May 2025 arXiv papers — page 23

Miaomiao Wei, Longjun Xiang, Fuming Xu, Bin Wang

A normal metallic system proximitized by a superconductor can exhibit a gapless superconducting state characterized by segmented Fermi surfaces, as confirmed experimentally. In such a state, quasiparticle states remain gapless along one direction, while a superconducting gap opens in the perpendicular direction. This anisotropy enables a novel Hall effect in

Maged S. Al-Shaibani, Moataz Ahmed

Large Language Models (LLMs) have achieved unprecedented capabilities in generating human-like text, posing subtle yet significant challenges for information integrity across critical domains, including education, social media, and academia, enabling sophisticated misinformation campaigns, compromising healthcare guidance, and facilitating targeted propagand

Guangyuan Liu, Yinqiu Liu, Ruichen Zhang, Hongyang Du

The rapid development of multimodal AI and Large Language Models (LLMs) has greatly enhanced real-time interaction, decision-making, and collaborative tasks. However, in wireless multi-agent scenarios, limited bandwidth poses significant challenges to exchanging semantically rich multimodal information efficiently. Traditional semantic communication methods,

Huachao Zhang, Chang-An Zhao

For a Kummer extension defined by the affine equation $y^{m}=\prod_{i=1}^{r} (x-\a_i)^{\lambda_i}$ over an algebraic extension $K$ of a finite field $\fq$, where $\la_i\in \Z\backslash\{0\}$ for $1\leq i\leq r$, $\gcd(m,q) = 1$, and $\a_1,\cdots,\a_r\in K$ are pairwise distinct elements, we propose a simple and efficient method to find all pure gaps at many

Jun Fan, Ailing Yan, Xianchao Xiu, Wanquan Liu

Phase retrieval (PR) is a popular research topic in signal processing and machine learning. However, its performance degrades significantly when the measurements are corrupted by noise or outliers. To address this limitation, we propose a novel robust sparse PR method that covers both real- and complex-valued cases. The core is to leverage the Huber function

Yazhou Zhang, Chunwang Zou, Qimeng Liu, Lu Rong

Can multi-modal large models (MLMs) that can ``see'' an image be said to ``understand'' it? Drawing inspiration from Searle's Chinese Room, we propose the \textbf{Visual Room} argument: a system may process and describe every detail of visual inputs by following algorithmic rules, without genuinely comprehending the underlying intention. This dilemma challen

Mao-Lin Luo, Zi-Hao Zhou, Tong Wei, Min-Ling Zhang

Continual learning with vision-language models like CLIP offers a pathway toward scalable machine learning systems by leveraging its transferable representations. Existing CLIP-based methods adapt the pre-trained image encoder by adding multiple sets of learnable parameters, with each task using a partial set of parameters. This requires selecting the expect

Haokun Chen, Yueqi Zhang, Yuan Bi, Yao Zhang

In recent years, Large Language Models (LLMs) have achieved remarkable advancements, drawing significant attention from the research community. Their capabilities are largely attributed to large-scale architectures, which require extensive training on massive datasets. However, such datasets often contain sensitive or copyrighted content sourced from the pub

Wei Zhuo, Zhaohuan Zhan, Han Yu

Federated Learning (FL) on graph-structured data typically faces non-IID challenges, particularly in scenarios where each client holds a distinct subgraph sampled from a global graph. In this paper, we introduce Federated learning with Auxiliary projections (FedAux), a personalized subgraph FL framework that learns to align, compare, and aggregate heterogene

Andrew Ng

We observe a criterion for groups to have vanishing virtual first Betti number and use it to give infinitely many examples of torsion-free, finitely generated, residually finite groups which aren't virtually diffuse. This answers a question raised by Kionke and Raimbault.

Spyros Barbakos, Charalampos Antoniadis, Gerasimos Potamianos, Gianluca Setti

Video consumption is a key part of daily life, but watching entire videos can be tedious. To address this, researchers have explored video summarization and highlight detection to identify key video segments. While some works combine video frames and transcripts, and others tackle video summarization and highlight detection using Reinforcement Learning (RL),

Jianlin Ye, Savvas Papaioannou, Panayiotis Kolios

Path planning is a fundamental capability of autonomous Unmanned Aerial Vehicles (UAVs), enabling them to efficiently navigate toward a target region or explore complex environments while avoiding obstacles. Traditional pathplanning methods, such as Rapidly-exploring Random Trees (RRT), have proven effective but often encounter significant challenges. These

Chunlong Xie, Jialing He, Shangwei Guo, Jiacheng Wang

We present Adversarial Object Fusion (AdvOF), a novel attack framework targeting vision-and-language navigation (VLN) agents in service-oriented environments by generating adversarial 3D objects. While foundational models like Large Language Models (LLMs) and Vision Language Models (VLMs) have enhanced service-oriented navigation systems through improved per

Zhiyuan Hu, Zheng Sun, Yi Wei, Long Yu

The performance of image generation has been significantly improved in recent years. However, the study of image screening is rare, and its performance with Multimodal Large Language Models (MLLMs) is unsatisfactory due to the lack of data and the weak spatial plausibility reasoning ability in MLLMs. In this work, we propose a complete solution to address th

Fangyikang Wang, Hubery Yin, Shaobin Zhuang, Huminhao Zhu

Recent Diffusion models (DMs) advancements have explored incorporating the second-order diffusion Fisher information (DF), defined as the negative Hessian of log density, into various downstream tasks and theoretical analysis. However, current practices typically approximate the diffusion Fisher by applying auto-differentiation to the learned score network.

Paolo Leonetti, Cihan Orhan

Given an ideal $\mathcal{I}$ on $\omega$, we denote by $\mathrm{SL}(\mathcal{I})$ the family of positive normalized linear functionals on $\ell_\infty$ which assign value $0$ to all characteristic sequences of sets in $\mathcal{I}$. We show that every element of $\mathrm{SL}(\mathcal{I})$ is a Choquet average of certain ultrafilter limit functionals. Also, w

Pengfei Xu, Donggen Wang

As a specific domain of subjective well-being, travel satisfaction has recently attracted much research attention. Previous studies primarily relied on statistical models and, more recently, machine learning models to explore its determinants. Both approaches,however, depend on sufficiently large sample sizes and appropriate statistical assumptions. The emer

Carlo Albert, Simone Ulzega, Simon Dirmeier, Andreas Scheidegger

Bayesian inference with stochastic models is often difficult because their likelihood functions involve high-dimensional integrals. Approximate Bayesian Computation (ABC) avoids evaluating the likelihood function and instead infers model parameters by comparing model simulations with observations using a few carefully chosen summary statistics and a toleranc

Budhaditya Halder, Shubhayan Pan, Koulik Khamaru

We consider the problem of statistical inference when the data is collected via a Thompson Sampling-type algorithm. While Thompson Sampling (TS) is known to be both asymptotically optimal and empirically effective, its adaptive sampling scheme poses challenges for constructing confidence intervals for model parameters. We propose and analyze a variant of TS,

An Vu, Jonas Oppenlaender

The rise of large language models (LLMs) has created a new job role: the Prompt Engineer. Despite growing interest in this position, we still do not fully understand what skills this new job role requires or how common these jobs are. In this paper, we present a data-driven analysis of global prompt engineering job trends on LinkedIn. We take a snapshot of t

Ziad Qais Al-Abbasi

It is envisioned that the next generations of wireless communication environment will be characterized with dense traffic demand due to the prediction that there will be large numbers of active users. Hence, it is important to find a solution to deal with such dense numbers of users. This paper investigates optimizing the connectivity and users scheduling to

Yiwei Sun

Educators teaching entry-level university engineering modules face the challenge of identifying which topics students find most difficult and how to support diverse student needs effectively. This study demonstrates a rigorous yet interpretable statistical approach -- hierarchical Bayesian modeling -- that leverages detailed student response data to quantify

Haojie Jia, Zhenhao Li, Gen Li, Minxian Xu

As securities trading systems transition to a microservices architecture, optimizing system performance presents challenges such as inefficient resource scheduling and high service response delays. Existing container orchestration platforms lack tailored performance optimization mechanisms for trading scenarios, making it difficult to meet the stringent 50ms

Priyotosh Bandyopadhyay, Snehashis Parashar

A real scalar triplet with zero hypercharge offers a minimal non-trivial extension of the Standard Model (SM) with a charged Higgs and a possible dark matter or custodial symmetry breaking signature. The $Z_2$-odd inert triplet model (ITM) provides a dark matter, while the non-inert Higgs triplet model (HTM) breaks the custodial symmetry, enabling rich colli

Yanxia Liu, Shu Chen

We unveil the mechanism for the formation of puzzled boundary-localized bound states in a spinless fermionic open lattice with nearest-neighbor interactions. By solving the Bethe-ansatz equation analytically, we uncover asymmetrical string solutions corresponding to the boundary-localized bound states, which emerge in systems with at least three particles. T

Yong-Cheng Liaw, Shuo-Han Chen

Owing to the huge success of generative artificial intelligence (AI), large language models (LLMs) have emerged as a core subclass, underpinning applications such as question answering, text generation, and code completion. While fine-tuning these models on domain-specific data can yield significant performance gains, it also poses daunting computational cha

Yixun Liang, Kunming Luo, Xiao Chen, Rui Chen

We present UniTEX, a novel two-stage 3D texture generation framework to create high-quality, consistent textures for 3D assets. Existing approaches predominantly rely on UV-based inpainting to refine textures after reprojecting the generated multi-view images onto the 3D shapes, which introduces challenges related to topological ambiguity. To address this, w

Bing Ma, Hai Zhuge

Approaches form the foundation for conducting scientific research. Querying approaches from a vast body of scientific papers is extremely time-consuming, and without a well-organized management framework, researchers may face significant challenges in querying and utilizing relevant approaches. Constructing multiple dimensions on approaches and managing them

Chang Liu, Bohao Zhao, Jingtao Ding, Huandong Wang

Understanding chaotic dynamics is a fundamental problem across scientific disciplines, including climate science, neuroscience, and fluid dynamics, yet direct experimentation and intervention in such systems are often infeasible. Chaotic system reconstruction aims to identify a surrogate dynamical model that preserves a system's invariant geometric and l

Yuanyuan Wang, Tianze Wei

We study the fair allocation problem of indivisible items with subsidy. In this paper, we focus on the notion of fairness - equitability (EQ), which requires that items be allocated such that all agents value the bundle they receive equally. First, we study the upper bounds of the minimum required subsidy to achieve EQ in different item settings and provide

Pascal J. Sager, Ashwini Kamaraj, Benjamin F. Grewe, Thilo Stadelmann

We present the methodology and results of the Deep Retrieval team for subtask 4b of the CLEF CheckThat! 2025 competition, which focuses on retrieving relevant scientific literature for given social media posts. To address this task, we propose a hybrid retrieval pipeline that combines lexical precision, semantic generalization, and deep contextual re-ranking

Guangyuan Liu, Yinqiu Liu, Jiacheng Wang, Hongyang Du

In next-generation wireless networks, supporting real-time applications such as augmented reality, autonomous driving, and immersive Metaverse services demands stringent constraints on bandwidth, latency, and reliability. Existing semantic communication (SemCom) approaches typically rely on static models, overlooking dynamic conditions and contextual cues vi

Yunliang Qi, Meng Lou, Yimin Liu, Lu Li

Remote sensing image super-resolution (RSISR) is a crucial task in remote sensing image processing, aiming to reconstruct high-resolution (HR) images from their low-resolution (LR) counterparts. Despite the growing number of RSISR methods proposed in recent years, a systematic and comprehensive review of these methods is still lacking. This paper presents a

Zonglin Yang, Zhexuan Gu, Houduo Qi, Yancheng Yuan

Reinforcement learning from human feedback (RLHF) is an essential technique for ensuring that large language models (LLMs) are aligned with human values and preferences during the post-training phase. As an effective RLHF approach, group relative policy optimization (GRPO) has demonstrated success in many LLM-based applications. However, efficient GRPO-based

Honoka Anada, Tatsuya Kaneko, Shinya Takamaeda-Yamazaki

Federated learning (FL) enables multiple clients to collaboratively train machine learning models without sharing local data. In particular, decentralized FL (DFL), where clients exchange models without a central server, has gained attention for mitigating communication bottlenecks. Evaluating participant contributions is crucial in DFL to incentivize active

Martin Vohralík, Soleiman Yousef

A posteriori estimates give bounds on the error between the unknown solution of a partial differential equation and its numerical approximation. We present here the methodology based on H1-conforming potential and H(div)-conforming equilibrated flux reconstructions, where the error bounds are guaranteed and fully computable. We consider any lowest-order loca

Emo Todorov

Policy gradients in continuous control have been derived for both stochastic and deterministic policies. Here we study the relationship between the two. In a widely-used family of MDPs involving Gaussian control noise and quadratic control costs, we show that the stochastic and deterministic policy gradients, natural gradients, and state value functions are

C. Pallis

We show that the coexistence of a non-minimal coupling to gravity $f_{\mathcal{R}}=1+c{\mathcal{R}} \phi^{n/2}$ with a kinetic mixing of the form $f_K = f_{\mathcal{R}}^m$ -- where $n=2$ and 4 and $0.5 \le m \le 10$ -- reconciles chaotic inflation based on the $\phi^n$ potential with the recent ACT results, if we adopt the Palatini formulation of gravity. Th

Jingxuan Wei, Nan Xu, Junnan Zhu, Yanni Hao

Chart question answering (CQA) has become a critical multimodal task for evaluating the reasoning capabilities of vision-language models. While early approaches have shown promising performance by focusing on visual features or leveraging large-scale pre-training, most existing evaluations rely on rigid output formats and objective metrics, thus ignoring the

Hugo Galeano, Orlando Chaljub

We make a classification of codimension one degree 3 distributions on the projective three space, giving possible Chern classes of the tangent sheaf and describing de zero and one dimensional components of the singular scheme of the distribution. Also, we show the existence and describe some moduli spaces of such distributions, using the concept of stability

Hemant Tyagi

Given an undirected and connected graph $G$ on $T$ vertices, suppose each vertex $t$ has a latent signal $x_t \in \mathbb{R}^n$ associated to it. Given partial linear measurements of the signals, for a potentially small subset of the vertices, our goal is to estimate $x_t$'s. Assuming that the signals are smooth w.r.t $G$, in the sense that the quadratic var

Lingkai Meng, Yu Shao, Long Yuan, Longbin Lai

Usability evaluation is critical to the impact and adoption of open source software (OSS), yet traditional methods relying on human evaluators suffer from high costs and limited scalability. To address these limitations, we introduce OSS-UAgent, an automated, configurable, and interactive agent-based usability evaluation framework specifically designed for o

Dennis-Magnus Welz

We prove the Riemann Hypothesis via an analytically regulated surface integral over the critical strip of the Riemann zeta function. The key idea is that the convergence of this normalized integral is equivalent to the condition that all non-trivial zeros lie on the critical line. By constructing a singularity-sensitive integrand and removing infinitesimal d

Jiuyu Sun, Jinzhe Han, Yongping Du, Erjun Kan

Spin-momentum locking (SML) excitons in two-dimensional semiconductors are appealing to programmable optical control of spin-polarized carriers in ultrafast spintronics. To address the current thirsty for long-lived excitons with zero-external-field stability and room-temperature spin-polarization, we hereby predict the existence of intrinsically SML exciton

Youjun Chen, Xurong Xie, Haoning Xu, Mengzhe Geng

This paper presents a novel end-to-end LLM-empowered explainable speech emotion recognition (SER) approach. Fine-grained speech emotion descriptor (SED) features, e.g., pitch, tone and emphasis, are disentangled from HuBERT SSL representations via alternating LLM fine-tuning to joint SER-SED prediction and ASR tasks. VAE compressed HuBERT features derived vi

Kin Shing Chan, Kei Fong Lam

We study a thermodynamically consistent phase field model for binary mixtures of micropolar fluids, i.e., fluids exhibiting internal rotations. Furnishing with classical no-slip, no-spin and no-flux boundary conditions, in a smooth and bounded three-dimensional domain, we establish the well-posedness of local-in-time strong solutions. Since the model studied

Duc Ninh Le, Thi Nhung Dao

In this report we discuss the definition of the polarized cross sections of the inclusive $W^+W^-$ production at the LHC. Results at the level of next-to-leading order (NLO) QCD+EW accuracy, published in our recent paper, are presented to highlight the effects of bottom-quark induced processes. Compared to the unpolarized case, the bottom-induced effects aft

Patrizia Schalk, Artem Polyvyanyy

Simple process models are key for effectively communicating the outcomes of process mining. An important question in this context is whether the complexity of event logs used as inputs to process discovery algorithms can serve as a reliable indicator of the complexity of the resulting process models. Although various complexity measures for both event logs a

Fabrizio Tamburini, Nicolò Leone, Matteo Sanna, Roberto Siagri

We present a theoretical framework that integrates Majorana's infinite-component relativistic equation within the algebraic structure of paraparticles through the minimal nontrivial $\mathbb{Z}_2 \times \mathbb{Z}_2$--graded Lie algebras and $R$-matrix quantization. By mapping spin-dependent mass spectra to graded sectors associated with generalized quantum

MSVPJ Sathvik, Manan Roy Choudhury, Rishita Agarwal, Sathwik Narkedimilli

The rise of online platforms has enabled covert illicit activities, including online prostitution, to pose challenges for detection and regulation. In this study, we introduce REDDIX-NET, a novel benchmark dataset specifically designed for moderating online sexual services and going beyond traditional NSFW filters. The dataset is derived from thousands of we

Janina Hansen, Daniel Angerhausen, Sascha P. Quanz, Derek Vance

Identifying key observables is essential for enhancing our knowledge of exoplanet habitability and biospheres, as well as improving future mission capabilities. While currently challenging, future observatories such as the Large Interferometer for Exoplanets (LIFE) will enable atmospheric observations of a diverse sample of temperate terrestrial worlds. Usin

Hao Lu, Yanchi Gu, Haoyuan Huang, Yulin Zhou

The integration of Monte Carlo Tree Search (MCTS) with Large Language Models (LLMs) has demonstrated significant success in structured, problem-oriented tasks. However, applying these methods to open-ended dialogues, such as those in psychological counseling, presents unique challenges. Unlike tasks with objective correctness, success in therapeutic conversa

Wanfu Gao, Jun Gao, Qingqi Han, Hanlin Pan

The rapid growth in feature dimension may introduce implicit associations between features and labels in multi-label datasets, making the relationships between features and labels increasingly complex. Moreover, existing methods often adopt low-dimensional linear decomposition to explore the associations between features and labels. However, linear decomposi

Liyiming Tao, Zechao Jiang, Shiyuan Hu, Lin Du

Highly aligned nanowire networks are essential for enabling anisotropic optical, electrical, and sensing functionalities in next-generation devices. However, achieving such alignment typically requires complex fabrication methods or high-energy processing. Here, we present a simple and scalable self-assembly strategy that uses a viscosity-enhancing polymer a

Zheng Gong, Ziyi Jiang, Weihao Gao, Yuanyuan Wang

The mRNA optimization is essential for mRNA vaccines, therapies, and industrial protein production. Based on current explorations, an ideal optimization approach should simultaneously (i) prevent unintended amino-acid changes, (ii) optimize multiple, biologically relevant objectives, and (iii) retain computational efficiency. However, existing methods are fo

Renye Zhang, Mengyun Yang, Qichang Zhao, Jianxin Wang

Drug repositioning aims to identify potential new indications for existing drugs to reduce the time and financial costs associated with developing new drugs. Most existing deep learning-based drug repositioning methods predominantly utilize graph-based representations. However, graph-based drug repositioning methods struggle to perform effective inference in

H. Hadi, Amin Rezaei Akbarieh, Emmanuel N. Saridakis

We suggest a quantum circuit model which simulates the black-hole evaporation process. In particular, Almheiri-Marolf-Polchinski-Sully (AMPS) paradox and the ER=EPR correspondence are reconsidered regarding our proposed model, which assumes a Maxwell's demon operating within a black hole interior. In other words, we form a quantum circuit, mimicking the beha

Fabiano Veglianti, Flavio Giorgi, Fabrizio Silvestri, Gabriele Tolomei

In this work, we investigate the relationship between model generalization and counterfactual explainability in supervised learning. We introduce the notion of $\varepsilon$-valid counterfactual probability ($\varepsilon$-VCP) -- the probability of finding perturbations of a data point within its $\varepsilon$-neighborhood that result in a label change. We p

Hugon Lee, Hyeonbin Moon, Junhyeong Lee, Seunghwa RYu

Artificial intelligence (AI) is reshaping inverse design in manufacturing, enabling high-performance discovery in materials, products, and processes. However, purely data-driven approaches often struggle in realistic manufacturing settings characterized by sparse data, high-dimensional design spaces, and complex constraints. This perspective proposes an inte

Giovanni Acampora, Andris Ambainis, Natalia Ares, Leonardo Banchi

This white paper discusses and explores the various points of intersection between quantum computing and artificial intelligence (AI). It describes how quantum computing could support the development of innovative AI solutions. It also examines use cases of classical AI that can empower research and development in quantum technologies, with a focus on quantu

Zhitao He, Sandeep Polisetty, Zhiyuan Fan, Yuchen Huang

In recent years, multimodal large language models (MLLMs) have made significant progress but continue to face inherent challenges in multimodal reasoning, which requires multi-level (e.g., perception, reasoning) and multi-granular (e.g., multi-step reasoning chain) advanced inferencing. Prior work on estimating model confidence tends to focus on the overall

Wenju Sun, Qingyong Li, Wen Wang, Yang Liu

Multi-task model merging aims to consolidate knowledge from multiple fine-tuned task-specific experts into a unified model while minimizing performance degradation. Existing methods primarily approach this by minimizing differences between task-specific experts and the unified model, either from a parameter-level or a task-loss perspective. However, paramete

Xingyuan Pan, Chenlu Ye, Joseph Melkonian, Jiaqi W. Ma

Training data attribution (TDA) methods aim to identify which training examples influence a model's predictions on specific test data most. By quantifying these influences, TDA supports critical applications such as data debugging, curation, and valuation. Gradient-based TDA methods rely on gradients and second-order information, limiting their applicability

J. L. Chen, J. S. Zhang, J. X. Ge, Y. X. Wang

Using the IRAM 30 m telescope, we presented observations of N2H+ J = 1-0, CCS JN = 87-76 and 77-66 lines toward a large sample of ultracompact HII regions (UC HIIs). Among our 88 UC HIIs, 87 and 33 sources were detected in the N2H+ J = 1-0 and CCS JN = 87-76 lines, respectively. For the CCS 77-66 transition, we detected emission in 10 out of 82 targeted sour

Geng Li, Chunjiang Shi, Ying Chen, Wei Sun

We investigate the $S$-wave $η_cη_c$ and $J/ψJ/ψ$ scattering in the $J^{PC}=(0,2)^{++}$ channels up to a center-of-mass energy of 6.6~GeV. The calculations are carried out at two unphysical pion masses, $m_π\approx 420$~MeV and 250~MeV, in $N_f=2$ lattice QCD. For each $m_π$, we extract the finite-volume energy levels on two lattices with an identical lattic

Jinhui Wei, Ye Huang, Yuhui Zhou, Jiazhi Jiang

In-situ LLM inference on end-user devices has gained significant interest due to its privacy benefits and reduced dependency on external infrastructure. However, as the decoding process is memory-bandwidth-bound, the diverse processing units in modern end-user devices cannot be fully exploited, resulting in slow LLM inference. This paper presents Ghidorah, a

Nguyen Sum

Let $P_k$ be the polynomial algebra $\mathbb F_2[x_1,x_2,\ldots ,x_k]$ over the field $\mathbb F_2$ with two elements, in $k$ variables $x_1, x_2, \ldots , x_k$, each variable of degree 1. Denote by $GL_k$ the general linear group over $\mathbb F_2$ which regularly acts on $P_k$. The algebra $P_k$ is a module over the mod-2 Steenrod algebra $\mathcal A$. In

Jessica Park, Susan Stepney, Irene D'Amico

Boson Sampling, a non-universal computing paradigm, has resulted in impressive claims of quantum supremacy. ORCA Computing have developed a time-bin interferometer (TBI) that claims to use the principles of boson sampling to solve a number of computational problems including optimisation and generative adversarial networks. We solve a dominating set problem

Armando Maria Monforte, Andrea Moiola

We propose a Trefftz discontinuous Galerkin (TDG) method for the approximation of plane wave scattering by periodic diffraction gratings, modelled by the two-dimensional Helmholtz equation. The periodic obstacle may include penetrable and impenetrable regions. The TDG method requires the approximation of the Dirichlet-to-Neumann (DtN) operator on the periodi

T. Jahn, J. Chemseddine, P. Hagemann, C. Wald

Accurately modeling time-continuous stochastic processes from irregular observations remains a significant challenge. In this paper, we leverage ideas from generative modeling of image data to push the boundary of time series generation. For this, we find new generators of SDEs and jump processes, inspired by trajectory flow matching, that have the marginal

Wenhao Xu, Shuchen Zheng, Changwei Wang, Zherui Zhang

Infrared small target detection (ISTD) is vital for long-range surveillance in military, maritime, and early warning applications. ISTD is challenged by targets occupying less than 0.15% of the image and low distinguishability from complex backgrounds. Existing deep learning methods often suffer from information loss during downsampling and inefficient globa

Qianwang Ye, Hanqing Dai, Yukun Yan, Liwei Wang

With the energy crisis and climate warming, the position of a new generation of smart windows is becoming increasingly important, and materials or systems that can have high blocking of near-infrared (NIR) and ultraviolet (UV) and high transmittance of visible light (VIS) are needed. Currently, it is difficult for smart heat-insulation materials to achieve h

Kohei Saijo, Wangyou Zhang, Samuele Cornell, Robin Scheibler

There has been a growing effort to develop universal speech enhancement (SE) to handle inputs with various speech distortions and recording conditions. The URGENT Challenge series aims to foster such universal SE by embracing a broad range of distortion types, increasing data diversity, and incorporating extensive evaluation metrics. This work introduces the

Atte Pennanen

Let $0<p<\infty$ and $\Psi: [0,1) \to (0,\infty)$, and let $\mu$ be a finite positive Borel measure on the unit disc $\mathbb{D}$ of the complex plane. We define the Lebesgue-Zygmund space $L^p_{\mu,\Psi}$ as the space of all measurable functions $f$ on $\mathbb{D}$ such that $\int_{\mathbb{D}}|f(z)|^p\Psi(|f(z)|)\,d\mu(z)<\infty$. The weighted Bergman-Zygmu

Paul Lutkus, Kaiyuan Wang, Lars Lindemann, Stephen Tu

We initiate a formal study on the use of low-dimensional latent representations of dynamical systems for verifiable control synthesis. Our main goal is to enable the application of verification techniques -- such as Lyapunov or barrier functions -- that might otherwise be computationally prohibitive when applied directly to the full state representation. Tow

Akash Dhasade, Divyansh Jhunjhunwala, Milos Vujasinovic, Gauri Joshi

Model merging has emerged as an efficient method to combine multiple single-task fine-tuned models. The merged model can enjoy multi-task capabilities without expensive training. While promising, merging into a single model often suffers from an accuracy gap with respect to the fine-tuned models. On the other hand, deploying all individual fine-tuned models

Aashwin Mishra, Matt Seaberg, Ryan Roussel, Fred Poitevin

With the increasing brightness of Light sources, including the Diffraction-Limited brightness upgrade of APS and the high-repetition-rate upgrade of LCLS, the proposed experiments therein are becoming increasingly complex. For instance, experiments at LCLS-II-HE will require the X-ray beam to be within a fraction of a micron in diameter, with pointing stabil

Subhajyoti Pal, Pradeep Thakur, Ashis Kumar Nandy, Anamitra Mukherjee

We investigate spin-flip excitations in the spin-1/2 trimer chain $\rm{Cu_3(P_2O_6OH)_2}$, featuring an antiferromagnetic exchange motif $J_1$-$J_1$-$J_2$ with $J_1 < J_2$. Using density matrix renormalization group (DMRG) simulations, we demonstrate that single-spin-flip processes induced by resonant inelastic X-ray scattering (RIXS) generate emergent gaple

Zhaokai Sun, Li Zhang, Qing Wang, Pan Zhou

Overlapping Speech Detection (OSD) aims to identify regions where multiple speakers overlap in a conversation, a critical challenge in multi-party speech processing. This work proposes a speaker-aware progressive OSD model that leverages a progressive training strategy to enhance the correlation between subtasks such as voice activity detection (VAD) and ove

Aldino Rizaldy, Richard Gloaguen, Fabian Ewald Fassnacht, Pedram Ghamisi

Multimodal remote sensing data, including spectral and lidar or photogrammetry, is crucial for achieving satisfactory land-use / land-cover classification results in urban scenes. So far, most studies have been conducted in a 2D context. When 3D information is available in the dataset, it is typically integrated with the 2D data by rasterizing the 3D data in

Massimo Bartoletti, Stefano Bonzio, Marco Ferrara

A numerical semigroup is a co-finite submonoid of the monoid of non-negative integers under addition. Many properties of numerical semigroups rely on some fundamental invariants, such as, among others, the set of gaps (and its cardinality), the Ap\'ery set or the Frobenius number. Algorithms for calculating invariants are currently based on computational too

Prajwal Hassan Puttasiddappa, Davi C Rodrigues, David F Mota

We investigate the observational features of exact vacuum solutions in Brans-Dicke (BD) gravity, focusing on their implications for black hole shadow imaging. Motivated by the Event Horizon Telescope (EHT) observations, we revisit a class of BD solutions that exhibit a naked singularity. These solutions, despite lacking a conventional event horizon, exhibit

Syu Kato

We characterize the $k$-Schur functions as the graded characters of simple objects in an additive module category. This confirms a set of conjectures formulated in the Ph.D. thesis of Chen, written under the direction of Mark Haiman, and thereby establishes the algebraic framework proposed therein. As a consequence, we deduce that the modified Macdonald poly

Hao Wu, Junzhou Chen, Ronghui Zhang, Nengchao Lyu

Object detection is a cornerstone of environmental perception in advanced driver assistance systems(ADAS). However, most existing methods rely on RGB cameras, which suffer from significant performance degradation under low-light conditions due to poor image quality. To address this challenge, we proposes WTEFNet, a real-time object detection framework specif

Navinder Singh

In 1991, V. J. Emery in his important review article entitled "Some aspects of the theory of high temperature superconductors"\cite{emery1} argued against the Zhang-Rice reduction of three-band to an effective one-band model. In his words "...therefore it seems that the simple $t-J$ model does not account for the properties of high temperature superconductor

Si-wen Li, Xun Chen

In recent years, the investigation of chaos has become a bridge connecting gravity theory and quantum field theory, especially within the framework of gauge-gravity duality. In this work, we study holographically the chaos in the matrix models for meson and baryon, which are derived from the $\mathrm{D4}/\mathrm{D6}/\overline{\mathrm{D6}}$ approach as a top-

Junyong Shin, Eunsung Jeon, Inhyoung Kim, Yo-Seb Jeon

To achieve higher throughput in next-generation Wi-Fi systems, a station (STA) needs to efficiently compress channel state information (CSI) and feed it back to an access point (AP). In this paper, we propose a novel deep learning (DL)-based CSI feedback framework tailored for next-generation Wi-Fi systems. Our framework incorporates a pair of encoder and de

Jatin Kumar Arora, Soutrik Bandyopadhyay, Sunil Sulania, Shubhendu Bhasin

Path planning for autonomous robots faces a fundamental trade-off between path length and obstacle clearance. While existing algorithms typically prioritize a single objective, we introduce the Unified Path Planner (UPP), a graph-search algorithm that dynamically balances safety and optimality via adaptive heuristic weighting. UPP employs a local inverse-dis

Eshant English, Christoph Lippert

Conformal prediction provides a model-agnostic framework for uncertainty quantification with finite-sample validity guarantees, making it an attractive tool for constructing reliable prediction sets. However, existing approaches commonly rely on residual-based conformity scores, which impose geometric constraints and struggle when the underlying distribution

Lifan Zhao, Yanyan Shen, Zhaoyang Liu, Xue Wang

Scaling laws motivate the development of Time Series Foundation Models (TSFMs) that pre-train vast parameters and achieve remarkable zero-shot forecasting performance. Surprisingly, even after fine-tuning, TSFMs cannot consistently outperform smaller, specialized models trained on full-shot downstream data. A key question is how to realize effective adaptati

Shiwei Li, Xiandi Luo, Xing Tang, Haozhao Wang

Low-rank adaptation (LoRA) is a widely used parameter-efficient fine-tuning method. In standard LoRA layers, one of the matrices, $A$ or $B$, is initialized to zero, ensuring that fine-tuning starts from the pretrained model. However, there is no theoretical support for this practice. In this paper, we investigate the impact of non-zero initialization on LoR

Sungjune Park, Hyunjun Kim, Beomchan Park, Yong Man Ro

Despite recent advancements in computer vision research, object detection in aerial images still suffers from several challenges. One primary challenge to be mitigated is the presence of multiple types of variation in aerial images, for example, illumination and viewpoint changes. These variations result in highly diverse image scenes and drastic alterations

Run Hao, Peng Ying

The rise of text-to-image (T2I) models has enabled the synthesis of photorealistic human portraits, raising serious concerns about identity misuse and the robustness of AIGC detectors. In this work, we propose an automated adversarial prompt generation framework that leverages a grammar tree structure and a variant of the Monte Carlo tree search algorithm to

Jinglong Gao, Xiao Ding, Lingxiao Zou, Bibo Cai

Recent studies provide large language models (LLMs) with textual task-solving experiences via prompts to improve their performance. However, previous methods rely on substantial human labor or time to gather such experiences for each task, which is impractical given the growing variety of task types in user queries to LLMs. To address this issue, we design a

Yekun Zhu, Min Tang, Zheng Ma

In this paper, we propose a novel neural network approach, termed DeepRTE, to address the steady-state Radiative Transfer Equation (RTE). The RTE is a differential-integral equation that governs the propagation of radiation through a participating medium, with applications spanning diverse domains such as neutron transport, atmospheric radiative transfer, he

Shaoan Wang, Jiazhao Zhang, Minghan Li, Jiahang Liu

Embodied visual tracking is a fundamental skill in Embodied AI, enabling an agent to follow a specific target in dynamic environments using only egocentric vision. This task is inherently challenging as it requires both accurate target recognition and effective trajectory planning under conditions of severe occlusion and high scene dynamics. Existing approac

Hong-Ming Cui, Zhong-Ying Fan

We extend the thermodynamics of quantum BTZ black holes by treating the quantum backreaction strength parameter $\nu$ as a thermodynamic variable. We find various novel features. The critical point appears at $\nu_c=1$ and a first order transition occurs either below the critical temperature for $\nu<\nu_c$ or above the critical temperature for $\nu>\nu_c$.

Yilong Li, Chen Qian, Yu Xia, Ruijie Shi

Large Language Model-based multi-agent systems (MAS) have shown remarkable progress in solving complex tasks through collaborative reasoning and inter-agent critique. However, existing approaches typically treat each task in isolation, resulting in redundant computations and limited generalization across structurally similar tasks. To address this, we introd

Junyi Guo, Jingxuan Zhang, Fangyu Wu, Huanda Lu

Diffusion-based garment synthesis tasks primarily focus on the design phase in the fashion domain, while the garment production process remains largely underexplored. To bridge this gap, we introduce a new task: Flat Sketch to Realistic Garment Image (FS2RG), which generates realistic garment images by integrating flat sketches and textual guidance. FS2RG pr

Shahaf E. Finder, Ron Shapira Weber, Moshe Eliasof, Oren Freifeld

Message-Passing Neural Networks (MPNNs) have become a cornerstone for processing and analyzing graph-structured data. However, their effectiveness is often hindered by phenomena such as over-squashing, where long-range dependencies or interactions are inadequately captured and expressed in the MPNN output. This limitation mirrors the challenges of the Effect