learning (ICL), where providing many examples ("many-shot" prompting) is often assumed to enhance performance. We investigate this assumption for the complex task of code translation. Through a large-scale empirical study of over 90,000 translations, we systematically eva

Dan Braha, Marcus A. M. de Aguiar

We generalize Condorcet's jury theorem (CJT) to socially connected populations in which agents revise discrete choices on a network in the presence of zealots. Free agents receive privately informative signals about the correct alternative and, at each update, either retain their state or imitate a uniformly chosen neighbor (free or zealot). For finite netwo

Zhoutong Wu, Yuan Zhang, Yiming Dong, Chenheng Zhang

Transformer models have driven breakthroughs across various language tasks by their strong capability to learn rich contextual representations. Scaling them to improve representation, however, often demands substantial memory and compute costs, such as the Key-Value (KV) cache used during auto-regressive decoding. Skip connections offer a promising way to im

Weilin Wan, Weizhong Zhang, Cheng Jin

Data selection improves computational efficiency by choosing informative subsets of training samples. However, existing methods ignore the compute budget, treating data selection and importance evaluation independently of compute budget constraints. Yet empirical studies show no algorithm can consistently outperform others (or even random selection) across v

Mariam Rakka, Marios Fournarakis, Olga Krestinskaya, Jinane Bazzi

The rapid scaling of language models (LMs) has resulted in unprecedented computational, memory, and energy requirements, making their training and deployment increasingly unsustainable. Quantization has emerged as an essential compression technique to reduce model size, alleviate memory bottlenecks, and accelerate inference. However, while uniform low-bit qu

Xiaokai Wei, Jiajun Wu, Daiyao Yi, Reza Shirkavand

Generative Recommendation (GR) models treat a user's interaction history as a sequence to be autoregressively predicted. When both items and actions (e.g., watch time, purchase, comment) are modeled, the layout-the ordering and visibility of item/action tokens-critically determines what information the model can use and how it generalizes. We present a unifi

Zishuai Zhang, Sihao Yu, Wenyi Xie, Ying Nie

The whole-page reranking plays a critical role in shaping the user experience of search engines, which integrates retrieval results from multiple modalities, such as documents, images, videos, and LLM outputs. Existing methods mainly rely on large-scale human-annotated data, which is costly to obtain and time-consuming. This is because whole-page annotation

Chao Li, Yuru Wang

Traditional knowledge graphs are constrained by fixed ontologies that organize concepts within rigid hierarchical structures. The root cause lies in treating domains as implicit context rather than as explicit, reasoning-level components. To overcome these limitations, we propose the Domain-Contextualized Concept Graph (CDC), a novel knowledge modeling frame

Jingtao Zhu, Yuying Zhao, Siqing Gan

In the study of McKean-Vlasov stochastic differential equations (MV-SDEs), numerical approximation plays a crucial role in understanding the behavior of interacting particle systems (IPS). Classical Milstein schemes provide strong convergence of order one under globally Lipschitz coefficients. Nevertheless, many MV-SDEs arising from applications possess supe

Zhenpeng Zhang, Yi Wang, Shanglei Chai, Yingying Liu

Lychee is a high-value subtropical fruit. The adoption of vision-based harvesting robots can significantly improve productivity while reduce reliance on labor. High-quality data are essential for developing such harvesting robots. However, there are currently no consistently and comprehensively annotated open-source lychee datasets featuring fruits in natura

C. T. Davies, M. Klein, A. Fumagalli, J. J. Mohr

Cosmic voids, vast underdensities in the large-scale structure, offer unique sensitivity to cosmological parameters. However, traditional 3D galaxy-based void finding is limited by many factors, including uncertainties in the galaxy-halo connection and distortions from redshift errors. Using alternative tracers and new 2D void definitions can alleviate these

Helene Charlotte Wiese Rytgaard, Mark van der Laan

This work develops a flexible inferential framework for nonparametric causal inference in time-to-event settings, based on stochastic interventions defined through multiplicative scaling of the intensity governing an intermediate event process. These interventions induce a family of estimands indexed by a scalar parameter {\alpha}, representing effects of mo

Amirkia Rafiei Oskooei, Mehmet S. Aktas

The proficiency of Large Language Models (LLMs) in processing structured data and adhering to syntactic rules is a capability that drives their widespread adoption but also makes them paradoxically vulnerable. In this paper, we investigate this vulnerability through BreakFun, a jailbreak methodology that weaponizes an LLM's adherence to structured schemas. B

Vera Pavlova, Mohammed Makhlouf

We introduce MOSAIC (Masked Objective with Selective Adaptation for In-domain Contrastive learning), a multi-stage framework for domain adaptation of text embedding models that incorporates joint domain-specific masked supervision. Our approach addresses the challenges of adapting large-scale general-domain text embedding models to specialized domains. By jo

Minghua Dou

Hartshorne developed a theory of generalized divisors on Gorenstein schemes to characterize codimension-one closed subschemes without embedded points. Generalized divisors can be viewed as a generalization of Weil divisors to non-normal schemes. The purpose of this paper is to extend generalized divisors on schemes to DMH stacks, where DMH stacks are Deligne

Jie Zhang, Meng Ding, Yang Liu, Jue Hong

We present a novel approach for attacking black-box large language models (LLMs) by exploiting their ability to express confidence in natural language. Existing black-box attacks require either access to continuous model outputs like logits or confidence scores (which are rarely available in practice), or rely on proxy signals from other models. Instead, we

Chuansen Peng, Xiaojing Shen

This paper tackles the challenging problem of jointly inferring time-varying network topologies and imputing missing data from partially observed graph signals. We propose a unified non-convex optimization framework to simultaneously recover a sequence of graph Laplacian matrices while reconstructing the unobserved signal entries. Unlike conventional decoupl

Sai Khadloya, Kush Juvekar, Arghya Bhattacharya, Utkarsh Saxena

Judicial work depends on close reading of long records, charge sheets, pleadings, annexures, orders, often spanning hundreds of pages. With limited staff support, exhaustive reading during hearings is impractical. We present CourtNav, a voice-guided, anchor-first navigator for legal PDFs that maps a judge's spoken command (e.g., "go to paragraph 23", "highli

Florian Besau, Christoph Thäle

Consider the triangle $T$ with vertices $(0,0)$, $(0,1)$, and $(1,0)$. The lower boundary of the convex hull of $(0,1)$, $(1,0)$, together with $n$ independent uniformly distributed random points in $T$, is called a random convex chain and denoted by $T_n$. We study the random variable $f_0(T_n)$, the number of vertices of this chain. Our first result gives

Zhi Gu, Wai Ho Mow

In wireless communications, the performance of non-orthogonal sequence sets significantly affects the level of multi-user interference when the number of users surpasses the sequence length. The design of non-orthogonal sequences plays a crucial role in both the non-orthogonality of the pilots in multi-cell systems and the signature sequences in overloaded c

Chengxuan Zhu, Shuchen Weng, Jiacong Fang, Peixuan Zhang

Photographic style, as a composition of certain photographic concepts, is the charm behind renowned photographers. But learning and transferring photographic style need a profound understanding of how the photo is edited from the unknown original appearance. Previous works either fail to learn meaningful photographic concepts from reference images, or cannot

Larkin Liu, Jalal Etesami

We explore the use of expert-guided bandit learning, which we refer to as online mixture-of-experts (OMoE). In this setting, given a context, a candidate committee of experts must determine how to aggregate their outputs to achieve optimal results in terms of aggregate accuracy. We propose two algorithms to address this problem. The first algorithm combines

Sara Hatami Rostami, Behrooz Nasihatkon

This paper presents a fully unsupervised approach for binary road segmentation (road vs. non-road), eliminating the reliance on costly manually labeled datasets. The method leverages scene geometry and temporal cues to distinguish road from non-road regions. Weak labels are first generated from geometric priors, marking pixels above the horizon as non-road a

Al Kari

The proliferation of Large Language Model (LLM) architectures presents a fundamental challenge: valuable, task-specific behaviors learned through fine-tuning methods like Low-Rank Adaptation (LoRA) are effectively trapped within their source model's architecture, herein referred to architectural lock-in. Existing transfer methods attempt to bridge this gap b

Yuya Arima

In this paper, we perform a multifractal analysis of Birkhoff averages for interval maps with finitely many branches and parabolic fixed points. Using the thermodynamic approach, we strengthen the results of Johansson et al. on the conditional variational principle for the multifractal spectra of Birkhoff averages. To do this, we develop several refined prop

Jonathan Nemirovsky, Maya Chuchem, Lee Peleg, Yakov Solomons

Quantum circuit synthesis and compilation are critical components in the quantum computing stack, both for contemporary quantum systems, where efficient use of limited resources is essential, as well as for large-scale fault-tolerant platforms, where computation time can be minimized. The specific characteristics of the quantum hardware determine which circu

Philani Rodney Majozi

Building on the recent work of Mushaandja and Olela-Otafudu~\cite{MushaandjaOlela2025} on modular metric topologies, this paper investigates extended structural properties of modular (pseudo)metric spaces. We provide necessary and sufficient conditions under which the modular topology $\tau(w)$ coincides with the uniform topology $\tau(\mathcal{V})$ induced

Pengfei Gao, Chao Peng

LLM-powered coding agents, which operate in iterative loops (turns) to solve software engineering tasks, are becoming increasingly powerful. However, their practical deployment is hindered by significant and unpredictable costs. This challenge arises from a combination of factors: quadratically growing token counts with each turn, the high price of models, t

Alex Acero, Daniel M. Jimenez-Gutierrez, Dario Pighin, Enrique Zuazua

Federated Learning (FL) enables collaborative decentralized training across multiple parties (nodes) while keeping raw data private. There are two main paradigms in FL: Horizontal FL (HFL), where all participant nodes share the same feature space but hold different samples, and Vertical FL (VFL), where participants hold complementary features for the same sa

Jiazhen Liu, Long Chen

Integrating diverse visual capabilities into a unified model is a significant trend in Multimodal Large Language Models (MLLMs). Among these, the inclusion of segmentation poses a distinct set of challenges. To equip MLLMs with pixel-level segmentation abilities, prevailing methods require finetuning the model to produce specific outputs compatible with a ma

Lord Sen, Shyamapada Mukherjee

This paper introduces an efficient quantum computing method for reducing special graphs in the context of the graph coloring problem. The special graphs considered include both symmetric and non-symmetric graphs where the axis passes through nodes only, edges only, and both together. The presented method reduces the number of coloring matrices, which is impo

Sheikh Jubair, Arwa Omayrah, Amal Alshammari, Alhanoof Althnian

Recent advancements in Large Language Models (LLMs) have demonstrated sophisticated capabilities, including the ability to process and comprehend extended contexts. These emergent capabilities necessitate rigorous evaluation methods to effectively assess their performance in long-context understanding. In this paper, we present \textbf{LC-Eval}, a bilingual,

Tongyang Li, Xinzhao Wang, Yexin Zhang

Computing Nash equilibria of zero-sum games in classical and quantum settings is extensively studied. For general-sum games, computing Nash equilibria is PPAD-hard and the computing of a more general concept called correlated equilibria has been widely explored in game theory. In this paper, we initiate the study of quantum algorithms for computing $\varepsi

Shihao Ji, Zihui Song

The remarkable zero-shot reasoning capabilities of large-scale Visual Language Models (VLMs) on static images have yet to be fully translated to the video domain. Conventional video understanding models often rely on extensive, task-specific training on annotated datasets, a process that is both costly and limited in scalability. This paper introduces a nove

Chang Wu, Zhiyuan Liu, Wen Shu, Liang Wang

Masked graph modeling (MGM) is a promising approach for molecular representation learning (MRL).However, extending the success of re-mask decoding from 2D to 3D MGM is non-trivial, primarily due to two conflicting challenges: avoiding 2D structure leakage to the decoder, while still providing sufficient 2D context for reconstructing re-masked atoms. To addre

Xiaoyu Guo, Minggu Wang, Jianjun Zhao

Large language models (LLMs) have demonstrated good performance in general code generation; however, their capabilities in quantum code generation remain insufficiently studied. This paper presents QuanBench, a benchmark for evaluating LLMs on quantum code generation. QuanBench includes 44 programming tasks that cover quantum algorithms, state preparation, g

Bin Jin, Bin Han, Wei Feng, Kuang Yu

Exact characterization of phase transitions requires sufficient configurational sampling, necessitating efficient and accurate potential energy surfaces. Molecular force fields with computational efficiency and physical interpretability are desirable but challenging to refine for complex interactions. To address this, we propose a force field refinement stra

Kush Juvekar, Arghya Bhattacharya, Sai Khadloya, Utkarsh Saxena

Large language models (LLMs) are entering legal workflows, yet we lack a jurisdiction-specific framework to assess their baseline competence therein. We use India's public legal examinations as a transparent proxy. Our multi-year benchmark assembles objective screens from top national and state exams and evaluates open and frontier LLMs under real-world exam

Junbo Li, Weimin Yuan, Yinuo Wang, Yue Zeng

Accurate 6D pose estimation of 3D objects is a fundamental task in computer vision, and current research typically predicts the 6D pose by establishing correspondences between 2D image features and 3D model features. However, these methods often face difficulties with textureless objects and varying illumination conditions. To overcome these limitations, we

Mingzheng Zhang, Jinfeng Gao, Dan Xu, Jiangrui Yu

X-ray image-based medical report generation (MRG) is a pivotal area in artificial intelligence that can significantly reduce diagnostic burdens for clinicians and patient wait times. Existing MRG models predominantly rely on Large Language Models (LLMs) to improve report generation, with limited exploration of pre-trained vision foundation models or advanced

Anil Kumar, Noritaka Shimizu, Takayuki Miyagi, Yusuke Tsunoda

The structure of low-lying states of $N=50$ nuclei is investigated by the advanced Monte Carlo shell model (MCSM) in the $\pi{(fp)}$-$\nu{(sdg)}$ model space. We have employed the shell-model Hamiltonian based on the valence-space in-medium similarity renormalization group, with minimal phenomenological adjustments to the single-particle energies. The MCSM r

Yuguang Yue, Irakli Salia, Samuel Hunt, Christopher Green

We argue that 3-D first-person video games are a challenging environment for real-time multi-modal reasoning. We first describe our dataset of human game-play, collected across a large variety of 3-D first-person games, which is both substantially larger and more diverse compared to prior publicly disclosed datasets, and contains text instructions. We demons

Gianluca Grassi

In the toric variety $\mathcal{T}$, with Cox ring graded by $°(z_{2i})=(1,-1,0)$, $°(z_{2i+1})=(1,0,-1)$ and $°(w_\pm)=(0,1,0),(0,0,1)$, we study hypersurfaces $\widetilde{X}^{2n}\subset\mathcal T$ of multidegree $(2d+1,-d,-d)$ over a field $k$. These are the strict transforms of odd-degree hypersurfaces in $\mathbb{P}^{2n+1}$ with multiplicity $d$ along two

Floris-Jan Willemsen, Niki van Stein, Ben van Werkhoven

Automatic performance tuning (auto-tuning) is essential for optimizing high-performance applications, where vast and irregular search spaces make manual exploration infeasible. While auto-tuners traditionally rely on classical approaches such as evolutionary, annealing, or surrogate-based optimizers, designing algorithms that efficiently find near-optimal co

Thuy Phuong Vu, Dinh-Cuong Hoang, Minhhuy Le, Phan Xuan Tan

Recent research has made significant progress in localizing and editing image regions based on text. However, most approaches treat these regions in isolation, relying solely on local cues without accounting for how each part contributes to the overall visual and semantic composition. This often results in inconsistent edits, unnatural transitions, or loss o

Xu He, Xiaolin Meng, Wenxuan Yin, Youdong Zhang

Developing universal Positioning, Navigation, and Timing (PNT) is our enduring goal. Today's complex environments demand PNT that is more resilient, energy-efficient and cognitively capable. This paper asks how we can endow unmanned systems with brain-inspired spatial cognition navigation while exploiting the high precision of machine PNT to advance universa

Jonathan R. Gair, Senwen Deng, Stanislav Babak

We present a probabilistic framework to quantify the impact of artefacts (glitches and gaps) in LISA data on transient gravitational wave signals. By modeling both artefacts and transient signals as independent Poisson processes, and characterising the contaminating effect of an artefact by an associated dead time, we estimate the probability distribution of

Shuo Han, Yukun Cao, Zezhong Ding, Zengyi Gao

Vision-language models (VLMs) have shown promise in graph structure understanding, but remain limited by input-token constraints, facing scalability bottlenecks and lacking effective mechanisms to coordinate textual and visual modalities. To address these challenges, we propose GraphVista, a unified framework that enhances both scalability and modality coord

Maciej Szymkat, Adam Korytowski

A method of optimal control computation is proposed for problems with control and state constraints. It uses a sequence of control structure adjustments in the form of generations and reductions of nodes and arcs, which do not change the current control but redefine the decision space. Several examples are given.

Jia Li, Guoxiang Zhao

Translating natural language instructions into executable motion plans is a fundamental challenge in robotics. Traditional approaches are typically constrained by their reliance on domain-specific expertise to customize planners, and often struggle with spatio-temporal couplings that usually lead to infeasible motions or discrepancies between task planning a

Shengyu Zhu, Congyi Fan, Fuxuan Zhang

Image restoration is a fundamental and challenging task in computer vision, where CNN-based frameworks demonstrate significant computational efficiency. However, previous CNN-based methods often face challenges in adequately restoring fine texture details, which are limited by the small receptive field of CNN structures and the lack of channel feature modeli

Francesco Vona, Julia Schorlemmer, Paulina Kaulard, Sebastian Fischer

This study compares user behavior between real and virtual supermarket shelves using eye tracking technology to assess behavior in both environments. A sample of 29 participants was randomly assigned to two conditions: a real world supermarket shelf with Tobii eye tracking and a virtual shelf using the Meta Quest Pro eye tracker. In both scenarios, participa

Victor Barroso-Nascimento, Maria Osório, Elaine Pimentel

Logical bilateralism challenges traditional concepts of logic by treating assertion and denial as independent yet opposed acts. While initially devised to justify classical logic, its constructive variants show that both acts admit intuitionistic interpretations. This paper presents a bilateral system where a formula cannot be both provable and refutable wit

Le Yang

Multimode fiber imaging requires strict matching between measurement value and measurement matrix to achieve image reconstruction. However, in practical applications, the measurement matrix often cannot be obtained due to unknown system configuration or difficulty in real-time alignment after arbitrary fiber bending, resulting in the failure of traditional r

Mahmut Elbistan, Peng-Ming Zhang, Peter Horvathy

The local Carroll symmetry of a gravitational wave found in Baldwin-Jeffery-Rosen coordinates is extended to a globally defined one by switching to Brinkmann coordinates. Two independent globally defined solutions of a Sturm-Liouville equation allow us to describe both the symmetries (translations and Carroll boosts) and the geodesic motions. One of them sat

Yikai Zhang, Ye Rong, Siyu Yuan, Jiangjie Chen

Existing language agents often encounter difficulties in dynamic adversarial games due to poor strategic reasoning. To mitigate this limitation, a promising approach is to allow agents to learn from game interactions automatically, without relying on costly expert-labeled data. Unlike static environments where agents receive fixed feedback or rewards, select

Yiliang Fan, Rongxiang Zhu, Tongshuai Zhu, Jianzhou Zhao

The MnBi$_2$Te$_4$ material family has emerged as a key platform for exploring magnetic topological phases, most notably exemplified by the experimental realization of the axion insulator state. While spin dynamics are known to significantly influence the axion state, a profound understanding of their interplay remains elusive. In this work, we employ an ant

Peter Jaksch

A series of numerical experiments are performed, where a symmetric potential is generated for the 1D time-independent Schr\"odinger equation, with an eigenspectrum that matches the imaginary part of the first nontrivial zeros of the Riemann Zeta Function. The potential is generated as a series of correction functions, where the starting point is a potential

Nicoletta Sanvitale, Claudio Gheller, Franco Vazza, Federica Govoni

We present TUNA, a Vision-Transformer based network adapted from segmentation to flux regression for faint, diffuse radio emission. Trained on LOFAR-like mock observations derived from cosmological simulations, TUNA accurately reconstructs low surface-brightness structures, with only mild smoothing and small brightness-dependent biases. Applied to LOFAR data

Young In Kim, Andrea Agiollo, Rajiv Khanna

Modern machine learning solutions require extensive data collection where labeling remains costly. To reduce this burden, open set active learning approaches aim to select informative samples from a large pool of unlabeled data that includes irrelevant or unknown classes. In this context, we propose Sharpness Aware Minimization for Open Set Active Learning (

Siyin Wang, Wenyi Yu, Xianzhao Chen, Xiaohai Tian

Human interaction is inherently multimodal and full-duplex: we listen while watching, speak while acting, and fluidly adapt to turn-taking and interruptions. Realizing these capabilities is essential for building models simulating humans. We present ELLSA (End-to-end Listen, Look, Speak and Act), which, to our knowledge, is the first full-duplex, end-to-end

Kyung-Hwan Kim, DongHyun Ahn, Dong-hyun Lee, JuYoung Yoon

State estimation is crucial for legged robots as it directly affects control performance and locomotion stability. In this paper, we propose an Adaptive Invariant Extended Kalman Filter to improve proprioceptive state estimation for legged robots. The proposed method adaptively adjusts the noise level of the contact foot model based on online covariance esti

Shihao Ji, Zihui Song

The Mixture of Experts (MoE) architecture enables the scaling of Large Language Models (LLMs) to trillions of parameters by activating a sparse subset of weights for each input, maintaining constant computational cost during inference. Concurrently, Low-Rank Adaptation (LoRA) has emerged as a dominant technique for parameter-efficiently fine-tuning LLMs on s

Soroush Khademi, Jesse J. Slim, Kiarn T. Laverick, Jin Chang

Weak quantum measurements enable real-time tracking and control of dynamical quantum systems, producing quantum trajectories -- evolutions of the quantum state of the system conditioned on measurement outcomes. For classical systems, the accuracy of trajectories can be improved by incorporating future information, a procedure known as smoothing. Here we appl

Wei Huang, Peining Li, Meiyu Liang, Xu Hou

Multimodal Knowledge Graphs (MKGs) extend traditional knowledge graphs by incorporating visual and textual modalities, enabling richer and more expressive entity representations. However, existing MKGs often suffer from incompleteness, which hinder their effectiveness in downstream tasks. Therefore, multimodal knowledge graph completion (MKGC) task is receiv