October 2025 arXiv papers — page 46

Nephtalí Eliceo Martínez-Pérez, Cupatitzio Ramírez

We study the quantum cosmology of supersymmetric, homogeneous and isotropic, higher derivative models. We recall superfield actions obtained in previous works and give classically equivalent actions leading to second order equations for the bosons, and first order for the fermions. Upon quantization, the algebra of fermions leads to a multi-component state,

Zhiyu Wang, Suman Raj, Rajkumar Buyya

Multiple Unmanned Aerial Vehicles (UAVs) cooperative Mobile Edge Computing (MEC) systems face critical challenges in coordinating trajectory planning, task offloading, and resource allocation while ensuring Quality of Service (QoS) under dynamic and uncertain environments. Existing approaches suffer from limited scalability, slow convergence, and inefficient

Zhanchao Zhou, Xiaodong Chen, Haoxing Chen, Zhenzhong Lan

Multi-head attention (MHA) has become the cornerstone of modern large language models, enhancing representational capacity through parallel attention heads. However, increasing the number of heads inherently weakens individual head capacity, and existing attention mechanisms - whether standard MHA or its variants like grouped-query attention (GQA) and groupe

Tengxue Zhang, Biao Ouyang, Yang Shu, Xinyang Chen

Pre-trained models exhibit strong generalization to various downstream tasks. However, given the numerous models available in the model hub, identifying the most suitable one by individually fine-tuning is time-consuming. In this paper, we propose \textbf{SwiftTS}, a swift selection framework for time series pre-trained models. To avoid expensive forward pro

Xu Cheng, Yue Li, Zehua Tian, Xingyu Zhao

The Unruh effect predicts an astonishing phenomenon that an accelerated detector would detect counts despite being in a quantum field vacuum in the rest frame. Since the required detector acceleration for its direct observation is prohibitively large, recent analog studies on quantum simulation platforms help to reveal various properties of the Unruh effect

Christos Thrampoulidis, Sadegh Mahdavi, Wenlong Deng

This note reconciles two seemingly distinct approaches to policy gradient optimization for the Pass@K objective in reinforcement learning with verifiable rewards: (1) direct REINFORCE-style methods, and (2) advantage-shaping techniques that directly modify GRPO. We show that these are two sides of the same coin. By reverse-engineering existing advantage-shap

Y. Alipour Fakhri

We develop a Finsler Ginzburg--Landau framework for the analysis of vortex interactions in anisotropic superconductors. Within this setting, the Finsler structure encodes directional dependence of the condensate energy, yielding a renormalized energy W_F that governs both equilibrium and dynamics of vortices. We derive the Gamma--limit, establish the analyti

Sina Kazemian, Ghazal Farhani, Amirhessam Yazdi

We present an uncertainty-aware, physics-informed neural network (PINN) for option pricing that solves the Black--Scholes (BS) partial differential equation (PDE) as a mesh-free, global surrogate over $(S,t)$. The model embeds the BS operator and boundary/terminal conditions in a residual-based objective and requires no labeled prices. For American options,

Jun-Lei Chen, Shan-Ping Wu, Shao-Wen Wei

A recently developed topological approach offers novel insights into photon spheres, which are fundamental to the formation of black hole shadows. In this study, we extend this topological analysis to higher-dimensional, static, spherically symmetric, and asymptotically flat black holes. By examining the asymptotic properties of the vector field associated w

Rohan E. Louis

Sunspots or active regions (ARs) with a delta-magnetic configuration are known to be associated with strong eruptions such as flares and mass ejections. This article investigates the relationship between delta-ARs and flares over the course of three solar cycles (SCs), from 1996 to 2024, with respect to the former's area, lifetime, latitudinal distribution,

Guiyao Tie, Pan Zhou, Lichao Sun

Artificial intelligence is undergoing a profound transition from a computational instrument to an autonomous originator of scientific knowledge. This emerging paradigm, the AI scientist, is architected to emulate the complete scientific workflow-from initial hypothesis generation to the final synthesis of publishable findings-thereby promising to fundamental

W. M. Snow, V. Kurmangaliyeva, O. Agyl-Mussapar, S. Amangeldinova

We propose a method to measure the energy dependence of the neutron optical potential $dV_{opt}/dE$ in the slow neutron energy regime. Our method makes essential use of a special property of the phase shift for a nonrelativistic neutron in moving matter, known as the neutron Fizeau effect. If a neutron traverses a medium which moves along the surfaces of its

Joungbin An, Kristen Grauman

Video temporal grounding, the task of localizing the start and end times of a natural language query in untrimmed video, requires capturing both global context and fine-grained temporal detail. This challenge is particularly pronounced in long videos, where existing methods often compromise temporal fidelity by over-downsampling or relying on fixed windows.

Harikrishnan KP, Xin Wei, Chia-Hao Lee, Dasol Yoon

The ability to tune electronic structure in twisted stacks of two-dimensional (2D) materials has motivated the exploration of similar moir\'e physics with twisted oxide membranes. Due to the intrinsic three-dimensional nature of bonding in many oxides, achieving atomic-level coupling is significantly more challenging than with van der Waals materials. Althou

Jinhu Ren, Linyuan Lü

Generated networks are widely used in network-based research as a convenient simulation environment. Generating universal networks that more accurately reflect real-world patterns is a cornerstone task. This study proposes a vari-linear network generation model that incorporates two core mechanisms: exponential probabilistic growth and vari-linear preferenti

Subhojyoti Khastagir, Kishalay Das, Pawan Goyal, Seung-Cheol Lee

Recent advances in generative modeling have shown significant promise in designing novel periodic crystal structures. Existing approaches typically rely on either large language models (LLMs) or equivariant denoising models, each with complementary strengths: LLMs excel at handling discrete atomic types but often struggle with continuous features such as ato

Ved Danait, Srijan Das, Sujoy Bhore

Approximate Nearest Neighbor (ANN) search and Approximate Kernel Density Estimation (A-KDE) are fundamental problems at the core of modern machine learning, with broad applications in data analysis, information systems, and large-scale decision making. In massive and dynamic data streams, a central challenge is to design compact sketches that preserve essent

Ran Xu, Jingjing Chen, Jiayu Ye, Yu Wu

Large Language Models (LLMs) are widely used as judges to evaluate response quality, providing a scalable alternative to human evaluation. However, most LLM judges operate solely on intrinsic text-based reasoning, limiting their ability to verify complex constraints or perform accurate computation. Motivated by the success of tool-integrated reasoning (TIR)

Shubham Barua, Sujit K. Dalui, Rikiya Okazaki, Shantanu Desai

In this work, we test the cosmic distance duality relation (CDDR) using the arbitrary redshift pivot Pad\'e-(2,1) expansion methodology developed in arXiv:2509.16196. This approach allows us to constrain cosmography parameters and test CDDR at any redshift. Further, it does not rely on data reconstructions or extrapolations of the cosmography parameters to h

Xudong Yang, Jincheng Li, Kaiwen Xing, Zhenjia Xiao

As the primary mechanism of digital authentication, user-created passwords exhibit common patterns and regularities that can be learned from leaked datasets. Password choices are profoundly shaped by external factors, including social contexts, cultural trends, and popular vocabulary. Prevailing password guessing models primarily emphasize patterns derived f

Jun Jiang, Weiming Zhang, Nenghai Yu, Kejiang Chen

Linguistic steganography enables covert communication through embedding secret messages into innocuous texts; however, current methods face critical limitations in payload capacity and security. Traditional modification-based methods introduce detectable anomalies, while retrieval-based strategies suffer from low embedding capacity. Modern generative stegano

Gokulnath Rajendran, Suman Deb, Anupam Chattopadhyay

Binarized Neural Networks (BNNs) are a class of deep neural networks designed to utilize minimal computational resources, which drives their popularity across various applications. Recent studies highlight the potential of mapping BNN model parameters onto emerging non-volatile memory technologies, specifically using crossbar architectures, resulting in impr

Fangchao Liu, Shota Nakagawa, Yuichiro Nakai, Yaoduo Wang

The strong CP problem remains one of the most important unresolved issues in the Standard Model. Spontaneous CP violation (SCPV) is a promising approach to the problem by assuming that CP is an exact symmetry of the Lagrangian but broken spontaneously at the vacuum, which enables the generation of the observed Cabibbo-Kobayashi-Maskawa (CKM) phase without re

Chen-Che Lu, Yun-Cheng Chou, Teng-Ruei Chen

Recent advances in large language models (LLMs) have enabled multi-agent reasoning systems capable of collaborative decision-making. However, in financial analysis, most frameworks remain narrowly focused on either isolated single-agent predictors or loosely connected analyst ensembles, and they lack a coherent reasoning workflow that unifies diverse data mo

Hamza Rehman, Saddam Hussain, G. Abbas, Tao Zhu

Black holes in dense astrophysical environments, such as globular clusters or in the vicinity of other massive objects, may possess accelerations. Such acceleration would modulate the characteristics of the quasi-periodic oscillations (QPOs) observed in X-ray black hole binaries. In this paper, we explore the influence of spin-aligned acceleration of a black

Hongyu Wu, Xuhui Fan, Zhangkai Wu, Longbing Cao

AutoRegressive (AR) models have demonstrated competitive performance in image generation, achieving results comparable to those of diffusion models. However, their token-by-token image generation mechanism remains computationally intensive and existing solutions such as VAR often lead to limited sample diversity. In this work, we propose a Nested AutoRegress

Di Zhang, Xun Wu, Shaohan Huang, Lingjie Jiang

Recent advances in reinforcement learning (RL) have substantially improved the training of large-scale language models, leading to significant gains in generation quality and reasoning ability. However, most existing research focuses on dense models, while RL training for Mixture-of-Experts (MoE) architectures remains underexplored. To address the instabilit

Crimson Stambaugh, Rajesh P. N. Rao

Recent studies demonstrate that diffusion planners benefit from sparse-step planning over single-step planning. Training models to skip steps in their trajectories helps capture long-term dependencies without additional memory or computational cost. However, predicting excessively sparse plans degrades performance. We hypothesize this temporal density thresh

I. R. Gabdrakhmanov, N. A Gramotkov, A. V. Kotikov, O. V. Teryaev

We investigate the Gross-Llewellyn Smith sum rule within the framework of analytic QCD. A comparison is performed between experimental data, lattice calculations, and perturbative QCD based on conventional and analytic versions of perturbation theory with different parametrizations for the twist-four contribution. We show that conventional perturbation theor

Chuan Yan, Zeng Li, Kunlin Cai, Liuhuo Wan

Virtual Reality (VR) has gained increasing traction among various domains in recent years, with major companies such as Meta, Pico, and Microsoft launching their application stores to support third-party developers in releasing their applications (or simply apps). These apps offer rich functionality but inherently collect privacy-sensitive data, such as user

Huixuan Zhang, Xiaojun Wan

With the rapid proliferation of image generative models, the authenticity of digital images has become a significant concern. While existing studies have proposed various methods for detecting AI-generated content, current benchmarks are limited in their coverage of diverse generative models and image categories, often overlooking end-to-end image editing an

Xibin Jin, Guoliang Li, Shuai Wang, Fan Liu

Integrated sensing and communication (ISAC) enables simultaneous localization, environment perception, and data exchange for connected autonomous vehicles. However, most existing ISAC designs prioritize sensing accuracy and communication throughput, treating all targets uniformly and overlooking the impact of critical obstacles on motion efficiency. To overc

Huixuan Zhang, Xiaojun Wan

Text-to-image models are known to struggle with generating images that perfectly align with textual prompts. Several previous studies have focused on evaluating image-text alignment in text-to-image generation. However, these evaluations either address overly simple scenarios, especially overlooking the difficulty of prompts with multiple different instances

Gurpreet Singh, Keshav Sood, P. Rajalakshmi, Yong Xiang

Federated learning (FL) offers a privacy-preserving paradigm for machine learning, but its application in intrusion detection systems (IDS) within IoT networks is challenged by severe class imbalance, non-IID data, and high communication overhead.These challenges severely degrade the performance of conventional FL methods in real-world network traffic classi

JaeEun Lim, Soomin Kim, Jaeyong Seo, Iori Ono

Multilingual e-commerce search is challenging due to linguistic diversity and the noise inherent in user-generated queries. This paper documents the solution employed by our team (EAR-MP) for the CIKM 2025 AnalytiCup, which addresses two core tasks: Query-Category (QC) relevance and Query-Item (QI) relevance. Our approach first normalizes the multilingual da

Bin Wang, Hui Li, AoFan Liu, BoTao Yang

Security in code generation remains a pivotal challenge when applying large language models (LLMs). This paper introduces RefleXGen, an innovative method that significantly enhances code security by integrating Retrieval-Augmented Generation (RAG) techniques with guided self-reflection mechanisms inherent in LLMs. Unlike traditional approaches that rely on f

Yuanhao Wu, Alexander van Teijlingen, Julie Watts, Zhiquan Yu

The energy landscape dictates pathways and outcomes in supramolecular selfassembly, yet harnessing it from the nano to the macro scales remains a major challenge. Here, we demonstrate liquid liquid phase separation (LLPS) as a powerful tool to navigate and engineer the energy landscapes of coassembly systems comprising disordered proteins and peptides. We qu

Zhuo Li, Junjia Liu, Dianxi Li, Tao Teng

Recent work has demonstrated the potential of diffusion models in robot bimanual skill learning. However, existing methods ignore the learning of posture-dependent task features, which are crucial for adapting dual-arm configurations to meet specific force and velocity requirements in dexterous bimanual manipulation. To address this limitation, we propose Ma

Wenxi Cai, Yuheng Wang, Naichen Shi

We introduce Coupled Flow Matching (CPFM), a framework that integrates controllable dimensionality reduction and high-fidelity reconstruction. CPFM learns coupled continuous flows for both the high-dimensional data x and the low-dimensional embedding y, which enables sampling p(y|x) via a latent-space flow and p(x|y) via a data-space flow. Unlike classical d

Giovanna Bimonte, Maria Russolillo, Han Lin Shang, Yang Yang

Model averaging techniques in the actuarial literature aim to forecast future longevity appropriately by combining forecasts derived from various models. This approach often yields more accurate predictions than those generated by a single model. The key to enhancing forecast accuracy through model averaging lies in identifying the optimal weights from a fin

Han Wu, Jie Yin

Few-shot knowledge graph relational learning seeks to perform reasoning over relations given only a limited number of training examples. While existing approaches largely adopt a meta-learning framework for enabling fast adaptation to new relations, they suffer from two key pitfalls. First, they learn relation meta-knowledge in isolation, failing to capture

Pravin Nair

The softmax function is a basic operator in machine learning and optimization, used in classification, attention mechanisms, reinforcement learning, game theory, and problems involving log-sum-exp terms. Existing robustness guarantees of learning models and convergence analysis of optimization algorithms typically consider the softmax operator to have a Lips

Chiung-Yi Tseng, Somshubhra Roy, Maisha Thasin, Danyang Zhang

There is a substantial body of literature examining the mathematical reasoning capabilities of large language models (LLMs), particularly their performance on precise arithmetic operations in autoregressive architectures. However, their ability to perform approximate reasoning in informal, fast-paced mathematical operations has received far less attention, e

Bin Wang, Zexin Liu, Hao Yu, Ao Yang

The Model Context Protocol (MCP) has emerged as a standardized interface enabling seamless integration between Large Language Models (LLMs) and external data sources and tools. While MCP significantly reduces development complexity and enhances agent capabilities, its openness and extensibility introduce critical security vulnerabilities that threaten system

Xinjian Zhao, Wei Pang, Zhongkai Xue, Xiangru Jian

Graph Neural Networks operate through bottom-up message-passing, fundamentally differing from human visual perception, which intuitively captures global structures first. We investigate the underappreciated potential of vision models for graph understanding, finding they achieve performance comparable to GNNs on established benchmarks while exhibiting distin

Sammriddh Gupta, Sonit Singh, Aditya Joshi, Mira Kim

Language educators strive to create a rich experience for learners, while they may be restricted in the extend of feedback and practice they can provide. We present the design and development of LangLingual, a conversational agent built using the LangChain framework and powered by Large Language Models. The system is specifically designed to provide real-tim

Ming-Tung Shen, Yuh-Jzer Joung

Agentic code generation requires large language models (LLMs) capable of complex context management and multi-step reasoning. Prior multi-agent frameworks attempt to address these challenges through collaboration, yet they often suffer from rigid workflows and high reasoning recovery costs. To overcome these limitations, we propose TALM (Tree-Structured Mult

Pan Zhao, Hui Yuan, Chongzhen Tian, Tian Guo

Lossy compression of point clouds reduces storage and transmission costs; however, it inevitably leads to irreversible distortion in geometry structure and attribute information. To address these issues, we propose a unified geometry and attribute enhancement (UGAE) framework, which consists of three core components: post-geometry enhancement (PoGE), pre-att

Youcan Xu, Zhen Wang, Jiaxin Shi, Kexin Li

While recent text-to-video models excel at generating diverse scenes, they struggle with precise motion control, particularly for complex, multi-subject motions. Although methods for single-motion customization have been developed to address this gap, they fail in compositional scenarios due to two primary challenges: motion-appearance entanglement and ineff

Shenran Wang, Timothy Tin-Long Tse, Jian Zhu

We perform in-depth evaluations of in-context learning (ICL) on state-of-the-art transformer, state-space, and hybrid large language models over two categories of knowledge-based ICL tasks. Using a combination of behavioral probing and intervention-based methods, we have discovered that, while LLMs of different architectures can behave similarly in task perf

Pak-Yeung Chan, Yi Lai, Man-Chun Lee

For any $n\geq 4$, we construct an $(n-2)$-parameter family of steady gradient Ricci solitons with non-negative curvature operator and prescribed by the eigenvalues of Ricci tensor at a critical point of the soliton potential. Among them lies an $(n-3)$-parameter subfamily of non-collapsed solitons. These solitons generalized the flying wings constructed by

Lei Zhang, Jiachen Guo, Shaoqiang Tang, Thomas J. R. Hughes

In this paper, we propose the MultiLevel Variational MultiScale (ML-VMS) method, a novel approach that seamlessly integrates a multilevel mesh strategy into the Variational Multiscale (VMS) framework. A key feature of the ML-VMS method is the use of the Convolutional Hierarchical Deep Neural Network (C-HiDeNN) as the approximation basis. The framework employ

ZhengKai Huang, YiKun Wang, ChenYu Hui, XiaoCheng

This paper introduces an intelligent water-saving irrigation system designed to address critical challenges in precision agriculture, such as inefficient water use and poor terrain adaptability. The system integrates advanced computer vision, robotic control, and real-time stabilization technologies via a multi-sensor fusion approach. A lightweight YOLO mode

Pravin Kumar

Small Coxeter groups are exactly those for which the Tits representation takes integral values, which makes the study of their congruence subgroups significant. In \cite{MR0938643}, Squier introduced a matrix representation of an Artin group defined over the ring $\mathbb Z[s^{\pm}, t^{\pm}]$ of Laurent polynomials in two variables. This representation simul

Diqi He, Xuehao Gao, Hao Li, Junwei Han

The Zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) task requires agents to navigate previously unseen 3D environments using natural language instructions, without any scene-specific training. A critical challenge in this setting lies in ensuring agents' actions align with both spatial structure and task intent over long-horizon

Julia Falcone, D. Michael Crenshaw, Mitchell Revalski, Travis C. Fischer

We present spatially resolved mass outflow rates of the ionized and molecular gas in the narrow line region of the Seyfert 1 galaxy NGC 3227. Using long-slit spectroscopy and [O III] imaging from from Hubble Space Telescope's Space Telescope Imaging Spectrograph and Apache Point Observatory's Kitt Peak Ohio State Multi-Object Spectrograph, in conjunction wit

Pengfei Gao, Haidi Wang

Universal machine learning interatomic potentials have emerged as efficient tools for materials simulation, yet their reliability for elastic property prediction remains unclear. Here, we present a systematic benchmark of four uMLIPs -- MatterSim, MACE, SevenNet, and CHGNet -- against first-principles data for nearly 11\,000 elastically stable materials from

Gilber A. Corrales, Carlos Andrés Ferro Sánchez, Reinel Tabares-Soto, Jesús Alfonso López Sotelo

ProfileXAI is a model- and domain-agnostic framework that couples post-hoc explainers (SHAP, LIME, Anchor) with retrieval - augmented LLMs to produce explanations for different types of users. The system indexes a multimodal knowledge base, selects an explainer per instance via quantitative criteria, and generates grounded narratives with chat-enabled prompt

Liang Chen, Xu-Feng Zhao, Shao-Zhe Lu

We investigate the finite-temperature Casimir effect for a (1+1)-dimensional scalar field interacting with a pair of delta-function potentials. We employ the canonical quantization method to compute the Casimir force and entropy, contrasting the results with those from the standard Lifshitz theory. At zero temperature, both frameworks yield identical forces.

Md Mostafijur Rahman, Radu Marculescu

U-shaped networks output logits at multiple spatial scales, each capturing a different blend of coarse context and fine detail. Yet, training still treats these logits in isolation - either supervising only the final, highest-resolution logits or applying deep supervision with identical loss weights at every scale - without exploring mixed-scale combinations

Quanjian Song, Donghao Zhou, Jingyu Lin, Fei Shen

Recent text-to-image models have revolutionized image generation, but they still struggle with maintaining concept consistency across generated images. While existing works focus on character consistency, they often overlook the crucial role of scenes in storytelling, which restricts their creativity in practice. This paper introduces scene-oriented story ge

Zidong Liu, Zhuoyan Xu, Zhenmei Shi, Yingyu Liang

Composing basic skills from simple tasks to accomplish composite tasks is crucial for modern intelligent systems. We investigate the in-context composition ability of language models to perform composite tasks that combine basic skills demonstrated in in-context examples. This is more challenging than the standard setting, where skills and their composition

Hitoshi Konno, Andrey Smirnov

We propose a new construction of vertex operators of the elliptic quantum toroidal algebra $U_{t_1,t_2,p}(\mathfrak{gl}_{N,tor})$ by combining representations of the algebra and formulas of the elliptic stable envelopes for the $A^{(1)}_{N-1}$ quiver variety ${\cal M}(v,w)$. Compositions of the vertex operators turn out consistent to the shuffle product form

Dimitris Bertsimas, Yubing Cui

Random Forests (RF) and Extreme Gradient Boosting (XGBoost) are two of the most widely used and highly performing classification and regression models. They aggregate equally weighted CART trees, generated randomly in RF or sequentially in XGBoost. In this paper, we propose Adaptive Forests (AF), a novel approach that adaptively selects the weights of the un

Youssef Megahed, Robin Ducharme, Aylin Erman, Mark Walker

Ultrasound imaging is one of the most widely used diagnostic modalities, offering real-time, radiation-free assessment across diverse clinical domains. However, interpretation of ultrasound images remains challenging due to high noise levels, operator dependence, and limited field of view, resulting in substantial inter-observer variability. Current Deep Lea

L. -H. Li, S. Benetti, Y. -Z. Cai, B. Wang

We present optical observations of the Type IIb supernova (SN) 2017ckj, covering approximately 180 days after the explosion. Its early-time multi-band light curves display no clear evidence of a shock-cooling tail, resembling the behavior of SN2008ax. The $V$-band light curve exhibits a short rise time of about 5 days and reaches an absolute fitted peak magn

Han Lin Shang, Steven Haberman

Age-specific life-table death counts observed over time are examples of densities. Non-negativity and summability are constraints that sometimes require modifications of standard linear statistical methods. The centered log-ratio transformation presents a mapping from a constrained to a less constrained space. With a time series of densities, forecasts are m

Mahsa Tavakoli, Rohitash Chandra, Cristian Bravo

Mortgage risk assessment traditionally relies on structured financial data, which is often proprietary, confidential, and costly. In this study, we propose a novel multimodal deep learning framework that uses cost-free, publicly available, unstructured data sources, including textual information, images, and sentiment scores, to generate credit scores that a

Junjie Huang, Minghua He, Jinyang Liu, Yintong Huo

Log-based anomaly detection (LogAD) is critical for maintaining the reliability and availability of large-scale online service systems. While machine learning, deep learning, and large language models (LLMs)-based methods have advanced the LogAD, they often suffer from limited interpretability, high inference costs, and extensive preprocessing requirements,

Nicholas DeJesse, Spencer Lyudovyk, Dhruv Pai

In this paper, we examine Quigley's "A Polynomial Time Algorithm for 3SAT" [Qui24]. Quigley claims to construct an algorithm that runs in polynomial time and determines whether a boolean formula in 3CNF form is satisfiable. Such a result would prove that 3SAT $\in \text{P}$ and thus $\text{P} = \text{NP}$. We show Quigley's argument is flawed by providing co

Chankyo Kim, Sicheng Zhao, Minghan Zhu, Tzu-Yuan Lin

Many scientific and geometric problems exhibit general linear symmetries, yet most equivariant neural networks are built for compact groups or simple vector features, limiting their reuse on matrix-valued data such as covariances, inertias, or shape tensors. We introduce Reductive Lie Neurons (ReLNs), an exactly GL(n)-equivariant architecture that natively s

Shuai Feng, Shiyin Shen, Yanmei Chen, Y. Sophia Dai

The morphology of ionized gas velocity maps provides a direct probe of the internal gas kinematics of galaxies. Using integral field spectroscopy from SDSS-IV MaNGA, we analyze a sample of 528 low-inclination, regular disk galaxies to investigate the correlations between velocity map morphology, star formation rate, and gas-phase metallicity. We quantify vel

Guanchen Du, Jianlong Xu, Mingtong Li, Ruiqi Wang

With the rapid advancement of internet technologies, network services have become critical for delivering diverse and reliable applications to users. However, the exponential growth in the number of available services has resulted in many similar offerings, posing significant challenges in selecting optimal services. Predicting Quality of Service (QoS) accur

Jin Hu, Jiakai Wang, Linna Jing, Haolin Li

Recently, semantically constrained adversarial examples (SemanticAE), which are directly generated from natural language instructions, have become a promising avenue for future research due to their flexible attacking forms. To generate SemanticAEs, current methods fall short of satisfactory attacking ability as the key underlying factors of semantic uncerta

Bhavya Vasudeva, Puneesh Deora, Yize Zhao, Vatsal Sharan

The growing adoption of spectrum-aware matrix-valued optimizers such as Muon and Shampoo in deep learning motivates a systematic study of their generalization properties and, in particular, when they might outperform competitive algorithms. We approach this question by introducing appropriate simplifying abstractions as follows: First, we use imbalanced data

Rikiya Okazaki, Shantanu Desai

We use model-independent luminosity distances of 186 HII galaxy observations to address the circularity problem in the Amati relation for Gamma-ray Bursts (GRBs). For this purpose, we used Artificial Neural Network based interpolation to reconstruct the luminosity distance corresponding to the GRB redshift. We then use two independent GRB datasets to test th

Ramaravind Kommiya Mothilal, Sally Zhang, Syed Ishtiaque Ahmed, Shion Guha

Reasoning is a distinctive human-like characteristic attributed to LLMs in HCI due to their ability to simulate various human-level tasks. However, this work argues that the reasoning behavior of LLMs in HCI is often decontextualized from the underlying mechanics and subjective decisions that condition the emergence and human interpretation of this behavior.

Chenlong Yin, Zeyang Sha, Shiwen Cui, Changhua Meng

Enhancing the reasoning capabilities of Large Language Models (LLMs) is a key strategy for building Agents that "think then act." However, recent observations, like OpenAI's o3, suggest a paradox: stronger reasoning often coincides with increased hallucination, yet no prior work has systematically examined whether reasoning enhancement itself causes tool hal

Lei Liu, Zhongyi Yu, Hong Wang, Huanshuo Dong

In recent years, Neural Operators(NO) have gradually emerged as a popular approach for solving Partial Differential Equations (PDEs). However, their application to large-scale engineering tasks suffers from significant computational overhead. And the fact that current models impose a uniform computational cost while physical fields exhibit vastly different c

Rui Hespanha, Elliot McGuire, João Hespanha

Conventional fluid simulations can be time consuming and energy intensive. We researched the viability of a neural network for simulating incompressible fluids in a randomized obstacle-heavy environment, as an alternative to the numerical simulation of the Navier-Stokes equation. We hypothesized that the neural network predictions would have a relatively low

Rishit Dagli, Donglai Xiang, Vismay Modi, Charles Loop

Physical simulation relies on spatially-varying mechanical properties, often laboriously hand-crafted. VoMP is a feed-forward method trained to predict Young's modulus ($E$), Poisson's ratio ($\nu$), and density ($\rho$) throughout the volume of 3D objects, in any representation that can be rendered and voxelized. VoMP aggregates per-voxel multi-view feature

Takeshi Fukuyama

Bose-Einstein Condensation (BEC) cosmology is analyzed in the framework of a string-inspired axion model. The dispersion relation of the axionic mode includes both gravitational and self-interaction terms, the latter being small in magnitude but crucial for inducing instability of the condensate. The generation rate of BEC around redshift $z\approx 30$ is pr

G. P. Zhang, Y. H. Bai, Thomas F. George

It is now well established that a laser pulse can demagnetize a ferromagnet. However, for a long time, it has not had an analytic theory because it falls into neither nonlinear optics (NLO) nor magnetism. Here we attempt to fill this gap by developing a nonlinear optical theory centered on the spin moment, instead of the more popular susceptibility. We first

Myeongseob Ko, Nikhil Reddy Billa, Adam Nguyen, Charles Fleming

The memorization of training data in large language models (LLMs) poses significant privacy and copyright concerns. Existing data extraction methods, particularly heuristic-based divergence attacks, often exhibit limited success and offer limited insight into the fundamental drivers of memorization leakage. This paper introduces Confusion-Inducing Attacks (C

Sudiksha Das, Ashish Kundu

Introduced by Juels and Rivest in 2013, Honeywords, which are decoy passwords stored alongside a real password, appear to be a proactive method to help detect password credentials misuse. However, despite over a decade of research, this technique has not been adopted by major authentication platforms. This position paper argues that the core concept of Honey

Zhangkai Wu, Xuhui Fan, Zhongyuan Xie, Kaize Shi

Recent advances in training-free video editing have enabled lightweight and precise cross-frame generation by leveraging pre-trained text-to-image diffusion models. However, existing methods often rely on heuristic frame selection to maintain temporal consistency during DDIM inversion, which introduces manual bias and reduces the scalability of end-to-end in

Kechen Meng, Sinuo Zhang, Rongpeng Li, Xiangming Meng

In wireless communication systems, efficient and adaptive resource allocation plays a crucial role in enhancing overall Quality of Service (QoS). Compared to the conventional Model-Free Reinforcement Learning (MFRL) scheme, Model-Based RL (MBRL) first learns a generative world model for subsequent planning. The reuse of historical experience in MBRL promises

Michael Hardy

Objective and scalable measurement of teaching quality is a persistent challenge in education. While Large Language Models (LLMs) offer potential, general-purpose models have struggled to reliably apply complex, authentic classroom observation instruments. This paper uses custom LLMs built on sentence-level embeddings, an architecture better suited for the l

Yucheng Ning, Xixun Lin, Fang Fang, Yanan Cao

The widespread adoption of Large Language Models (LLMs) raises critical concerns about the factual accuracy of their outputs, especially in high-risk domains such as biomedicine, law, and education. Existing evaluation methods for short texts often fail on long-form content due to complex reasoning chains, intertwined perspectives, and cumulative information

Qing Chen, Shuang-Yong Zhou

We develop a general framework for the computation of light portal dark matter direct detection, incorporating a consistent treatment of finite momentum transfer. In this framework, dark matter interacts with Standard Model matter through a light mediator, which simultaneously serves as the force carrier for dark matter self-interaction, potentially with a d

De-Xin Wang, Su-Ya-La-Tu Zhang, Wei Jiang, Rui-Rui Fan

The neutron capture yield of 165Ho have been measured at the Back-streaming White neutron beam line (Back-n) of the China Spallation Neutron Source (CSNS) using a 4{\pi} BaF2 Gamma Total Absorption Facility (GTAF). The resonance shapes in the 1eV to 1.0keV region were analyzed with the Bayesian R-matrix code SAMMY. For 18 s-wave resonances below 100eV, the r

Liling Yang, Ning Chen, Jun Yue, Yidan Liu

Foundation models have transformed natural language processing and computer vision, and their impact is now reshaping remote sensing image analysis. With powerful generalization and transfer learning capabilities, they align naturally with the multimodal, multi-resolution, and multi-temporal characteristics of remote sensing data. To address unique challenge

Takeshi Fukuyama

The QCD axion is investigated within the minimal supersymmetric SO(10) grand unified theory, where the Yukawa sector involves Higgs multiplets ${\bf 10}$ and $\overline{{\bf 126}}$. The relative phase between the VEVs of $({\bf 10,1,3})\subset\overline{{\bf 126}}$ and $({\bf \overline{10},1,3})\subset{\bf 126}$ under ${\rm SU}(4)_C\times{\rm SU}(2)_L\times{\

Jing-Xuan Zhang, Genshun Wan, Jin Li, Jianqing Gao

While speech foundation models (SFMs) have demonstrated remarkable performance in audio-only tasks, their adaptation to multimodal scenarios remains underexplored. This work presents UASR-LLM, a novel framework that adapts frozen SFMs to unified visual speech recognition (VSR), automatic speech recognition (ASR), and audio-visual speech recognition (AVSR) by

Zhangkai Wu, Xuhui Fan, Zhongyuan Xie, Kaize Shi

Training-free video editing (VE) models tend to fall back on gender stereotypes when rendering profession-related prompts. We propose \textbf{FAME} for \textit{Fairness-aware Attention-modulated Video Editing} that mitigates profession-related gender biases while preserving prompt alignment and temporal consistency for coherent VE. We derive fairness embeddi

Hai N. Ngo, Dieu D. Nguyen, Tinh T. Q. Le, Tien H. T. Ho

Intermediate-mass black holes (IMBHs; $M_{BH} \approx 10^{3-5} M_\odot$) play a critical role in understanding the formation of supermassive black holes in the early universe. In this study, we expand on Nguyen et al. simulated measurements of IMBH masses using stellar kinematics, which will be observed with the High Angular Resolution Monolithic Optical and

Yuhao Chen, Chengcai Shen, Zhixing Mei, Jing Ye

Newly emerging flux (NEF) has been widely studied as a trigger of solar filament eruptions, but its influence on the subsequent dynamics remains poorly explored. Because NEF typically emerges adjacent to filaments, it imposes magnetic asymmetry that can drive non-radial eruptions and complicate space-weather forecasting. We bridge analytic catastrophe theory

Ziyang Gao, Yuanliang Qu, Yi Han

Cross-lingual sponsored search is crucial for global advertising platforms, where users from different language backgrounds interact with multilingual ads. Traditional machine translation methods often fail to capture query-specific contextual cues, leading to semantic ambiguities that negatively impact click-through rates (CTR) and conversion rates (CVR). T

Anwesan Pal, Karen Hovsepian, Tinghao Guo, Mengnan Zhao

Recent investigations into effective context lengths of modern flagship large language models (LLMs) have revealed major limitations in effective question answering (QA) and reasoning over long and complex contexts for even the largest and most impressive cadre of models. While approaches like retrieval-augmented generation (RAG) and chunk-based re-ranking a

Junhao Fan, Wenrui Liang, Wei-Qiang Zhang

Accurate prediction of remaining useful life (RUL) is essential to enhance system reliability and reduce maintenance risk. Yet many strong contemporary models are fragile around fault onset and opaque to engineers: short, high-energy spikes are smoothed away or misread, fixed thresholds blunt sensitivity, and physics-based explanations are scarce. To remedy