November 2025 arXiv papers — page 33

Rosana Caro, Lorena Cruz, Arturo Martinez, Pablo S. Naharro

Designing Zero-Emissions Buildings (ZEBs) involves balancing numerous complex objectives that traditional methods struggle to address. Fenestration, encompassing fa\c{c}ade openings and shading systems, plays a critical role in ZEB performance due to its high thermal transmittance and solar radiation admission. This paper presents a novel simulation-based op

T. Moustapha Mai, C. Azzaro-Pantel, M. Chin Choi, M. Hajajji

This study investigates the potential of hydrogen as a sustainable energy carrier for mobility applications in island territories, which are traditionally dependent on fossil fuel imports. Green hydrogen is identified as a key component of the energy transition. A Mixed Integer Linear Programming (MILP) model with a multi-period, multi-objective framework is

Gian Caspari, Manshu Khanna

We analyze the problem of matching asylum seekers to member states, incorporating wait times, preferences of asylum seekers, and the priorities, capacities, and burden-sharing commitments of member states. We identify a unique choice rule that addresses feasibility while balancing priorities and capacities. We examine the effects of both homogeneous and hete

Xinyue Guo, Xiaoran Yang, Lipan Zhang, Jianxuan Yang

Sound effect editing-modifying audio by adding, removing, or replacing elements-remains constrained by existing approaches that rely solely on low-level signal processing or coarse text prompts, often resulting in limited flexibility and suboptimal audio quality. To address this, we propose AV-Edit, a generative sound effect editing framework that enables fi

Jiaming He, Guanyu Hou, Hongwei Li, Zhicong Huang

Text-to-Video (T2V) models are capable of synthesizing high-quality, temporally coherent dynamic video content, but the diverse generation also inherently introduces critical safety challenges. Existing safety evaluation methods,which focus on static image and text generation, are insufficient to capture the complex temporal dynamics in video generation. To

Taejun Kim, Amy Karlson, Aakar Gupta, Tovi Grossman

While text entry is an essential and frequent task in Augmented Reality (AR) applications, devising an efficient and easy-to-use text entry method for AR remains an open challenge. This research presents STAR, a smartphone-analogous AR text entry technique that leverages a user's familiarity with smartphone two-thumb typing. With STAR, a user performs thumb

Zhongze Guo, Bei Xu, Qiang Gu

Zitterbewegung (ZBW), the trembling motion predicted by the Dirac equation, has long remained unobservable in free electrons due to its sub-Compton scale. We elaborately construct a relativistic vortex electron wave packet as a coherent superposition of both positive- and negative-energy Dirac states and derive their space-time dynamics. Our analysis demonst

Mannque Rho

How to arrive at the densest matter in massive compact stars starting from Walecka's linear $\omega$-$\sigma$ mean-field model is described in a series of arguments anchored on hidden local symmetry, hidden scale symmetry and emergent parity-doublet symmetry. I follow the bottom-up approach from chiral symmetry with pions, coupled to hidden local and scale s

Baoli Sun, Xinzhu Ma, Ning Wang, Zhihui Wang

Referring video object segmentation (RVOS) is an emerging cross-modality task that aims to generate pixel-level maps of the target objects referred by given textual expressions. The main concept involves learning an accurate alignment of visual elements and language expressions within a semantic space. Recent approaches address cross-modality alignment throu

Yuxuan Yang

Given a number field $F$ and $R$ be the ring of integers of $F$, the problem of embedding a field extension $K/F$ into a central simple algebra $B$ is classical. This paper proves that when the central simple algebra has degree $p$, the $R$-order $S\subset K$ can be optimal embedded into all maximal $R$-orders $O\subset B$, unless satisfies the optimal selec

Changlin Li, Jiawei Zhang, Shuhao Liu, Sihao Lin

Human video generation has advanced rapidly with the development of diffusion models, but the high computational cost and substantial memory consumption associated with training these models on high-resolution, multi-frame data pose significant challenges. In this paper, we propose Entropy-Guided Prioritized Progressive Learning (Ent-Prog), an efficient trai

Ziyi Chen, Yingnan Guo, Zedong Chu, Minghua Luo

Embodied navigation that adheres to social norms remains an open research challenge. Our SocialNav is a foundational model for socially-aware navigation with a hierarchical "brain-action" architecture, capable of understanding high-level social norms and generating low-level, socially compliant trajectories. To enable such dual capabilities, we construct the

Maxim Zabelkin, Sergey Drozdov, Oleg Skorikov, Sergey Pilipenko

We analyze the possibility of using observational data on nearby dwarf galaxies -- their luminosity functions and spatial distributions -- to constrain deviations of the cosmological power spectrum from the standard one. Specifically, we consider a cosmological model with a "bump" in the power spectrum at a wavelength of 1.3~Mpc and a dimensionless amplitude

Xi Zhang, Miguel Bernal, Wei-Ning Lee

Two-dimensional (2D) fully-addressed arrays can conveniently realize three-dimensional (3D) ultrasound imaging while fully controlled such arrays usually demands thousands of independent channels, which is costly. Sparse array technique using stochastic optimization methods is one of promising techniques to reduce channel counts while due to the stochastic n

Xintian Mao, Haofei Song, Yin-Nian Liu, Qingli Li

It is well-known that if a network aims to learn how to deblur, it should understand the blur process. Blurring is naturally caused by the convolution of the sharp image with the blur kernel. Thus, allowing the network to learn the blur process in the kernel-level can significantly improve the image deblurring performance. But, current deep networks are stil

Taejun Kim, Auejin Ham, Sunggeun Ahn, Geehyuk Lee

We present Lattice Menu, a gaze-based marking menu utilizing a lattice of visual anchors that helps perform accurate gaze pointing for menu item selection. Users who know the location of the desired item can leverage target-assisted gaze gestures for multilevel item selection by looking at visual anchors over the gaze trajectories. Our evaluation showed that

Pietro Carlo Boldini, Ryo Hirai, Benjamin Bugeat, Rene Pecnik

We present a perturbation-based framework that captures buoyancy effects on modal instabilities in stratified boundary-layer flows within the fully compressible, non-Oberbeck-Boussinesq formulation. Treating the Richardson number as a small parameter and recasting the stability problem into an adjoint-residual form, we derive a first-order correction for the

Dianbing Xi, Jiepeng Wang, Yuanzhi Liang, Xi Qiu

We tackle the dual challenges of video understanding and controllable video generation within a unified diffusion framework. Our key insights are two-fold: geometry-only cues (e.g., depth, edges) are insufficient: they specify layout but under-constrain appearance, materials, and illumination, limiting physically meaningful edits such as relighting or materi

Hyeongwoo Lee, Taeyoung Moon, Hyeonmin Oh, Kijeong Park

Atom-like defects in hexagonal boron nitride (hBN) provide room-temperature single-photon emission and coherent spin states, making them attractive for quantum-computing and -sensing applications. However, their random spatial and spectral characteristics hamper deterministic coupling with nano-optical cavities, limiting their use as bright single-photon sou

Huizheng Wang, Hongbin Wang, Shaojun Wei, Yang Hu

Attention-based Transformers have revolutionized natural language processing (NLP) and shown strong performance in computer vision (CV) tasks. However, as the input sequence varies, the computational bottlenecks in Transformer models exhibit dynamic behavior across stages, which calls for a cross-stage sparse acceleration strategy. Unfortunately, most existi

Xiaozhu Zhang, Nir Keret, Ali Shojaie, Armeen Taeb

We study the problem of learning a directed acyclic graph from data generated according to an additive, non-linear structural equation model with Gaussian noise. We express each non-linear function through a basis expansion, and derive a maximum likelihood estimator with a group l0-regularization that penalizes the number of edges in the graph. The resulting

A. I. Frank, V. A. Bushuev, M. A. Zakharov, G. V. Kulin

The problem of changing the energy of a neutron when it passes through an accelerating crystal under conditions close to the Bragg condition is considered. It is shown that, similar to the case of the passage of long-wavelength neutrons through a refractive sample, the accelerated motion of a crystal results in a change in neutron energy. The physical nature

Patrick Fournier, Fabrice Larribe

The ancestral recombination graph (ARG) is the model of choice in statistical genetics to model population ancestries. Software capable of simulating ARGs on a genome scale within a reasonable amount of time are now widely available for most practical use cases. While the inverse problem of inferring ancestries from a sample of haplotypes has seen major prog

Thomas Dedieu

We discuss, following Mikhalkin, Brugall\'e, and many others, the counting of curves on toric surfaces with prescribed genus, Newton polygon, and intersection pattern with the toric boundary divisor, both at assigned and unassigned points. The first lecture is dedicated to the proof of a correspondence theorem (for plane curves) with the counting of floor di

Changlin Li, Jiawei Zhang, Zeyi Shi, Zongxin Yang

Large-scale vision generative models, including diffusion and flow models, have demonstrated remarkable performance in visual generation tasks. However, transferring these pre-trained models to downstream tasks often results in significant parameter redundancy. In this paper, we propose EntPruner, an entropy-guided automatic progressive pruning framework for

Anup Roy, Rishabh Gyanendra Upadhyay, Animesh Rameshbhai Panara, Robin Mills

Document centric RAG pipelines usually begin with OCR, followed by brittle heuristics for chunking, table parsing, and layout reconstruction. These text first workflows are costly to maintain, sensitive to small layout shifts, and often lose the spatial cues that contain the answer. Vision first retrieval has emerged as a strong alternative. By operating dir

Mengran Li, Zelin Zang, Wenbin Xing, Junzhou Chen

Understanding how chemical perturbations propagate through biological systems is essential for robust molecular property prediction. While most existing methods focus on chemical structures alone, recent advances highlight the crucial role of cellular responses such as morphology and gene expression in shaping drug effects. However, current cell-aware approa

Riccardo Fantoni

We study a fluid of quantum hard-spheres treated with affine-quantization. Assuming that the fluid obeys to Bose-Einstein statistics we solve for its thermodynamic properties using the path integral Monte Carlo method.

Pius Onobhayedo, Paul Osemudiame Oamen

Artificial intelligence is retracing the Internet's path from centralized provision to distributed creation. Initially, resource-intensive computation concentrates within institutions capable of training and serving large models.Eventually, as federated learning matures, billions of edge devices holding sensitive data will be able to collectively improve mod

Sheng-nan Han, Jing Wu, Yong-rui Chen, Yi-zhen Huang

We investigate the influence of QCD phase transition and critical fluctuations of the critical end point (CEP) on the deuteron yield within the functional renormalization group (fRG) approach, by using the nucleon coalescence model and a low energy effective field theory of quarks and mesons. It is found that the two-point baryon density correlation function

Lachlan McGinness, Yutong Ma, Mohammad Attar, Andrew Carse

We present Particle Builder, an online board game which teaches students about concepts from the Standard Model of Particle Physics at a high school level. This short activity resulted in a gain of 0.16, indicating that students learned a significant amount of particle physics knowledge. Students found the activity was more engaging and less difficult than a

Lechen Feng, Haoran Li, Lucky Li, Xingqiu Zhao

This paper investigates the partial linear model by Least Absolute Deviation (LAD) regression. We parameterize the nonparametric term using Deep Neural Networks (DNNs) and formulate a penalized LAD problem for estimation. Specifically, our model exhibits the following challenges. First, the regularization term can be nonconvex and nonsmooth, necessitating th

Xin Honga, Jie Lin, Minghui Wang

Alzheimer's disease (AD), a degenerative brain condition, can benefit from early prediction to slow its progression. As the disease progresses, patients typically undergo brain atrophy. Current prediction methods for Alzheimers disease largely involve analyzing morphological changes in brain images through manual feature extraction. This paper proposes a nov

YuAn Wang, Xiaofan Li, Chi Huang, Wenhao Zhang

In controllable driving-scene reconstruction and 3D scene generation, maintaining geometric fidelity while synthesizing visually plausible appearance under large viewpoint shifts is crucial. However, effective fusion of geometry-based 3DGS and appearance-driven diffusion models faces inherent challenges, as the absence of pixel-wise, 3D-consistent editing cr

Swathi Shetty, Sayinath Udupa N. V., B. R. Rakshith

Let $G$ be graph with vertex set $V(G)$ and order $n$. A coalition in a graph $G$ consists of two disjoint sets of vertices $V_1$ and $V_2$, neither of which is a dominating set but whose union $V_1 \cup V_2$ is a dominating set. A coalition partition, abbreviated $c$-partition, in a graph $G$ is a vertex partition $\pi=\left\{V_1 , V_2,\dots, V_k\right\}$ s

Venkata Siddharth Dhara, Pawan Kumar

In recent times, denoising diffusion probabilistic models (DPMs) have proven effective for medical image generation and denoising, and as representation learners for downstream segmentation. However, segmentation performance is limited by the need for dense pixel-wise labels, which are expensive, time-consuming, and require expert knowledge. We propose FastT

Andrea Zacheo, Marco Marangi, Nilo Mata-Cervera, Yijie Shen

A major challenge in polariton fluids is achieving deterministic control over the spin texture of the macroscopic condensate wavefunction, which dictates the nucleation and dynamics of topological excitations, such as vortices, solitons, and strings. Existing approaches typically rely on external gauge fields to indirectly access the polariton pseudospin, re

A. Arziev, K. Kudaybergenov

The paper is devoted to characterizing convex trace ranges in finite atomic von Neumann algebras. The main result provides us with the necessary and sufficient condition for the range of a faithful normal trace on a finite atomic von Neumann algebra to be convex. In order to prove this result we will prove the following result, which has independent interest

Mudi Jiang, Jiahui Zhou, Xinying Liu, Zengyou He

Fair clustering has gained increasing attention in recent years, especially in applications involving socially sensitive attributes. However, existing fair clustering methods often lack interpretability, limiting their applicability in high-stakes scenarios where understanding the rationale behind clustering decisions is essential. In this work, we address t

Sourendu Gupta

We review the current state of knowledge of the phase diagram of QCD through lattice, effective field theories, and chiral models. Several sections through the three dimensional phase diagram are known for $N_f=2+1$ with good precision. Due to technical advances in lattice techniques over the last decade or so, new aspects of the phase diagram can now be exp

Tongkai Lu, Shuai Ma, Chongyang Tao

Mixed Integer Linear Programming (MILP) is a fundamental class of NP-hard problems that has garnered significant attention from both academia and industry. The Branch-and-Bound (B\&B) method is the dominant approach for solving MILPs and the branching plays an important role in B\&B methods. Neural-based learning frameworks have recently been developed to en

Ze Feng, Sen Yang, Boqiang Duan, Wankou Yang

Efficient Multimodal Large Language Models (MLLMs) compress vision tokens to reduce resource consumption, but the loss of visual information can degrade comprehension capabilities. Although some priors introduce Knowledge Distillation to enhance student models, they overlook the fundamental differences in fine-grained vision comprehension caused by unbalance

Pushkal Mishra, Kshitiz Bansal, Dinesh Bharadia

Radar sensors provide reliable perception across adverse weather, lighting, and long-range conditions, yet existing machine learning approaches remain fragmented and task-specific, with each downstream task employing distinct architectures and training objectives. We present RadarVLM, a vision-language framework that learns unified scene-level representation

Hengyu Fu, Baihe Huang, Virginia Adams, Charles Wang

Diffusion Language Models (DLMs) have recently emerged as a strong alternative to autoregressive language models (LMs). DLMs offer comparable accuracy with faster inference speed via parallel decoding. However, standard DLM decoding strategies relying on high-confidence tokens encounter an inherent information-theoretic bottleneck that restricts decoding pro

Narinder Kumar Wadhawan, Priyanka Wadhawan

We have proved in this paper that natural logarithm of consecutive number ratio, x/(x-1) approximates to 2/(2x - 1) where x is a real number except 1. Using this relation, we, then proved, x approximates to double the sum of odd harmonic series having first and last terms 1/3 and 1/(2x - 1) respectively. Thereafter, not limiting to consecutive number ratios,

Manish Jain, Satheesh Kumar Ponnambalam, Salman Faroz, Chandrakanth Lns

Large Language Models (LLMs) demonstrate exceptional capabilities across general domains, yet their application to specialized sectors such as mortgage finance requires domain-specific knowledge augmentation while preserving instruction-following fidelity. We present MortgageLLM, a novel domain-specific large language model that addresses this dual challenge

Feifan Han, Xiongbin Yu, Qun Zhang, Zebin Huang

Terahertz (THz) frequency combs (TFCs) are promising for numerous applications in spectroscopy, metrology, sensing, and wireless communications. However, the practical applications of TFCs have been hindered thus far by the need for cryogenic cooling, limited bandwidth, and bulky configuration, largely due to the lack of advanced THz sources. Here, we report

Chandra Chekuri, Pooja Kulkarni, Ruta Mehta, Jan Vondrak

We consider discrete allocation problem where $m$ indivisible goods are to be divided among $n$ agents. When agents' valuations are additive, the well-known cycle cancelling lemma by Lenstra, Shmoys, and Tardos plays a key role in design and analysis of rounding algorithms. In this paper, we prove an analogous lemma for the case of submodular valuations. Our

Gayoung Lee, Junho Kim, Jin-Hwa Kim, Junmo Kim

Understanding reflection remains a long-standing challenge in 3D reconstruction due to the entanglement of appearance and geometry under view-dependent reflections. In this work, we present the Pygmalion Effect in Vision, a novel framework that metaphorically "sculpts" reflective objects into clay-like forms through image-to-clay translation. Inspired by the

Geetanjali Sharma, Gaurav Jaswal, Aditya Nigam, Raghavendra Ramachandra

Iris authentication algorithms have achieved impressive recognition performance, making them highly promising for real-world applications such as border control, citizen identification, and both criminal investigations and commercial systems. However, their robustness is still challenged by variations in rotation, scale, specular reflections, and defocus blu

Caihong Yang, Jiasheng Zeng, Xiao-Dong Zhang

Recently, Alon and Frankl (JCTB, 2024) determined the maximum number of edges in $K_{\ell+1}$-free $n$-vertex graphs with bounded matching number. For integers $\ell\ge r \ge 2$, the family $\mathcal{K}_{\ell+1}^{r}$ consists of all $r$-graphs $F$ with at most $\binom{\ell+1}{2}$ edges such that, for some $(\ell+1)$-set $K$, every pair $\{x,y\} \subseteq K$

M. A. H. B. Md Yusoff, H. B. Ambalampitiya, J. M. Ngoko Djiokap

Attosecond streaking camera is an ex situ technique in which a linearly polarized (LP) XUV attopulse produces an electron wavepacket by photoionization in the presence of an IR femtopulse. By moving the two synchronous oppositely circularly polarized XUV pulses (that make the ionizing LP pulse) apart in time, we propose an attosecond double-slit streak camer

Jinwoo Jeong, Minchul Kang, Younghun Go, Changyong Shin

As deep learning models in agentic AI systems grow in scale and complexity, GPU memory requirements increase and often exceed the available GPU memory capacity, so that out-of-memory (OoM) errors occur. It is well known that OoM interrupts the whole training itself and wastes substantial computational resources. Therefore, to prevent OoM, accurate prediction

Jaewoo Kim, Yeonwoo Nam, Chung-Kil Hur

Contemporary proof assistants impose restrictive syntactic guardedness conditions that reject many valid corecursive definitions. Existing approaches to overcome these restrictions present a fundamental trade-off between coverage and automation. We present Compositional Heterogeneous Productivity (CHP), a theoretical framework that unifies high automation wi

Seunghun Baek, Jaejin Lee, Jaeyoon Sim, Minjae Jeong

Various neuroimaging studies suffer from small sample size problem which often limit their reliability. Meta-analysis addresses this challenge by aggregating findings from different studies to identify consistent patterns of brain activity. However, traditional approaches based on keyword retrieval or linear mappings often overlook the rich hierarchical stru

Wuyang Zhang, Lejun Xu

The United Nations' Sustainable Development Goals (SDGs) represent a complex, interdependent framework where progress in one area can synergistically promote or competitively inhibit progress in others. For policymakers in international development, a critical challenge is identifying "leverage points" - specific goals where limited resource allocation yield

Qin Wu, Ralph-C Bayer

Contest participants often have strong incentives to engage in cheating. Sanctions serve as a common deterrent against such conduct. Often, other agents on the contestant's team (e.g., a coach of an athlete) or a company (a manager of an R\&D engineer) have a vested interest in outcomes and can influence the cheating decision. An agency problem arises when o

Ivan Novikov

Large Language Models (LLMs) are predominantly deployed as dense transformers, where every parameter in every feed-forward block is activated for every token. While architecturally simple, this is computationally inefficient, since inference costs scale linearly with parameter count. Recent upcycling methods such as MoEfication, CMoE, ToMoE, and MoORE reveal

Ye Bhone Lin, Thura Aung, Ye Kyaw Thu, Thazin Myint Oo

This paper investigates sequence-to-sequence Transformer models for automatic speech recognition (ASR) error correction in low-resource Burmese, focusing on different feature integration strategies including IPA and alignment information. To our knowledge, this is the first study addressing ASR error correction specifically for Burmese. We evaluate five ASR

Ziyun Zeng, Hang Hua, Jiebo Luo

Instruction-guided image editing offers an intuitive way for users to edit images with natural language. However, diffusion-based editing models often struggle to accurately interpret complex user instructions, especially those involving compositional relationships, contextual cues, or referring expressions, leading to edits that drift semantically or fail t

Bryan E. Tuck, Rakesh M. Verma

Large language models must satisfy hard orthographic constraints during controlled text generation, yet systematic cross-family evaluation remains limited. We evaluate 39 configurations spanning three model families (Qwen3, Claude Haiku 4.5, GPT-5-mini) on 58 word puzzles requiring character-level constraint satisfaction. Cross-family differences produce sub

Bo-Lun Huang, Zhen-Zhao Tao, Tong-Jie Zhang, Vishal Gajjar

We report a narrowband technosignature search toward five Milky Way globular clusters (NGC 6171, NGC 6218, NGC 6254, NGC 6838, and IC 1276) using the Five-hundred-meter Aperture Spherical radio Telescope (FAST) 19-beam L-band receiver (1.05-1.45 GHz). We adapt the MultiBeam Point-source Scanning (MBPS) strategy to extended targets by gating detections to gen

Ahmadreza Majlesara, Ali Majlesi, Ali Mamaghani, Alireza Shokrani

This demonstration showcases the integration of a lightweight, locally deployed Large Language Model (LLaMA-3 8b Q-4b) empowered by retrieval augmented generation (RAG) to automate 5G network management, with a strong emphasis on privacy. By running the LLM on local or edge devices ,we eliminate the need for external APIs, ensuring that sensitive data remain

Feiyang Pan, Shenghe Zheng, Chunyan Yin, Guangbin Dou

Visual-Inertial Odometry (VIO) is a critical component for robust ego-motion estimation, enabling foundational capabilities such as autonomous navigation in robotics and real-time 6-DoF tracking for augmented reality. Existing methods face a well-known trade-off: filter-based approaches are efficient but prone to drift, while optimization-based methods, thou

James V. Raj, Xiaohan Sun, Charles Emmett Maher, Katherine A. Newhall

Hyperuniformity, which is a type of long-range order that is characterized by the suppression of long-range density fluctuations in comparison to the fluctuations in standard disordered systems, has emerged as a powerful concept to aid in the understanding of diverse natural and engineered phenomena. In the present paper, we harness hyperuniform point patter

Thura Aung, Eaint Kay Khaing Kyaw, Ye Kyaw Thu, Thazin Myint Oo

In low-resource languages like Burmese, classification tasks often fine-tune only the final classification layer, keeping pre-trained encoder weights frozen. While Multi-Layer Perceptrons (MLPs) are commonly used, their fixed non-linearity can limit expressiveness and increase computational cost. This work explores Kolmogorov-Arnold Networks (KANs) as altern

Yeswanth Ravichandran, Duoduo Liao, Charan Teja Kurakula

Subsurface defects such as delamination, voids, and honeycombing critically affect the durability of concrete bridge decks but are difficult to detect reliably using visual inspection or manual sounding. This paper presents a machine learning based Impact Echo (IE) framework that automates both defect localization and multi-class classification of common con

Qianyun Wu, Yukie Sano, Hideki Takayasu, Misako Takayasu

Social media discourse on COVID-19 vaccination provides a valuable context for studying opinion formation, emotional expression, and social influence during a global crisis. While prior studies have examined emotional strategies within communities and the link between emotions and vaccine hesitancy, few have investigated dynamic emotion changes across collec

Yusuke Koide, Takato Ishida, Takashi Uneyama, Yuichi Masubuchi

We investigate extensional viscosity and polymer conformation in dilute polymer solutions under uniaxial extensional flow using dissipative particle dynamics simulations. At high extension rates, polymers are significantly stretched by extensional flows, and the extensional viscosity growth function exhibits strain hardening. To reveal their quantitative rel

Aaditya L. Kachhadiya

Inverse problems in the physical sciences are often ill-conditioned in input space, making progress step-size sensitive. We propose the Deceptron, a lightweight bidirectional module that learns a local inverse of a differentiable forward surrogate. Training combines a supervised fit, forward-reverse consistency, a lightweight spectral penalty, a soft bias ti

Zhenchao Tang, Fang Wang, Haohuai He, Jiale Zhou

Engineering LLMs to accelerate life sciences research requires a robust alignment with biomedical knowledge. We observe that biomedical text exhibits a fundamentally different uncertainty structure from general text: dense low-confidence runs encode epistemic knowledge gaps (dense causal chains, rare entities) rather than the sparse aleatoric stylistic varia

Hongrui Chen, Rong Ma

The rapid growth of high-dimensional datasets across a wide range of scientific domains has created an urgent need for new statistical methods to compare distributions with underlying low-dimensional structure. Assessing similarity between high-dimensional datasets whose observations concentrate near low-dimensional manifolds is particularly challenging due

Runqi Kang, Xing Rong

Exotic interactions between fermions mediated by new bosons beyond the Standard Model may hold the key to several fundamental conundrums on the frontier of physics.However, laboratory searching for exotic interactions on small length scales is fundamentally held back by the short force range.Here we propose that the force range of exotic interactions tends t

Samia Heshmat, Satoshi Tomioka, Naoki Miyamoto, Yuji Yamauchi

Wavefront reconstruction in lateral shearing interferometry typically assumes that the shear amount is an integer multiple of the sampling interval. When the shear is fractional, approximating it with the nearest integer value leads to noticeable reconstruction errors. To address this, we propose a weighted integer shear averaging method. The approach combin

Yu He, Zhihao Zhou, Lina Yang, Nuo Yang

Understanding nanoscale hotspot thermal transport is crucial in electronic devices. Contrary to common perception, recent experiments show that closely spaced nanoscale multiple hotspots can enhance heat dissipation. Here, the thermal transport in nanoscale multiple hotspot systems is investigated by solving the phonon Boltzmann transport equation. The local

Andrius Burnelis, Daniel R. Phillips

We extend previous halo effective field theory analyses of low-energy elastic scattering of $^{3}$He-$^{4}$He, including the $\frac{7}{2}^{-}$ $f$-wave resonance as an explicit degree of freedom. The presence of this resonance necessitates a changing power counting scheme depending on the kinematic region. Therefore, we construct a theory uncertainty model a

Anirshu Devroy, Gregor Fritz, Mathias Brandstoetter

Current rehabilitation of lower limb prostheses has significant challenges, especially with skin conditions, irritation and discomfort. Understanding the skin temperature and having comfortable wearable sensors that would monitor skin temperature in a real-time outdoor environment would be useful. The system would help the user and orthopedic technician to p

Aritra Banik, Praneet Kumar Patra, Adele Anna Rescigno, Abhishek Sahu

The Identifying Code (IC) problem seeks a vertex subset whose intersection with every vertex's closed neighborhood is unique, enabling fault detection in multiprocessor systems and practical uses in identity verification, environmental monitoring, and dynamic localization. A closely related problem is the Locating-Dominating Set (LD), which requires each non

Sanjar M. Abrarov, Rehan Siddiqui, Rajinder Kumar Jagpal, Brendan M. Quine

In this work, we prove the irrationality of $π$ motivated by the nested radicals with roots of $2$ of kind $c_k = \sqrt{2 + c_{k - 1}}$ and $c_0 = 0$. Sample computations showing how the rational approximation tends to $π$ with increasing the integer $k$ are presented.

Michael Iskandardinata, William Christian, Derwin Suhartono

Detecting sarcasm remains a challenging task in the areas of Natural Language Processing (NLP) despite recent advances in neural network approaches. Currently, Pre-trained Language Models (PLMs) and Large Language Models (LLMs) are the preferred approach for sarcasm detection. However, the complexity of sarcastic text, combined with linguistic diversity and

Daniella Catalá, Miriam Vollmayr-Lee, Alejandro Bravo-Doddoli

This paper investigates sub-Riemannian geodesics within the jet space of curves. We establish the existence of two distinct families of metric lines, that is, globally minimizing geodesics, in the $2$-jet space of plane curves. This result provides an initial contribution toward the broader classification of metric lines in jet spaces. Additionally, we prese

Mingcheng Zhu, Mingxuan Liu, Han Yuan, Yilin Ning

Artificial intelligence (AI) has shown remarkable promise in dermatology, offering accurate and non-invasive diagnosis of skin cancer. While extensive research has addressed skin tone-related bias, gender bias in dermatologic AI remains underexplored, leading to unequal care and reinforcing existing gender disparities. In this study, we developed LesionAttn,

Chujie Wang, Jianyu Lu, Zhiyuan Luo, Xi Chen

Open-Vocabulary Object Detection (OVOD) aims to enable detectors to generalize across categories by leveraging semantic information. Although existing methods are pretrained on large vision-language datasets, their inference is still limited to fixed category names, creating a gap between multimodal training and unimodal inference. Previous work has shown th

Alireza Aghasi, Nicholas Marshall, Saeid Pourmand, Wyatt Whiting

We propose a novel randomized algorithm for constructing binary neural networks with tunable accuracy. This approach is motivated by hyperdimensional computing (HDC), which is a brain-inspired paradigm that leverages high-dimensional vector representations, offering efficient hardware implementation and robustness to model corruptions. Unlike traditional low

Yin Zhang, Baolin Tan, Quan Wang, Jing Huang

We present the observational evidence of the existence of a double-decker filament channel (FC) by using observations in extreme ultraviolet and H{\alpha} wavelengths. For both FCs, the east foot-point roots in the active region (AR), while the west one roots in the remote quiet region. The bottom FC (FC1) appears as intermittent filaments. Within the AR, th

Chi-Hsiang Hsiao, Yi-Cheng Wang, Tzung-Sheng Lin, Yi-Ren Yeh

Retrieval-augmented generation (RAG) enables large language models (LLMs) to dynamically access external information, which is powerful for answering questions over previously unseen documents. Nonetheless, they struggle with high-level conceptual understanding and holistic comprehension due to limited context windows, which constrain their ability to perfor

Vladimir Berman

Zipf's law in language lacks a definitive origin, debated across fields. This study explains Zipf-like behavior using geometric mechanisms without linguistic elements. The Full Combinatorial Word Model (FCWM) forms words from a finite alphabet, generating a geometric distribution of word lengths. Interacting exponential forces yield a power-law rank-frequenc

Pete Rigas

We characterize optimality of Quantum strategies for the Odd-Cycle game. Separate from other game-theoretic settings, parallel repetition for the Odd-Cycle game is related to the foam problem, which can be formulated through a minimization of the surface area. In comparison to previous works on minimizing the surface area, we quantify how properties of the m

Lintao Liu, Nariya Uchida

Chimera states, characterized by the coexistence of coherent and incoherent domains, represent a paradigm of self-organization in complex systems. In this study, we introduce a topological analysis method based on winding numbers to characterize the dynamics of spiral wave chimeras in a two-dimensional phase oscillator network. Our investigation reveals dist

Xin Hong, Xinze Sun, Yinhao Li, Yen-Wei Chen

Image generation can provide physicians with an imaging diagnosis basis in the prediction of Alzheimer's Disease (AD). Recent research has shown that long-term AD predictions by image generation often face difficulties maintaining disease-related characteristics when dealing with irregular time intervals in sequential data. Considering that the time-related

Spiro Karigiannis, Sébastien Picard, Caleb Suan

We study flows of $G_2$-structures guided by the principle of dimensional reduction: natural geometric flows in $G_2$-geometry reduce to natural flows in complex geometry. Our main examples are the $G_2$-Laplacian coflow, which lifts the K\"ahler--Ricci flow, and a 7-dimensional lift of the anomaly flow on complex threefolds. The $G_2$-lift of the anomaly fl

Jiaming Guo, Rui Zhang, Zerun Li, Yunkai Gao

Diffusion planning is a promising method for learning high-performance policies from offline data. To avoid the impact of discrepancies between planning and reality on performance, previous works generate new plans at each time step. However, this incurs significant computational overhead and leads to lower decision frequencies, and frequent plan switching m

Chenglizhao Chen, Shaofeng Liang, Runwei Guan, Xiaolou Sun

Referring Multi-Object Tracking (RMOT) aims to achieve precise object detection and tracking through natural language instructions, representing a fundamental capability for intelligent robotic systems. However, current RMOT research remains mostly confined to ground-level scenarios, which constrains their ability to capture broad-scale scene contexts and pe

P. Djorwé, R. Altuijri, A. J. Almalki, S. Abdel-Khalek

We propose a scheme to generate multi-paths entanglement in an optomechanical system by exploiting polarized electromagnetic fields and dark mode control. Our system consists of two mechanically coupled mechanical resonators, which are driven by a common electromagnetic field. An inclusion of a polarizer induces linear polarizations of the electromgnetic fie

Yingjie Xia, Xi Wang, Jinglei Shi, Vicky Kalogeiton

Images evoke emotions that profoundly influence perception, often prioritized over content. Current Image Emotional Synthesis (IES) approaches artificially separate generation and editing tasks, creating inefficiencies and limiting applications where these tasks naturally intertwine, such as therapeutic interventions or storytelling. In this work, we introdu

Dongkyu Derek Cho, Huan Song, Arijit Ghosh Chowdhury, Haotian An

Fine-tuning large language models (LLMs) for downstream tasks typically exhibit a fundamental safety-capability tradeoff, where improving task performance degrades safety alignment even on benign datasets. This degradation persists across standard approaches including supervised finetuning (SFT) and reinforcement learning from human feedback (RLHF). While re

Jesús Arturo Sánchez-Sánchez, Thomas Stegmann

We investigate electronic transport in gapped bilayer graphene (gBLG) devices. For certain edge terminations -typically a combination of zigzag, armchair, and bearded types - we observe edge state conduction within the band gap, which is opened by a potential bias between the two layers. The edge states can generate a non-local resistance, in line with recen

Jingtao Guo, Yuyi Mao, Ivan Wang-Hei Ho

Wi-Fi channel state information (CSI)-based sensing provides a non-invasive, device-free approach for tasks such as human activity recognition and crowd counting, but large-scale deployment is hindered by the need for extensive site-specific training data. Federated learning (FL) offers a way to avoid raw data sharing but is challenged by heterogeneous sensi

Changjian Xie

High-fidelity numerical simulation serves as a cornerstone for exploring magnetization dynamics in micromagnetics. This work introduces a novel third-order temporally accurate and stable numerical scheme for the Landau-Lifshitz-Gilbert (LLG) equation, aiming to address the limitations in accuracy and efficiency often encountered with conventional approaches.

Zhongchun Zhou, Chengtao Lai, Wei Zhang

Large Language Models (LLMs) have achieved unprecedented success across various applications, but their substantial memory requirements pose significant challenges to current memory system designs, especially during inference. Our work targets last-level cache (LLC) based architectures, including GPUs (e.g., NVIDIA GPUs) and AI accelerators. We introduce LLa