October 2025 arXiv papers — page 135

Xi Xiao, Yunbei Zhang, Lin Zhao, Yiyang Liu

In computer vision, Visual Prompting (VP) and Visual Prompt Tuning (VPT) have recently emerged as lightweight and effective alternatives to full fine-tuning for adapting large-scale vision models within the "pretrain-then-finetune" paradigm. However, despite rapid progress, their conceptual boundaries remain blurred, as VP and VPT are frequently used interch

Xiaofan Wang, Haitao Lu, Hengyan Wang, Zhihuang Luo

Nonlinear spin systems exhibit rich and exotic dynamical phenomena, offering promising applications ranging from spin masers and time crystals to precision measurement. Recent theoretical work [T. Wang et al., Commun. Phys. 8, 41 (2025)] predicted intriguing nonlinear dynamical phases arising from inhomogeneous magnetic fields and feedback interactions. Howe

Akhilesh Yadav, Tarun Saxena

The aim of this paper is to study geometrical aspects of static spacetime admitting an almost gradient Ricci soliton. Among others, We first determine the conditions under which the base manifold of static spacetime possess an almost gradient Ricci soliton and we show that the almost gradient Ricci soliton become steady gradient Ricci soliton when static spa

Joy Jia Yin Lim, Ye He, Jifan Yu, Xin Cong

Personalized Learning Path Planning (PLPP) aims to design adaptive learning paths that align with individual goals. While large language models (LLMs) show potential in personalizing learning experiences, existing approaches often lack mechanisms for goal-aligned planning. We introduce Pxplore, a novel framework for PLPP that integrates a reinforcement-based

Zehui Ling, Deshu Chen, Yichi Zhang, Yuchen Liu

Recent advances in Large Language Models (LLMs) demonstrate that chain-of-thought prompting and deep reasoning substantially enhance performance on complex tasks, and multi-agent systems can further improve accuracy by enabling model debates. However, applying deep reasoning to all problems is computationally expensive. To mitigate these costs, we propose a

Zizhuo Zhang, Qizhou Wang, Shanshan Ye, Jianing Zhu

Large language model (LLM) alignment is typically achieved through learning from human preference comparisons, making the quality of preference data critical to its success. Existing studies often pre-process raw training datasets to identify valuable preference pairs using external reward models or off-the-shelf LLMs, achieving improved overall performance

Prawaal Sharma, Navneet Goyal, Poonam Goyal, Vishnupriyan R

Linguistic diversity across the world creates a disparity with the availability of good quality digital language resources thereby restricting the technological benefits to majority of human population. The lack or absence of data resources makes it difficult to perform NLP tasks for low-resource languages. This paper presents a novel scalable and fully auto

Yasushi Hasegawa, Masayuki Ohzeki

We compare Ising ({-1,+1}) and QUBO ({0,1}) encodings for Boltzmann machine learning under a controlled protocol that fixes the model, sampler, and step size. Exploiting the identity that the Fisher information matrix (FIM) equals the covariance of sufficient statistics, we visualize empirical moments from model samples and reveal systematic, representation-

Lipeng Zhu, Haobin Mao, Ge Yan, Wenyan Ma

The growing demands of 6G mobile communication networks necessitate advanced antenna technologies. Movable antennas (MAs) and reconfigurable antennas (RAs) enable dynamic control over antenna's position, orientation, radiation, polarization, and frequency response, introducing rich electromagnetic-domain degrees of freedom for the design and performance enha

Rutvik Kokate, Pranati Kompella, Prasad Onkar

The creative potential of computers has intrigued researchers for decades. Since the emergence of Generative AI (Gen AI), computer creativity has found many new dimensions and applications. As Gen AI permeates mainstream discourse and usage, researchers are delving into how it can improve and complement what humans do. Creative potential is a highly relevant

Lianlian Liu, YongKang He, Zhaojie Chu, Xiaofen Xing

Generating stylized 3D human motion from speech signals presents substantial challenges, primarily due to the intricate and fine-grained relationships among speech signals, individual styles, and the corresponding body movements. Current style encoding approaches either oversimplify stylistic diversity or ignore regional motion style differences (e.g., upper

Feng Chen, Shu Li, Xin Chen, Dennis Wong

Owing to its high scalability and computational efficiency, machine learning methods have been increasingly integrated into various scientific research domains, including ab initio-based materials design. It has been demonstrated that, by incorporating modern machine learning algorithms, one can predict material properties with practically acceptable accurac

Liam Packer, Kihoon Seong, Philippe Sosoe

We prove the large deviation principle for the conditional Gibbs measure associated with the focusing Gross Pitaevskii equation in the low temperature regime. This conditional measure is of mixed type, being canonical in energy and microcanonical in particle number. In particular, our result extends the large deviation principle for the mixed ensemble studie

Amirhossein Mozafari, Kourosh Hashemi, Erfan Shafagh, Soroush Motamedi

Healthcare fraud detection remains a critical challenge due to limited availability of labeled data, constantly evolving fraud tactics, and the high dimensionality of medical records. Traditional supervised methods are challenged by extreme label scarcity, while purely unsupervised approaches often fail to capture clinically meaningful anomalies. In this wor

Mehdi Zekriyapanah Gashti

In this paper we represent a new framework for integrated distributed systems. In the proposed framework we have used three parts to increase Satisfaction and Performance of this framework. At first we analyse integrated systems and their evolution process and also ERPSD and ERPDRT framework briefly then we explain the new FDIRS framework. Finally we compare

Sai Suhruth Reddy Karri, Yashwanth Sai Nallapuneni, Laxmi Narasimha Reddy Mallireddy, Gopichand G

Bias in AI systems, especially those relying on natural language data, raises ethical and practical concerns. Underrepresentation of certain groups often leads to uneven performance across demographics. Traditional fairness methods, such as pre-processing, in-processing, and post-processing, depend on protected-attribute labels, involve accuracy-fairness tra

Janghan Yoon, Jaegwan Cho, Junhyeok Kim, Jiwan Chung

Large language models can generate visually coherent web UIs from natural language requests, but they frequently violate Web Content Accessibility Guidelines (WCAG), excluding users with diverse needs and contexts. We address this by introducing A11yn, a post-training framework for web accessibility-aware web UI generation. A11yn converts off-the-shelf WCAG

Jing Yang, Qiyao Wei, Jiaxin Pei

The rapid growth of AI conferences is straining an already fragile peer-review system, leading to heavy reviewer workloads, expertise mismatches, inconsistent evaluation standards, superficial or templated reviews, and limited accountability under compressed timelines. In response, conference organizers have introduced new policies and interventions to prese

Luka Filin

In this note, we study extension properties of finite abelian subgroups of $\mathrm{Bir}(X)$ where $X$ is a rational (or rationally connected) variety of dimension at most $4$. We are guided by the following question: is it true that if a finite group $G$ faithfully acts on a rationally connected variety of dimension $n$, then $G$ can faithfully act on a ter

Jongwon David Kim, Jack Xin

Tumor angiogenesis involves a collection of tumor cells moving towards blood vessels for nutrients to grow. Angiogenesis, and in general chemotaxis systems have been modeled using partial differential equations (PDEs) and as such require numerical methods to approximate their solutions in 3 space dimensions (3D). This is an expensive computation when solutio

Rongtao Xu, Jinzhou Lin, Jialei Zhou, Jiahua Dong

Camera-based occupancy prediction is a mainstream approach for 3D perception in autonomous driving, aiming to infer complete 3D scene geometry and semantics from 2D images. Almost existing methods focus on improving performance through structural modifications, such as lightweight backbones and complex cascaded frameworks, with good yet limited performance.

Yang Cao, Sikun Yang, Yujiu Yang, Lianyong Qi

Two-step approaches combining pre-trained large language model embeddings and anomaly detectors demonstrate strong performance in text anomaly detection by leveraging rich semantic representations. However, high-dimensional dense embeddings extracted by large language models pose challenges due to substantial memory requirements and high computation time. To

Shrey Pandit, Xuan-Phi Nguyen, Yifei Ming, Austin Xu

Web-based 'deep research' agents aim to solve complex question - answering tasks through long-horizon interactions with online tools. These tasks remain challenging, as the underlying language models are often not optimized for long-horizon reasoning and exploration. Prior work has proposed workflows for constructing instruction-tuning datasets, often levera

Ajinkya Desai, Antonio Quim Cervantes, Tirtha Banerjee

The interaction of a buoyant plume with a plant canopy results in turbulent flow features distinct from those in a grassland environment. In this work, we model the turbulence dynamics of a buoyant plume in a homogeneous plant canopy with a crosswind using large-eddy simulations. As the plume interacts with the crosswind, we observe increased vorticity at th

Qun Ma, Xiao Xue, Xuwen Zhang, Zihan Zhao

The advent of large language models (LLMs) has enabled agents to represent virtual humans in societal simulations, facilitating diverse interactions within complex social systems. However, existing LLM-based agents exhibit severe limitations in affective cognition: They fail to simulate the bounded rationality essential for bridging virtual and real-world se

Xi Chen, Yuchen Song, Satoshi Nakamura

We propose a stress-aware speech-to-speech translation (S2ST) system that preserves word-level emphasis by leveraging LLMs for cross-lingual emphasis conversion. Our method translates source-language stress into target-language tags that guide a controllable TTS model. To overcome data scarcity, we developed a pipeline to automatically generate aligned train

Yikuan Hu, Jifeng Zhu, Lanrui Tang, Chen Huang

Knowledge graphs (KGs), with their structured representation capabilities, offer promising avenue for enhancing Retrieval Augmented Generation (RAG) systems, leading to the development of KG-RAG systems. Nevertheless, existing methods often struggle to achieve effective synergy between system effectiveness and cost efficiency, leading to neither unsatisfying

Qiang Du, Shreeharshini Murthy, Victoria Moore, Angel Jurado Lopez

The Advanced Light Source (ALS) at LBNL is upgrading several LLRF systems for its Linac and Sub-Harmonic Bunchers, where it is desired to have a unified LLRF system design to support various RF frequencies (at 125MHz, 500MHz and 3GHz) and configurations. This paper demonstrates an open-source, direct sampling RFSoC based LLRF system design, featuring: sample

Jiamin Chen, Yuchen Li, Xinyu Ma, Xinran Chen

Retrieval-Augmented Generation (RAG) has become an essential approach for extending the reasoning and knowledge capacity of large language models (LLMs). While prior research has primarily focused on retrieval quality and prompting strategies, the influence of how the retrieved documents are framed, i.e., context format, remains underexplored. We show that s

Junlin Yang

This study examines how institutional differences and external crises shape volatility dynamics in emerging Asian stock markets. Using daily stock index returns for Indonesia, Malaysia, and the Philippines from 2010 to 2024, we estimate EGARCH(1,1) and TGARCH(1,1) models in a by-window design. The sample is split into the 2013 Taper Tantrum, the 2020-2021 CO

Juan Ren, Mark Dras, Usman Naseem

Large Vision-Language Models (LVLMs) unlock powerful multimodal reasoning but also expand the attack surface, particularly through adversarial inputs that conceal harmful goals in benign prompts. We propose SHIELD, a lightweight, model-agnostic preprocessing framework that couples fine-grained safety classification with category-specific guidance and explici

Ajinkya Desai, Antonio Quim Cervantes, Tirtha Banerjee

Tracking the structure and geometric properties of a buoyant plume in cross-wind is critical for managing smoke hazards and improving disaster mitigation efforts. Plume features, such as the tilt angle, centerline trajectory, plume height, and curvature changes with height, are impacted by a range of forcing parameters, with the altered turbulence patterns i

Sudipta Paul, Amanda W. Lund, George Jour, Iman Osman

The structural and spatial arrangements of cells within tissues represent their functional states, making graph-based learning highly suitable for histopathology image analysis. Existing methods often rely on fixed graphs with predefined edges, limiting their ability to capture the true biological complexity of tissue interactions. In this work, we propose A

A. P. Mahtessian, G. S. Karapetian, H. F. Khachatryan, M. A. Hovhannisyan

We show that the parameters used to standardize the luminosity of Type 1a supernovae in the SALT2 and SiFTO models are strongly dependent on the redshift z. Consequently, when standardized with increasing z, the average absolute magnitudes of Type 1a supernovae are artificially increased. This means that for a given apparent magnitude they are, on average, a

Zhen Li, Xibin Jin, Guoliang Li, Shuai Wang

Edge Gaussian splatting (EGS), which aggregates data from distributed clients (e.g., drones) and trains a global GS model at the edge (e.g., ground server), is an emerging paradigm for scene reconstruction in low-altitude economy. Unlike traditional edge resource management methods that emphasize communication throughput or general-purpose learning performan

Hong Jian Zhao, Laurent Bellaiche, Yanming Ma

Polar distortion, the collective off-center displacements of atoms, is a fingerprint of a ferroelectric that governs its properties and functionalities. Since the 1970s, the concepts of proper, improper and triggered ferroelectrics have been established to shed light on a diversity of polar distortion mechanisms. Such concepts assign a single nature to polar

Haolin Pan, Jinyuan Dong, Mingjie Xing, Yanjun Wu

Compiler optimization relies on sequences of passes to improve program performance. Selecting and ordering these passes automatically, known as compiler auto-tuning, is challenging due to the large and complex search space. Existing approaches generally assume a linear sequence of passes, a model compatible with legacy compilers but fundamentally misaligned

Ming Dong, Jinkui Zhang, Bolong Zheng, Xinhui Tu

Detoxification in large language models (LLMs) remains a significant research challenge. Existing decoding detoxification methods are all based on external constraints, which require additional resource overhead and lose generation fluency. This work proposes Detoxification with Self-Constrained Decoding (DSCD), a novel method for LLM detoxification without

Rongrong Xie, Yizhou Xu, Guido Sanguinetti

The rapid increase in multimodal data availability has sparked significant interest in cross-modal knowledge distillation (KD) techniques, where richer "teacher" modalities transfer information to weaker "student" modalities during model training to improve performance. However, despite successes across various applications, cross-modal KD does not always re

Qi Chen, Hao Jia, Dongyi Wei, Zhifei Zhang

In this paper we prove the asymptotic stability of the Kolmogorov flow on a non-square torus for perturbations $\omega_0$ satisfying $\|\omega_0\|_{H^3}\ll\nu^{1/3}$, where $0<\nu\ll1$ is the viscosity. Kolmogorov flows are important metastable states to the two dimensional incompressible Navier Stokes equations in the high Reynolds number regime. Our result

Qi Qi, Abdelhamid Tayebi, Daizhan Cheng, Jun-e Feng

In compressed sensing (CS), sparse signals can be reconstructed from significantly fewer samples than required by the Nyquist-Shannon sampling theorem. While non-sparse signals can be sparsely represented in appropriate transformation domains, conventional CS frameworks rely on the incoherence of the measurement matrix columns to guarantee reconstruction per

Himanshi Singh, Abhik Ghosh, Nil Kamal Hazra

Traditional likelihood based methods for parameter estimation get highly affected when the given data is contaminated by outliers even in a small proportion. In this paper, we consider a robust parameter estimation method, namely the minimum logarithmic norm relative entropy (LNRE) estimation procedure, and study different (generalized) sufficiency principle

Zhen Zhang, Zhencheng Xie, Walter Kob

Silica is the paradigmatic network glass-former and understanding its response to pressure is essential for comprehending the mechanical properties of silica-based materials and the behavior of silicate melts in the Earth's interior. While pressure-induced changes in the short-range structure - particularly the breakdown of tetrahedral symmetry - have been w

Árpád Baricz, Pranav Kumar, Sanjeev Singh

Motivated by the pioneering work of M.S. Robertson [Ro54] and R.K. Brown [Br60], [Br62], who examined the geometric properties of some normalised solutions of second-order homogeneous differential equations, in this paper we investigate the radii of univalence and starlikeness for two kind of normalised regular Coulomb wave functions. Moreover, a generalized

Takeshi Chiba, Hiroki Matsui, Keiju Murata

The quantum nature of the Schwarzschild black hole interior is investigated through the Wheeler-DeWitt (WDW) equation. The interior of a static, spherically symmetric black hole is described by the Kantowski-Sachs (KS) metric, which represents a homogeneous but anisotropic cosmology. We derive the Hamiltonian for the gravitational system corresponding to the

Himanshi Singh, Tanmay Sahoo, Nil Kamal Hazra

From the perspective of data reduction, the notions of minimal sufficient and complete statistics together play an important role in determining optimal statistics (estimators). The classical notion of sufficiency and completeness are not adequate in many robust estimations that are based on different divergences. Recently, the notion of generalized sufficie

Abhinav Jangir, Anees Ahmed

We investigate the combined effects of carrier envelope phase and laser pulse shape on electron-positron pair production in the presence of an external asymmetric super-Gaussian electric field by solving the quantum Vlasov equation. By varying the field asymmetry, the pulse shape from Gaussian to super-Gaussian, and the carrier envelope phase, we show the mo

J. Bae, M. Bergevin, E. P. Bernard, D. S. Bhattacharya

The BUTTON-30 detector is a 30-tonne technology demonstrator designed to evaluate the potential of hybrid event detection, simultaneously exploiting both Cherenkov and scintillation light to detect particles produced in neutrino interactions. The detector is installed at a depth of 1.1 km in the Boulby Underground Laboratory allowing to test the performance

Yoshiyasu Ozeki, Manabu Yoshida

In this paper, for every prime $p$ and every $0\le n\le \infty$, we classify the structure of the torsion subgroup of the group of $\mathbb{Q}_p(\mu_{p^n})$-rational points of elliptic curves over $\mathbb{Q}_p$ with good reduction, where $\mu_{p^n}$ is the set of the $p^n$-th roots of unity.

Jun Qian, Ross Murch, Khaled B. Letaief

Active reconfigurable intelligent surfaces (RISs) employ amplification to overcome attenuation caused by the RIS cascaded link. In this paper, we analyze the effects of phase errors and channel aging in active simultaneously transmitting and reflecting (STAR) RIS-assisted cell-free massive multiple-input multiple-output (MIMO) systems. By leveraging a spatia

Xiaoshu Chen, Sihang Zhou, Ke Liang, Duanyang Yuan

Chain of thought (CoT) fine-tuning aims to endow large language models (LLMs) with reasoning capabilities by training them on curated reasoning traces. It leverages both supervised and reinforced fine-tuning to cultivate human-like reasoning skills in LLMs, including detailed planning, divergent thinking, intuitive judgment, timely reflection, internal think

Tong Qiao, Ao Zhou, Yingjie Qi, Yiou Wang

Graph Neural Networks (GNNs) have been widely adopted due to their strong performance. However, GNN training often relies on expensive, high-performance computing platforms, limiting accessibility for many tasks. Profiling of representative GNN workloads indicates that substantial efficiency gains are possible on resource-constrained devices by fully exploit

Jiacheng Cen, Anyi Li, Ning Lin, Tingyang Xu

Equivariant Graph Neural Networks (GNNs) have demonstrated significant success across various applications. To achieve completeness -- that is, the universal approximation property over the space of equivariant functions -- the network must effectively capture the intricate multi-body interactions among different nodes. Prior methods attain this via deeper a

Michael P. Lamoureux, Matt Yedlin

We propose a novel foundation for calculus that focuses on the notion of approximations while avoiding the use of limits altogether. Continuity is defined as approximation at a point, while differentiability is defined as approximation with a linear function. The errors in approximation are defined as a class of functions with certain properties; rules for c

Aditya Ganeshan, Kurt Fleischer, Wenzel Jakob, Ariel Shamir

Traditional integral wood joints, despite their strength, durability, and elegance, remain rare in modern workflows due to the cost and difficulty of manual fabrication. CNC milling offers a scalable alternative, but directly milling traditional joints often fails to produce functional results because milling induces geometric deviations, such as rounded inn

Wen-Mei Li, Jianbo Lu, Shu-Min Wu

We investigate tripartite quantum-memory-assisted entropic uncertain and quantum coherence for GHZ and W states of a fermionic field in the background of a spherically symmetric black hole of Einstein-Gauss-Bonnet (EGB) gravity. Two distinct scenarios are analyzed: (i) the quantum memories (held by Bob and Charlie) are near the horizon while the measured par

Kehua Feng, Keyan Ding, Zhihui Zhu, Lei Liang

While chain-of-thought (CoT) distillation from advanced large language models (LLMs) has proven effective in general reasoning tasks, it struggles in scientific domains where even advanced models often produce incorrect or superficial reasoning due to high complexity and specialized knowledge requirements. Directly distilling from such flawed outputs results

Xiaoxin Chen, Zhaoyang Yin

In this paper, we consider the fifth order Camassa-Holm model. Firstly, we improve the local well-posedness results in \cite{TangLiu2015,FOCH2021}. Secondly, we give the blow up criteria and conditions for global existence. Finally, when $b=\frac 53$ in the model, we obtain the ill-posedness in $B^1_{\infty,1}$ and $B^{\frac 32}_{2,q}$ with $q\in(1,+\infty]$

Yiyao Lian, Zhaoyang Yin

In this paper, we study a new fifth-order Camassa-Holm type equation derived by Li \cite{Li.Z}. We firstly establish the local well-posedness in the sense of Hadamard for the Cauchy problem of the new fifth-order Camassa-Holm type equation in Besov spaces. Secondly, we obtain blow-up criteria. Building upon this, by utilizing the conservation laws and establ

Nyx Iskandar, Hisham Bedri, Andy Tsen

Most large language models (LLMs) today excel at generating raw, sequential code with minimal abstractions and custom structures. However, there has been little work on graph-based abstract code generation, where significant logic is encapsulated in predefined nodes and execution flow is determined by edges. This is relevant for visual programming languages,

Taylor Robinson, Rikke Bjerg Jensen

We report on two months of ethnographic fieldwork in a women's centre in Pattaya, and interviews with 76 participants. Our findings, as they relate to digital security, show how (i) women in Pattaya, often working in the sex and massage industries, perceived relationships with farang men as their best, and sometimes only, option to achieve security; (ii) the

Nikhil Bhendawade, Kumari Nishu, Arnav Kundu, Chris Bartels

Speculative decoding accelerates LLM inference by using a draft model to look ahead, but gains are capped by the cost of autoregressive draft generation: increasing draft size elevates acceptance rates but introduces additional latency overhead exacerbating the speed-accuracy tradeoff. Prior methods (Medusa, Hydra, EAGLE) partially reduce draft cost but eith

Meng Yang, Kecheng Chen, Wei Luo, Xianjie Chen

Transient Electromagnetic (TEM) method is widely used in various geophysical applications, providing valuable insights into subsurface properties. However, time-domain TEM signals are often submerged in various types of noise. While recent deep learning-based denoising models have shown strong performance, these models are mostly trained on simulated or sing

Hung Hung, Zhi-Yu Jou, Su-Yun Huang, Shinto Eguchi

Principal component analysis (PCA) is a fundamental tool in multivariate statistics, yet its sensitivity to outliers and limitations in distributed environments restrict its effectiveness in modern large-scale applications. To address these challenges, we introduce the $\phi$-PCA framework which provides a unified formulation of robust and distributed PCA. T

Haolin Pan, Jinyuan Dong, Hongbin Zhang, Hongyu Lin

Learning effective numerical representations, or embeddings, of programs is a fundamental prerequisite for applying machine learning to automate and enhance compiler optimization. Prevailing paradigms, however, present a dilemma. Static representations, derived from source code or intermediate representation (IR), are efficient and deterministic but offer li

Subhendu Khatuya, Shashwat Naidu, Pawan Goyal, Niloy Ganguly

Despite continuous advancements in the capabilities of large language models (LLMs), numerical reasoning remains a challenging area. Techniques like chain-of-thought prompting, tree-of-thought prompting, and program-of-thought prompting guide LLMs through intermediate reasoning steps. Although in-context learning with few-shot prompting has improved performa

Oleg Makarenkov, Marianne Bezaire, Michael Hasselmo

The theta rhythm is important for many cognitive functions including spatial processing, memory encoding, and memory recall. The information processing underlying these functions is thought to rely on consistent, phase-specific spiking throughout a theta oscillation that may fluctuate significantly in baseline (center of oscillations), frequency, or amplitud

Seyed Naseh Sajadi, Supakchai Ponglertsakul

We study New Massive Gravity (NMG) with Chern-Simons (CS), cubic, and quartic terms under the Comp\`ere-Song-Strominger (CSS) boundary conditions. By employing a semi-product of a Virasoro and a $U(1)$ Kac-Moody current algebra as the asymptotic symmetry algebra, we calculate the entropy of BTZ black holes via the degeneracy of states belonging to a Warped-C

Bin-Hui Chen, Sandeep Kumar Kataria, Juntai Shen, Meng Guo

Bars are among the most prominent structures in disk galaxies. While the widely accepted swing-amplification theory provides a qualitative framework for their formation, the detailed physical processes remain incompletely understood. Previous studies have shown that the bar formation timescale in isolated galaxies depends exponentially on the disk mass fract

Bin-Hui Chen, Juntai Shen

Bars are one of the most prominent galactic structures. The classical swing-amplification theory can qualitatively describe the spontaneous bar instability of stellar disks. Still, it cannot quantify the bar formation process or explain why some disk galaxies do not have a bar. Recent studies found that the bar formation timescale depends exponentially on th

María Victoria Carro, Denise Alejandra Mester, Facundo Nieto, Oscar Agustín Stanchi

The core premise of AI debate as a scalable oversight technique is that it is harder to lie convincingly than to refute a lie, enabling the judge to identify the correct position. Yet, existing debate experiments have relied on datasets with ground truth, where lying is reduced to defending an incorrect proposition. This overlooks a subjective dimension: lyi

Lifeng Qiu Lin, Henry Kam, Qi Sun, Kaan Akşit

Steganography finds its use in visual medium such as providing metadata and watermarking. With support of efficient latent representations and foveated rendering, we trained models that improve existing capacity limits from 100 to 500 bits, while achieving better accuracy of up to 1 failure bit out of 2000, at 200K test bits. Finally, we achieve a comparable

Tai Xiang, Yue-Hui Lu, Jacquelyn Ho, Tsai-Chen Lee

Ladder-type two-photon excitation of an atom from a ground state $|g\rangle$, to an intermediate excited state $|e\rangle$, and, finally, to a Rydberg state $|r\rangle$, has a variety of uses from quantum information to sensing. A common scheme for detecting this transition optically is through electromagnetically induced transparency (EIT). However, in inve

Yangtao Chen, Zixuan Chen, Nga Teng Chan, Junting Chen

Enabling robots to flexibly schedule and compose learned skills for novel long-horizon manipulation under diverse perturbations remains a core challenge. Early explorations with end-to-end VLA models show limited success, as these models struggle to generalize beyond the training distribution. Hierarchical approaches, where high-level planners generate subgo

Tatsuru Kikuchi

I develop a nonparametric framework for identifying spatial boundaries of treatment effects without imposing parametric functional form restrictions. The method employs local linear regression with data-driven bandwidth selection to flexibly estimate spatial decay patterns and detect treatment effect boundaries. Monte Carlo simulations demonstrate that the n

Faraz Tahmasebi, Michael Pelluer, Hyoukjun Kwon

The computation and memory costs of large language models kept increasing over last decade, which reached over the scale of 1T parameters. To address the challenges from the large scale models, model compression techniques such as low-rank decomposition have been explored. Previous model decomposition works have focused on weight decomposition to avoid costl

Lyu Yi, Weiqi Feng, Yuanbiao Wang, Yuhong Kan

Cardinality estimation is a key component of database query optimization. Recent studies have demonstrated that learned cardinality estimation techniques can surpass traditional methods in accuracy. However, a significant barrier to their adoption in production systems is their tendency to violate fundamental logical principles such as monotonicity. In this

Tianyue Li, Wenyu Gao, Boyan Fu, Tianhua Shao

Optical manipulation techniques offer exceptional contactless control but are fundamentally limited in their ability to perform parallel multitasking. To achieve high-density, versatile manipulation with subwavelength photonic devices, it is essential to sculpt light fields in multiple dimensions. Here, we overcome this challenge by introducing generalized o

Kento Yasuda, Kenta Ishimoto, Shigeyuki Komura

The interplay between information, dissipation, and control is reshaping our understanding of thermodynamics in feedback-regulated systems. We develop the informational Onsager-Machlup principle, a generalized variational framework that unifies energetic, dissipative, and informational contributions within a single formalism. This framework introduces a cond

Shijie Bao, Qi'an Guan, Xun Sun

In the present paper, we generalize the notion of the $p$-Bergman kernel and the $\xi$-Bergman kernel to the $p$-Bergman kernel with respect to a functional $\xi$, and establish some properties of the $p$-Bergman kernel with respect to $\xi$. We also study the relations between the $L^p$ versions of higher order Bergman kernels and $\xi$-Bergman kernels, and

Junichiro Niimi

Large language models (LLMs) have achieved remarkable results in wide range of domains. However, the accuracy and robustness of one-shot LLM predictions remain highly sensitive to the examples and the diversity among ensemble members. This study systematically investigates the effects of example representativeness (one-shot strategy) and output diversity (sa

Hiromichi Nakazato, Saverio Pascazio

We analytically derive the exact -- though formal -- master equation for a two-level quantum system (qubit) interacting with a bosonic environment within the rotating-wave approximation, assuming the environment is initially in an arbitrary thermal state. The long-time behavior of the evolution operator governing the dynamics of both the system and the envir

F. Thiesset, C. Federrath

This paper investigates the statistical properties of isothermal turbulence in both the subsonic and supersonic regimes. The focus is on the influence of the Mach number ($Ma$) and the Reynolds number ($Re$) on both the space-local and scale-dependent fluctuations of relevant gas variables, the density, velocity, their derivatives, and the kinetic energy. We

Crystal L. Martin, Weida Hu, Isak G. B. Wold, Andreas Faisst

Since the launch of JWST, the sample size of reionization-era Lyman-alpha-emitters (LAEs) has been steadily growing; yet inferences about the neutral hydrogen fraction in the intergalactic medium exhibit increasing variance at redshift z ~ 7, possibly indicating significant field-to-field fluctuations in the progression of cosmic reionization. In this paper,

Xiaoyu Yan, Tianxing Dai, Yu Marco Nie

A key challenge in transportation planning is that the collective preferences of heterogeneous travelers often diverge from the policies produced by model-driven decision tools. This misalignment frequently results in implementation delays or failures. Here, we investigate whether large language models (LLMs), noted for their capabilities in reasoning and si

Madhumati Pol, Anvay Anturkar, Anushka Khot, Ayush Andure

This study investigates the performance of 3D Convolutional Neural Networks (3D CNNs) and Long Short-Term Memory (LSTM) networks for real-time American Sign Language (ASL) recognition. Though 3D CNNs are good at spatiotemporal feature extraction from video sequences, LSTMs are optimized for modeling temporal dependencies in sequential data. We evaluate both

Tan Le, Van Le, Sachin Shetty

Indoor robotic systems within Cyber-Physical Systems (CPS) are increasingly exposed to Denial of Service (DoS) attacks that compromise localization, control and telemetry integrity. We propose a privacy-aware malware detection framework for indoor robotic systems, which leverages hybrid quantum computing and deep neural networks to counter DoS threats in CPS

Nils A. Nilsson

These proceedings summarise some recent efforts in understanding a class of vector-tensor theories known as {\it bumblebee} models, which spontaneously break local Lorentz and diffeomorphism invariance. Using cosmological perturbation theory on an FLRW background, we find that for non-minimal coupling to gravity, the theory contains a ghost mode unless degen

Jia Zhang, Bodong Du, Yitong Miao, Dongwei Sun

Oral mucosal diseases such as leukoplakia, oral lichen planus, and recurrent aphthous ulcers exhibit diverse and overlapping visual features, making diagnosis challenging for non-specialists. While vision-language models (VLMs) have shown promise in medical image interpretation, their application in oral healthcare remains underexplored due to the lack of la

Antonio Álvarez-López, Martín Hernández

We study dropout regularization in continuous-time models through the lens of random-batch methods -- a family of stochastic sampling schemes originally devised to reduce the computational cost of interacting particle systems. We construct an unbiased, well-posed estimator that mimics dropout by sampling neuron batches over time intervals of length $h$. Traj

Michael Döring

Protons and neutrons are the building blocks of matter, glued together in nuclei by strong interactions. They can be excited by pions, real and virtual photons, neutrinos and other probes. These excitations are referred to as light baryon resonances. A short, pedagogical overview of the field is presented including experimental progress, interpretation of li

Jieping Luo, Qiyue Li, Zhizhang Liu, Hang Qi

We study the client selection problem in Federated Learning (FL) within mobile edge computing (MEC) environments, particularly under the dependent multi-task settings, to reduce the total time required to complete various learning tasks. We propose CoDa-FL, a Cluster-oriented and Dependency-aware framework designed to reduce the total required time via clust

Rongjun Chen, Chengsi Yao, Jinchang Ren, Xianxian Zeng

Text-image alignment constitutes a foundational challenge in multimedia content understanding, where effective modeling of cross-modal semantic correspondences critically enhances retrieval system performance through joint embedding space optimization. Given the inherent difference in information entropy between texts and images, conventional approaches ofte

Ali Sarikhani, Steven M. Smith, Suzana Filipovic, William G. Fahrenholtz

The synthesis and characterization, along with the resulting properties, of fully dense \((\mathrm{Cr, Mo, Ta, V, W})\mathrm{C}\) high-entropy carbide ceramics were studied. The ceramics were synthesized from metal oxide and carbon powders by carbothermal reduction, followed by spark plasma sintering at various temperatures for densification. Increasing the

Romain Teyssier

In these lecture notes, we describe the current state-of-the-art for numerical simulations of large-scale structure and galaxy formation. Numerical simulations play a central role in the preparation and the exploitation of large-scale galaxy surveys, in which galaxies are the fundamental observational objects. We first describe basic methods for collisionles

Yujie Liu, Mingxuan Zhu, Shengyu Cheng, Dan Hao

Compilers are essential to software systems, and their bugs can propagate to dependent software. Ensuring compiler correctness is critical. However, isolating compiler bugs remains challenging due to the internal complexity of compiler execution. Existing techniques primarily mutate compilation inputs to generate passing and failing tests, but often lack cau

Cheuk Sau Au

This independent research investigates methods to improve the precision of cyclic peptide generation targeting the HIV gp120 trimer using AlphaFold. The study explores proximity-based hotspot mapping at the CD4 binding site, centroid distance penalization, generative loss tuning, and custom loss function development. These enhancements produced cyclic peptid

Sheng-Zhe Wang, Qian-Lan Cai, Zhi-Xin Meng, Yi-Cheng Deng

We report a compact single-cell source of a continuous cold-atom beam with three-dimensional (3D) cooling. By integrating an off-axis moving optical molasses (OM) with a two-dimensional magneto-optical trap (MOT), we achieve simultaneous 3D cooling within a 50 mm interaction region. The source delivers a continuous flux up to 4.9(5)x10^9 atoms/s, with a tran

John Alexander Cruz Morales

Starting from Greg Moore's description about Physical Mathematics, a framework is proposed in order to understand it, based on Gilles Ch\^atelet's philosophy. It will be argued that Ch\^atelet's ideas of inverting, splitting, augmenting and virtuality are crucial in the discussion about the nature of Physical Mathematics. Along this line, it will be proposed

Sina Jafari Ghalekohneh, Bo Zhao

Controlling photon mediated energy flow is central to the future of communications, thermal management, and energy harvesting technologies. Recent breakthroughs have revealed that many body systems violating Lorentz reciprocity can sustain persistent photon heat current at thermal equilibrium, hinting at a new paradigm of heat flow akin to superconductivity.