November 2025 arXiv papers — page 124

V. Asiryan, V. Volchkov, N. Papulovskaya

This paper is devoted to the development and research of a new compression technology based on Weyl-Heisenberg bases (WH-technology) for modifying the JPEG compression standard and improving its characteristics. For this purpose, the paper analyzes the main stages of the JPEG compression algorithm, notes its key features and problems that limit further enhan

Yong Sun, Zhengjie Zhang, Junyu Shi, Zhiyuan Zhang

Identification of fine-grained embryo developmental stages during In Vitro Fertilization (IVF) is crucial for assessing embryo viability. Although recent deep learning methods have achieved promising accuracy, existing discriminative models fail to utilize the distributional prior of embryonic development to improve accuracy. Moreover, their reliance on sing

Jun-Hyoung Park, Ho-Jun Song, Seong-Whan Lee

Deep learning-based molecular generation models have shown great potential in efficiently exploring vast chemical spaces by generating potential drug candidates with desired properties. However, these models often produce chemically invalid molecules, which limits the usable scope of the learned chemical space and poses significant challenges for practical a

Adel Bechihi, Aristotelis Kapnopoulos

Estimating the Region of Attraction (RoA) for nonlinear dynamical systems is a fundamental problem in control theory, with direct implications for stability analysis and safe controller design. Traditional approaches rely on analytically derived Lyapunov functions, which are often conservative and challenging to construct for high-dimensional or highly nonli

Jirong Zha, Yuxuan Fan, Tianyu Zhang, Geng Chen

Multimodal Large Language Models (MLLMs) have shown promise in single-agent vision tasks, yet benchmarks for evaluating multi-agent collaborative perception remain scarce. This gap is critical, as multi-drone systems provide enhanced coverage, robustness, and collaboration compared to single-sensor setups. Existing multi-image benchmarks mainly target basic

Ali Ellouze, Bastien Fernandez

Given the combined evidences of bounded rationality, limited information and short-term optimization, over-the-counter (OTC) fresh product markets provide a perfect instance where to develop a behavioural approach to the analysis of micro-economic systems. Aiming at characterizing via a rigorous mathematical analysis, the main features of the spontaneous org

Yaoxin Ge, Yao Zhang, Dengji Zhao

Incentives for early arrival (I4EA) was recently proposed for studying online cooperative games. In an online cooperative game, players arrive in an unknown order, and the value increase after each player arrived should be distributed immediately among all the arrived players. Although there is only one arriving order in the game, we also hope that the value

Voraprapa Nakavachara

This paper investigates the relationship between AI use and worker well-being outcomes such as mental health, job enjoyment, and physical health and safety, using microdata from the OECD AI Surveys across seven countries. The results reveal that AI users are significantly more likely to report improvements across all three outcomes, with effects ranging from

Farhad Abtahi, Fernando Seoane, Iván Pau, Mario Vega-Barbas

Healthcare AI systems face major vulnerabilities to data poisoning that current defenses and regulations cannot adequately address. We analyzed eight attack scenarios in four categories: architectural attacks on convolutional neural networks, large language models, and reinforcement learning agents; infrastructure attacks exploiting federated learning and me

Zichao Wei, Jun Zeng, Ming Wen, Zeliang Yu

Software vulnerabilities are increasing at an alarming rate. However, manual patching is both time-consuming and resource-intensive, while existing automated vulnerability repair (AVR) techniques remain limited in effectiveness. Recent advances in large language models (LLMs) have opened a new paradigm for AVR, demonstrating remarkable progress. To examine t

Xiaokun Luan, Zeming Wei, Yihao Zhang, Meng Sun

Large language models (LLMs) are increasingly tasked with generating structured outputs. While structured generation methods ensure validity, they often lack output diversity, a critical limitation that we confirm in our preliminary study. We propose a novel method to enhance diversity in automaton-based structured generation. Our approach utilizes automata

Dimitar Peshevski, Riste Stojanov, Dimitar Trajanov

The rapid expansion of e-commerce platforms generates vast amounts of unstructured product data, creating significant challenges for information retrieval, recommendation systems, and data analytics. Knowledge Graphs (KGs) offer a structured, interpretable format to organize such data, yet constructing product-specific KGs remains a complex and manual proces

Sumin Yu, Taesup Moon

While diffusion-based T2I models have achieved remarkable image generation quality, they also enable easy creation of harmful content, raising social concerns and highlighting the need for safer generation. Existing inference-time guiding methods lack both adaptivity--adjusting guidance strength based on the prompt--and selectivity--targeting only unsafe reg

S Deion, F Adersh, M Sahoo

We theoretically explore the dynamics of a chiral active Ornstein Uhlenbeck particle confined in a two-dimensional anisotropic harmonic trap. The particle is driven by chirality and is coupled to two orthogonal heat baths, potentially at two different temperatures. Using both analytical approach and numerical simulation, we explore the rotational dynamics of

Zhiwei Zhang, Hui Zhang, Kaihong Huang, Chenghao Shi

World models enable robots to conduct counterfactual reasoning in physical environments by predicting future world states. While conventional approaches often prioritize pixel-level reconstruction of future scenes, such detailed rendering is computationally intensive and unnecessary for planning tasks like navigation. We therefore propose that prediction and

Fuxiang Huang, Xiaowei Fu, Shiyu Ye, Lina Ma

Unsupervised domain adaptation (UDA) aims to transfer knowledge from a label-rich source domain to an unlabeled target domain by addressing domain shifts. Most UDA approaches emphasize transfer ability, but often overlook robustness against adversarial attacks. Although vanilla adversarial training (VAT) improves the robustness of deep neural networks, it ha

Marcello Benedetti, Gabriel Marin-Sanchez, Jordi Weggemans, Matthias Rosenkranz

Testing the predictions of quantum mechanics has been one of the main experimental endeavors for decades. Recent advancements in technology led to a number of demonstrations which test non-classicality via specific computational tasks. Limitations of these experiments include dependence on complexity theory assumptions, susceptibility to hardware noise and i

Xinlei Yu, Chengming Xu, Guibin Zhang, Zhangquan Chen

Despite the remarkable success of Vision-Language Models (VLMs), their performance on a range of complex visual tasks is often hindered by a "visual processing bottleneck": a propensity to lose grounding in visual evidence and exhibit a deficit in contextualized visual experience during prolonged generation. Drawing inspiration from human cognitive memory th

HongYu Liu, Ruijie Wan, Yueju Han, Junxin Li

Audio classification plays an essential role in sentiment analysis and emotion recognition, especially for analyzing customer attitudes in marketing phone calls. Efficiently categorizing customer purchasing propensity from large volumes of audio data remains challenging. In this work, we propose a novel Multi-Segment Multi-Task Fusion Network (MSMT-FN) that

Sungheon Jeong, Ryozo Masukawa, Jihong Park, Sanggeon Yun

While recent Large Vision-Language Models (LVLMs) exhibit strong multimodal reasoning abilities, they often produce ungrounded or hallucinated responses because they rely too heavily on linguistic priors instead of visual evidence. This limitation highlights the absence of a quantitative measure of how much these models actually use visual information during

Yiran Song, Yikai Zhang, Shuang Zhou, Guojun Xiong

Multiple instance learning (MIL) has emerged as the dominant paradigm for whole slide image (WSI) analysis in computational pathology, achieving strong diagnostic performance through patch-level feature aggregation. However, existing MIL methods face critical limitations: (1) they rely on attention mechanisms that lack causal interpretability, and (2) they f

Jeonghwan Lee, Cong Ma

Distribution shift between the training domain and the test domain poses a key challenge for modern machine learning. An extensively studied instance is the \emph{covariate shift}, where the marginal distribution of covariates differs across domains, while the conditional distribution of outcome remains the same. The doubly-robust (DR) estimator, recently in

Zongyang Qiu, Bingyuan Wang, Xingbei Chen, Yingqing He

Emotion plays a pivotal role in video-based expression, but existing video generation systems predominantly focus on low-level visual metrics while neglecting affective dimensions. Although emotion analysis has made progress in the visual domain, the video community lacks dedicated resources to bridge emotion understanding with generative tasks, particularly

Anindya Mukherjee, Pabitra Barik

Let X be a smooth projective variety carrying an Ulrich bundle. In the first part of this note, we construct an Ulrich sheaf on n-th symmetric power of X, which is a singular variety when $DimX >1$. As a consequence, we get the existence of an Ulrich bundle on Hillb^{n}C, where C is a smooth projective curve. Let A be an abelian variety which carries an Ulri

Jiaxi Li, Yue Zhu, Eun Kyung Lee, Klara Nahrstedt

Different from traditional Large Language Model (LLM) serving that colocates the prefill and decode stages on the same GPU, disaggregated serving dedicates distinct GPUs to prefill and decode workload. Once the prefill GPU completes its task, the KV cache must be transferred to the decode GPU. While existing works have proposed various KV cache transfer path

HongYu Liu, Junxin Li, Changxi Guo, Hao Chen

Recognizing speaker intent in long audio dialogues among speakers has a wide range of applications, but is a non-trivial AI task due to complex inter-dependencies in speaker utterances and scarce annotated data. To address these challenges, an end-to-end framework, namely DialogGraph-LLM, is proposed in the current work. DialogGraph-LLM combines a novel Mult

Sylvester Willys Namagwa

As life expectancy in Kenya increases, so does the need for efficient pension schemes that can secure a dignified retirement and protect members from old age poverty. Limited research, however, has explored the efficiency of these schemes under existing governance structures. This study addresses that gap by examining the combined effects of corporate govern

M. Jabłońska, T. Różański, L. Casagrande, H. Shah

In the era of large time-domain spectro-photometric surveys, surface variations such as starspots, chemical inhomogeneities, pulsations, rotational distortions, and binary interactions can now be directly detected and modelled. Accurately interpreting these phenomena requires stellar spectral synthesis frameworks that go beyond the assumption of homogeneous

Jiajun Chen, Sai Cheng, Yutao Yuan, Yirui Zhang

Multimodal models integrating natural language and visual information have substantially improved generalization of representation models. However, their effectiveness significantly declines in real-world situations where certain modalities are missing or unavailable. This degradation primarily stems from inconsistent representation learning between complete

Jianfei Cao, Michael P. Leung

This paper studies double/debiased machine learning (DML) methods applied to weakly dependent data. We allow observations to be situated in a general metric space that accommodates spatial and network data. Existing work implements cross-fitting by excluding from the training fold observations sufficiently close to the evaluation fold. We find in simulations

Xiuding Cai, Xueyao Wang, Sen Wang, Yaoyao Zhu

Intraoperative monitoring and prediction of vital signs are critical for ensuring patient safety and improving surgical outcomes. Despite recent advances in deep learning models for medical time-series forecasting, several challenges persist, including the lack of standardized benchmarks, incomplete data, and limited cross-center validation. To address these

Indranil Biswas

For a $\Gamma$--equivariant holomorphic Lie algebroid $(V,\, \phi)$, on a compact Riemann surface $X$ equipped with an action of a finite group $\Gamma$, we investigate the equivariant holomorphic Lie algebroid connections on holomorphic principal $G$--bundles over $X$, where $G$ is a connected affine complex reductive group. If $(V,\,\phi)$ is nonsplit, the

Keunwoo Park, Jihye Chae, Joong Ho Ahn, Jihoon Kweon

Text-to-image synthesis models require the ability to generate diverse images while maintaining stability. To overcome this challenge, a number of methods have been proposed, including the collection of prompt-image datasets and the integration of additional data modalities during training. Although these methods have shown promising results in general domai

Daxin Li, Yuanchao Bai, Kai Wang, Wenbo Zhao

Autoregressive (AR) models, the theoretical performance benchmark for learned lossless image compression, are often dismissed as impractical due to prohibitive computational cost. This work re-thinks this paradigm, introducing a framework built on hierarchical parallelism and progressive adaptation that re-establishes pure autoregression as a top-performing

Prab Prasertying, Paulo Garcia, Warisa Sritriratanarak

Coordination of movement and configuration in robotic swarms is a challenging endeavor. Deciding when and where each individual robot must move is a computationally complex problem. The challenge is further exacerbated by difficulties inherent to physical systems, such as measurement error and control dynamics. Thus, how to best determine the optimal path fo

Bin Wang, Xi-Yu Luo, Bo-Feng Gao, Jian-Long Liu

Quantum networks and remote quantum entanglement serve as vital future quantum communication resources with broad applicability. A key direction lies in extending the baseline of optical interferometers to enhance angular resolution in interferometric imaging. Here, by measuring a simulated thermal light field, we report the demonstration of a memory-assiste

Wenbin Bai, Qiyu Chen, Xiangbo Lin, Jianwen Li

The inherent difficulty and limited scalability of collecting manipulation data using multi-fingered robot hand hardware platforms have resulted in severe data scarcity, impeding research on data-driven dexterous manipulation policy learning. To address this challenge, we present a hand-agnostic manipulation transfer system. It efficiently converts human han

Tarik Akan

In this work, the spin-averaged $B_c$ spectrum is computed in a naive Cornell framework, treating the meson as a nonrelativistic system in a spin-independent potential. The Cornell parameters are calibrated directly to the spin-averaged $B_c$ tower by anchoring the $1S$ centroid and scanning a grid in $(\sigma,\kappa)$, with the additive constant $V_0$ fixed

Aladin Djuhera, Farhan Ahmed, Swanand Ravindra Kadhe, Syed Zawad

Aligning large language models (LLMs) is a central objective of post-training, often achieved through reward modeling and reinforcement learning methods. Among these, direct preference optimization (DPO) has emerged as a widely adopted technique that fine-tunes LLMs on preferred completions over less favorable ones. While most frontier LLMs do not disclose t

Le Xu, Jiayu Chen

Offline reinforcement learning (RL) is vulnerable to real-world data corruption, with even robust algorithms failing under challenging observation and mixture corruptions. We posit this failure stems from data corruption creating sharp minima in the loss landscape, leading to poor generalization. To address this, we are the first to apply Sharpness-Aware Min

Xiying Zhao, Zhoufutu Wen, Zhixuan Chen, Jingzhe Ding

The evaluation of discourse-level translation in expert domains remains inadequate, despite its centrality to knowledge dissemination and cross-lingual scholarly communication. While these translations demand discourse-level coherence and strict terminological precision, current evaluation methods predominantly focus on segment-level accuracy and fluency. To

Niklas Erdmann, Lars Bentsen, Roy Stenbro, Heine Nygard Riise

Probabilistic forecasting is not only a way to add more information to a prediction of the future, but it also builds on weaknesses in point prediction. Sudden changes in a time series can still be captured by a cumulative distribution function (CDF), while a point prediction is likely to miss it entirely. The modeling of CDFs within forecasts has historical

Rongbin Hu, Jeffrey Liu

We propose a training-free, binary verification workflow for zero-shot vision with off-the-shelf VLMs. It comprises two steps: (i) quantization, which turns the open-ended query into a multiple-choice question (MCQ) with a small, explicit list of unambiguous candidates; and (ii) binarization, which asks one True/False question per candidate and resolves dete

Haoran Zhang, Haotao Zhu, Ruihua He, Yan Zhang

Quantum Key Distribution (QKD) supports the negotiation and sharing of private keys with unconditional security between authorized parties. Over the years, theoretical advances and experimental demonstrations have successfully transitioned QKD from laboratory research to commercial applications. As QKD expands its reach globally, it encounters challenges suc

Satyajit Puhan, Shubham Sharma, Narinder Kumar, Harleen Dahiya

We investigate the partonic structure of the $\rho$ meson, the lightest spin-$1$ vector meson, within the light-front quark model (LFQM). To explore the sensitivity to model assumptions, we employ two distinct types of spin wave functions in the LFQM. Using light-front helicity wave functions, we derive explicit expressions for the leading-twist and subleadi

Bowen Sun, Yujun Cai, Ming-Hsuan Yang, Hang Wu

Video LLMs suffer from temporal inconsistency: small shifts in frame timing can flip attention and suppress relevant frames. We trace this instability to the common extension of Rotary Position Embeddings to video through multimodal RoPE. The induced inverse Fourier time kernel exhibits frame-scale ripples that multiply adjacent frames by different factors,

Liangwei Dong, Mingjing Fan, Boris A. Malomed, Yaroslav V. Kartashov

It is known that, under appropriate conditions, mean-field interactions can be canceled in binary BEC, leading to the formation of the Lee-Huang-Yang (LHY) superfluid, in which the nonlinearity is solely represented by the quartic LHY term. In this work we systematically investigate the existence, stability and evolution of hopfion states in this species of

Yukiho Matsumoto, Keisuke Yoshida, Tomohiko G. Sano

Fitting two different materials is a versatile methodology in manufacturing complex structures. One of the canonical models for fitting is the snap-fit model, in which flexible materials and rigid structures are assembled by pushing their interlocking components together. The assembly via snap-fit is often accompanied by large deformations of flexible struct

Haoran Chen, Houze Xu, Micah Goldblum, Daoguo Dong

Class-incremental learning (CIL) enables models to continuously learn new categories from sequential tasks without forgetting previously acquired knowledge. While recent advances in vision-language models such as CLIP have demonstrated strong generalization across domains, extending them to continual settings remains challenging. In particular, learning task

Hikaru Yamamoto

For an immersed Lagrangian submanifold $L$ in a Kähler manifold $(M,ω)$, there exists a symplectic local diffeomorphism from a tubular neighborhood of the image of the zero section in the normal bundle $T^{\bot}L$ of $L$, equipped with a canonical symplectic form $\tildeω$, to $(M,ω)$ whose restriction to $L$ is the identity map by Weinstein's Lagrangian

Bo Fu, Kai-Zhi Bai, Shi-Hao Bi, Shun-Qing Shen

Electric conductivities may reveal the topological and magnetic properties of band structures in solids, especially for two-dimensional unpaired Dirac fermions. In this work, we evaluate the longitudinal and Hall conductivity for unpaired Dirac fermions in the framework of the self-consistent Born approximation and find a nearly semi-elliptic relation betwee

Anzhe Cheng, Shukai Duan, Shixuan Li, Chenzhong Yin

Mixture-of-Experts (MoE) architectures expand model capacity by sparsely activating experts but face two core challenges: misalignment between router logits and each expert's internal structure leads to unstable routing and expert underutilization, and load imbalances create straggler bottlenecks. Standard solutions, such as auxiliary load-balancing losses,

Qingyan Ren, Liming Tang

This is the second paper in our series of papers dedicated to the study of the generalized loop Heisenberg-Virasoro algebra. The first paper is dedicated to the study of maps on the generalized loop Heisenberg-Virasoro algebra, including derivations, $2$-local derivations, biderivations the automorphism groups. The present paper is dedicated to the study of

Ye Zhang, Zichang Huang

By using the heat kernel method, we construct diffeomorphism-covariant coherent states for the $SU(3)$ gauge group. We numerically demonstrate that these states exhibit the required semiclassical properties in the semiclassical limit: the peakedness property of the probability distribution and the peakedness property of the overlap function. We also provide

Chong Yang, Nobuhiro Okabe, Yasushi Fukazawa

We report the discovery of a faint X-ray bridge connecting between the comma-shaped gas and the main cluster in MCXC J0157.4-0550, using {\it XMM-Newton} image. The filamentary structure is found in a model-independent manner in both topological features and Gaussian Gradient Magnitude filtering. The X-ray surface brightness profile perpendicular to the fila

Jingyi Zhang, James C. Spall

The Metropolis-Hastings algorithm has been extensively studied in the estimation and simulation literature, with most prior work focusing on convergence behavior and asymptotic theory. However, its covariance structure-an important statistical property for both theory and implementation-remains less understood. In this work, we provide new theoretical insigh

Yoshihito Tanaka

This paper investigates neighborhood and algebraic models for predicate modal logics with $\omega$-rules, including non-normal cases. We establish sufficient conditions under which such logics have neighborhood models with constant domains and satisfy the completeness theorem with respect to neighborhood frames with constant domains. Related results for norm

Khushboo Suman

Rheology, the study of flow, plays a vital role in diverse industries such as pharmaceuticals, cosmetics and food. In this work, we provide a comprehensive introduction to fundamental rheological experiments and offer a strategic approach to understand the rheological behavior of any soft condensed material. We emphasize the importance of design of a good rh

Andrea Maurino

Machine Learning (ML) models are being increasingly employed for credit risk evaluation, with their effectiveness largely hinging on the quality of the input data. In this paper we investigate the impact of several data quality issues, including missing values, noisy attributes, outliers, and label errors, on the predictive accuracy of the machine learning m

Yi-Jian Du, Chih-Hao Fu, Yihong Wang, Chongsi Xie

We study the algebraic structure of one-loop BCJ numerators in Yang-Mills and related theories. Starting from the propagator matrix that connects colour-ordered integrands to numerators, we identify the consistency conditions that ensure the existence of Jacobi-satisfying numerator solutions and determine the unique construction. The relation between one-loo

Xintian Han, Honggang Chen, Quan Lin, Jingyue Gao

Traditional ID-based recommender systems often struggle with cold-start and generalization challenges. Multimodal recommendation systems, which leverage textual and visual data, offer a promising solution to mitigate these issues. However, existing industrial approaches typically adopt a two-stage training paradigm: first pretraining a multimodal model, then

Riju Marwah, Vishal Pallagani, Ritvik Garimella, Amit Sheth

LLMs are increasingly being deployed as chatbots, but today's interfaces offer little to no friction: users interact through seamless conversations that conceal when the model is drifting, hallucinating or failing. This lack of transparency fosters blind trust, even as models produce unstable or repetitive outputs. We introduce an interactive demo that surfa

Han Zhang, Qingyan Meng, Jiaqi Wang, Baiyu Chen

Weight quantization in spiking neural networks (SNNs) could further reduce energy consumption. However, quantizing weights without sacrificing accuracy remains challenging. In this study, inspired by astrocyte-mediated synaptic modulation in the biological nervous systems, we propose Temporal-adaptive Weight Quantization (TaWQ), which incorporates weight qua

Fan Wang, Sandy Irani

We study the problem of constructing cycle bases of graphs with low maximum edge participation, defined as the maximum number of basis cycles that share any single edge. This quantity, though less studied than total weight or length, plays a critical role in quantum fault tolerance because it directly impacts the overhead of lattice surgery procedures used t

M. Ryzhii, V. Ryzhii, C. Tang, T. Otsuji

We examine the robustness of the S-shaped current-voltage characteristics associated with hot-carrier-induced electrical breakdown in perforated graphene metasurfaces (PGMs) as a function of doping. The perforation of the graphene layer forms interdigital arrays of graphene microribbons (GMRs) interconnected by graphene nanoribbon (GNR) bridges. These GNR co

Rhea Palak Bakshi, Anthony Christiana, Huizheng Guo, Dionne Ibarra

Over the past thirty-seven years, the study of linear and quadratic skein modules has produced a rich and far-reaching skein theory, intricately connected to diverse areas of mathematics and physics, including algebraic geometry, hyperbolic geometry, topological quantum field theories, and statistical mechanics. However, despite these advances, skein modules

P. R. Gordoa, A. Pickering, D. Puertas-Centeno, E. V. Toranzo

We present a new generalization of the well-known power-type Sundman transformation, involving not only powers of the function but also of its derivative, along with its inverse. Our aim is to explore the use of such transformations in the derivation of solutions of ordinary differential equations and in the study of their properties. We then show their usef

Zhe Yang, Wenrui Li, Hongtao Chen, Penghong Wang

Multimodal learning aims to improve performance by leveraging data from multiple sources. During joint multimodal training, due to modality bias, the advantaged modality often dominates backpropagation, leading to imbalanced optimization. Existing methods still face two problems: First, the long-term dominance of the dominant modality weakens representation-

Allana Tavares Bastos, Tiago Alves Schieber, Renato Hadad, Laura Carpi

Fraud in public procurement remains a persistent challenge, especially in large, decentralized systems like Brazil's Unified Health System. We introduce Heron's Information Coefficient (HIC), a geometric measure that quantifies how subgraphs deviate from the global structure of a network. Applied to over eight years of Brazilian bidding data for medical supp

Nikitha Donekal Chandrashekar, Sehrish Basir Nizamani, Margaret Ellis, Naren Ramakrishnan

We transitioned our post-CS1 course that introduces various subfields of computer science so that it integrates Large Language Models (LLMs) in a structured, critical, and practical manner. It aims to help students develop the skills needed to engage meaningfully and responsibly with AI. The course now includes explicit instruction on how LLMs work, exposure

Shu-Tong Zhang, Wen Luo, Dan-Yang Pang, Zhi-Cai Li

We demonstrate to extract $^{88}\mathrm{Y}(n,\gamma)$ cross sections using the $(p,p'\gamma)$ surrogate reaction with proper treatment of the spin-parity distribution of the compound nucleus $^{89}\mathrm{Y}$. Experimental data of both $\gamma$-decay probability and $\gamma$-ray strength function are used to constrain the nuclear model parameters within a co

Daemo Kang, Tien-Tien Yeh, Takahiro Morimoto, Alexander V. Balatsky

We present a novel framework for controlling Higgs mode and vortex dynamics in superconductors using structured light. We propose a phenomenon analog of the Kapitza-Dirac effect in superconductors, where Higgs waves scatter off light-induced vortex lattices, generating interference patterns akin to matter wave diffraction. We also find that the vortices enab

Wenrui Li, Wei Han, Hengyu Man, Wangmeng Zuo

With the rapid growth of video content on social media, video summarization has become a crucial task in multimedia processing. However, existing methods face challenges in capturing global dependencies in video content and accommodating multimodal user customization. Moreover, temporal proximity between video frames does not always correspond to semantic pr

Steven J. Jones, Robert E. Wray, John E. Laird

Deployed, autonomous AI systems must often evaluate multiple plausible courses of action (extended sequences of behavior) in novel or under-specified contexts. Despite extensive training, these systems will inevitably encounter scenarios where no available course of action fully satisfies all operational constraints (e.g., operating procedures, rules, laws,

Weilin Liu, Xianlei Huang, Li-Guo Dou, Qianglong Fang

Controllable gas adsorption is critical for both scientific and industrial fields, and high-capacity adsorption of gases on solid surfaces provides a significant promise due to its high-safety and low-energy consumption. However, the adsorption of nonpolar gases, particularly noble gases, poses a considerable challenge under atmospheric pressure and room tem

Ayumi Mutoh, Junoh Heo

Gaussian processes (GPs) are widely used metamodels for approximating expensive computer simulations, particularly in engineering design and spatial prediction. However, their performance can deteriorate significantly when covariance parameters are poorly estimated, highlighting the importance of accurate inference. The most common approach involves maximizi

Nirmit Arora, Sathvik Joel, Ishan Kavathekar, Palak

LLM-based agents are increasingly deployed in multi-agent systems (MAS). As these systems move toward real-world applications, their security becomes paramount. Existing research largely evaluates single-agent security, leaving a critical gap in understanding the vulnerabilities introduced by multi-agent design. However, existing systems fall short due to la

Ren Zhang, Huilai Li, Chao qi, Guoliang Xu

Micro expression recognition (MER) is crucial for inferring genuine emotion. Applying a multimodal large language model (MLLM) to this task enables spatio-temporal analysis of facial motion and provides interpretable descriptions. However, there are still two core challenges: (1) The entanglement of static appearance and dynamic motion cues prevents the mode

Alexander A. Donabedian, Deva D. Chan

Correlating articular cartilage material properties to quantitative magnetic resonance imaging biomarkers is a powerful approach to biofidelic finite element models. However, subject-specific relationships between imaging biomarkers such as T2 and material properties like dynamic modulus are uncertain. To evaluate the sensitivity of finite element models to

Yifan Liu, Fangneng Zhan, Kaichen Zhou, Yilun Du

Vision-language models (VLMs) struggle with 3D-related tasks such as spatial cognition and physical understanding, which are crucial for real-world applications like robotics and embodied agents. We attribute this to a modality gap between the 3D tasks and the 2D training of VLM, which led to inefficient retrieval of 3D information from 2D input. To bridge t

Xingyue Zhao, Wenke Huang, Xingguang Wang, Haoyu Zhao

Federated learning enables multiple medical institutions to train a global model without sharing data, yet feature heterogeneity from diverse scanners or protocols remains a major challenge. Many existing works attempt to address this issue by leveraging model representations (e.g., mean feature vectors) to correct local training; however, they often face tw

Tobias Heibges, Diksha Garg, Claire Guépin, Julia Burton-Heibges

Earth-skimming tau neutrinos with energies above $\sim 10$ PeV can convert to tau leptons that decay in the atmosphere and initiate upward-going extensive air showers that generate optical Cherenkov signals. On a curtailed NASA balloon flight in May 2023, the Cherenkov telescope (CT) on the Extreme Universe Space Observatory on a Super Pressure Balloon 2 (EU

Jialin Wu, Jian Yang, Handing Wang, Jiajun Wen

Model merging combines expert models for multitask performance but faces challenges from parameter interference. This has sparked recent interest in controllable model merging, giving users the ability to explicitly balance performance trade-offs. Existing approaches employ a compile-then-query paradigm, performing a costly offline multi-objective optimizati

Liuchi Xu, Hao Zheng, Lu Wang, Lisheng Xu

Knowledge distillation (KD)transfers the dark knowledge from a complex teacher to a compact student. However, heterogeneous architecture distillation, such as Vision Transformer (ViT) to ResNet18, faces challenges due to differences in spatial feature representations.Traditional KD methods are mostly designed for homogeneous architectures and hence struggle

Wenkai Liu, Nan Ma, Jianqiao Chen, Xiaoxuan Qi

Diffusion model (DM)-based channel estimation, which generates channel samples via a posteriori sampling stepwise with denoising process, has shown potential in high-precision channel state information (CSI) acquisition. However, slow sampling speed is an essential challenge for recent developed DM-based schemes. To alleviate this problem, we propose a novel

Umma Aymon, Nur Shazwani Kamarudin, Ahmad Fakhri Ab. Nasir

Facial Expression Recognition remains a challenging task, especially in unconstrained, real-world environments. This study investigates the performance of two lightweight models, YOLOv11n and YOLOv12n, which are the nano variants of the latest official YOLO series, within a unified detection and classification framework for FER. Two benchmark classification

Yuki Fujii, Toyohiro Tsurumaru

We establish a new decomposition formula for two orthogonal projections P and Q on a separable Hilbert space V. This formula yields an orthogonal direct sum decomposition of V into invariant subspaces under P and Q, each of which is either at most two dimensional or infinite dimensional. On every infinite dimensional component, the pair (P,Q) admits a matrix

R Sri Prakash, Nikhil Karamchandani, Sharayu Moharir

Motivated by the challenges of edge inference, we study a variant of the cascade bandit model in which each arm corresponds to an inference model with an associated accuracy and error probability. We analyse four decision-making policies-Explore-then-Commit, Action Elimination, Lower Confidence Bound (LCB), and Thompson Sampling-and provide sharp theoretical

Sophia Gershman, Mikhail N. Shneider, Yevgeny Raitses

Decomposition of methane using non-thermal plasmas is an attractive route for producing hydrogen-rich gases and valuable carbon nanomaterials. Understanding how plasma discharge modes influence methane decomposition in optimizing plasma-assisted chemical conversion remains unexplored. This study explores the coupling between the discharge structure and produ

Ying Song, Balaji Palanisamy

Graph unlearning has emerged as a promising solution to comply with "the right to be forgotten" regulations by enabling the removal of sensitive information upon request. However, this solution is not foolproof. The involvement of multiple parties creates new attack surfaces, and residual traces of deleted data can still remain in the unlearned graph neural

Yifan Zhuang, Calvin Huang, Zepeng Yu, Yongjie Zou

Brain-computer interface (BCI) speech decoding has emerged as a promising tool for assisting individuals with speech impairments. In this context, the integration of electroencephalography (EEG) and electromyography (EMG) signals offers strong potential for enhancing decoding performance. Mandarin tone classification presents particular challenges, as tonal

Shigeki Sugita

This article presents reflections from the perspective of a university librarian involved in the establishment and management of institutional repositories in Japan. It examines the historical evolution of scholarly communication, from the oral exchanges of ancient Greek philosophers, through the advent of printing and the rise of academic journals, to the c

Yunyi Ni, Ziyu Yang, Ze Niu, Emily Davis

Robust invisible watermarking embeds hidden information in images such that the watermark can survive various manipulations. However, the emergence of powerful diffusion-based image generation and editing techniques poses a new threat to these watermarking schemes. In this paper, we investigate the intersection of diffusion-based image editing and robust ima

Changjian Xie, Cheng Wang

Micromagnetics depends on high-fidelity numerical methods for magnetization dynamics. This work proposes a third-order temporal accuracy scheme for the Landau-Lifshitz-Gilbert equation, addressing accuracy-efficiency trade-offs in existing methods. Validated via nanostrip simulations (representative of real devices), the scheme offers two key advantages: rig

Richard J. Young, Alice M. Matthews

Biomedical text embeddings have primarily been developed using research literature from PubMed, yet clinical cardiology practice relies heavily on procedural knowledge and specialized terminology found in comprehensive textbooks rather than research abstracts. This research practice gap limits the effectiveness of existing embedding models for clinical appli

Kai Huang, Yongzhang Yang, Yuhao Chen, Yining Zhang

China is planing to launch the Tianwen-4 mission around the year 2030, with its aim being the exploration of Jupiter and its moon, Callisto. Within the realm of deep space exploration, the accuracy of ephemerides is of great importance. Current ephemerides employ a simplified rotation model for Callisto, which this study addresses by proposing a novel dynami

Kabenge Hamiss, Mohammed M. Alshahrani, Mujahid N. Syed

In this work, we propose two derivative-free methods to address the problem of large-scale nonlinear equations with convex constraints. These algorithms satisfy the sufficient descent condition. The search directions can be considered generalizations of the Modified Optimal Perry conjugate gradient method and the conjugate gradient projection method or the S

A. Moullet, D. Burgarella, T. Kataria, H. Beuther

The PRobe far-Infrared Mission for Astrophysics (PRIMA) mission concept is a proposed mission to NASA's Astrophysics Probe Explorer (APEX) call. The concept features a cryogenically cooled 1.8 m diameter telescope, and is designed to carry two science instruments covering the 24 to 264 $\mu$m wavelength range: an imaging polarimeter (PRIMAger) and a spectrom

Ryota Yamamoto, Kazushi Okamoto

For complex crowdsourcing tasks that require collaboration between multiple individuals, teams should be formed by considering both worker compatibility and expertise. Furthermore, the nature of crowdsourcing dictates the budget for tasks and workers' remuneration, and excessively large team sizes may reduce collaborative performance. To address these challe

Ha-Thanh Nguyen, Wachara Fungwacharakorn, Ken Satoh

Legal compliance in AI-driven data transfer planning is becoming increasingly critical under stringent privacy regulations such as the Japanese Act on the Protection of Personal Information (APPI). We propose a multi-agent legal verifier that decomposes compliance checking into specialized agents for statutory interpretation, business context evaluation, and