May 2025 arXiv papers — page 46

Roberto F. Pitzalis, Nicholas Cartocci, Christian Di Natali, Luigi Monica

Musculoskeletal disorders (MSD) are the most common cause of work-related injuries and lost production involving approximately 1.7 billion people worldwide and mainly affect low back (more than 50%) and upper limbs (more than 40%). It has a profound effect on both the workers affected and the company. This paper provides an ergonomic assessment of different

Chongjie Si, Yidan Cui, Fuchao Yang, Xiaokang Yang

Partial Multi-Label Learning (PML) extends the multi-label learning paradigm to scenarios where each sample is associated with a candidate label set containing both ground-truth labels and noisy labels. Existing PML methods commonly rely on two assumptions: sparsity of the noise label matrix and low-rankness of the ground-truth label matrix. However, these a

Deepesh Gavit, Debajyoti Mazumder, Samiran Das, Jasabanta Patro

In this paper, we present a comprehensive and systematic analysis of vision-language models (VLMs) for disparate meme classification tasks. We introduced a novel approach that generates a VLM-based understanding of meme images and fine-tunes the LLMs on textual understanding of the embedded meme text for improving the performance. Our contributions are three

Subhaditya Bhattacharya, Soumyajit Datta, Abhik Sarkar

We investigate lepton number violation (LNV) induced by $\Delta L = 2$ dimension-seven Standard Model Effective Field Theory (SMEFT) operators in the context of same-sign muon colliders. Specifically, we study $\mu^+ \mu^+ \rightarrow W^+W^+/\;W^+qq'$ production at the $\mu$TRISTAN at $\sqrt{s} =$ 2 TeV with an integrated luminosity of 1 ab$^{-1}$, using the

Max Collins, Jordan Vice, Tim French, Ajmal Mian

Adversarial samples exploit irregularities in the manifold `learned' by deep learning models to cause misclassifications. The study of these adversarial samples provides insight into the features a model uses to classify inputs, which can be leveraged to improve robustness against future attacks. However, much of the existing literature focuses on constraine

Duzhen Zhang, Yong Ren, Chenxing Li, Dong Yu

Continual Text Classification (CTC) aims to continuously classify new text data over time while minimizing catastrophic forgetting of previously acquired knowledge. However, existing methods often focus on task-specific knowledge, overlooking the importance of shared, task-agnostic knowledge. Inspired by the complementary learning systems theory, which posit

Ningyuan Tang, Minghao Fu, Hao Yu, Jianxin Wu

Network quantization is arguably one of the most practical network compression approaches for reducing the enormous resource consumption of modern deep neural networks. They usually require diverse and subtle design choices for specific architecture and tasks. Instead, the QwT method is a simple and general approach which introduces lightweight additional st

Ahmet Burak Ozyurt, Shreesh Mohalik, John S. Thompson

With the increasing frequency and intensity of natural disasters, there is a necessity for advanced technologies that can provide reliable situational awareness and communication. Conventional systems are often inadequate due to unreliable infrastructure, power grid failures, high investment costs and scalability challenges. This paper explores the potential

Ruiqi Zhang, Simon H. Tindemans

Multilevel Monte Carlo (MLMC) is a flexible and effective variance reduction technique for accelerating reliability assessments of complex power system. Recently, data-driven surrogate models have been proposed as lower-level models in the MLMC framework due to their high correlation and negligible execution time once trained. However, in resource adequacy a

Yunhan Du, Takaaki Aoki, Naoya Fujiwara

Understanding human mobility is vital to solving societal challenges, such as epidemic control and urban transportation optimization. Recent advancements in data collection now enable the exploration of dynamic mobility patterns in human flow. However, the vast volume and complexity of mobility data make it difficult to interpret spatiotemporal patterns dire

Francesco Cordiano, Matin Jafarian, Bart De Schutter

We propose a novel distribution-free scheme to solve optimization problems where the goal is to minimize the expected value of a cost function subject to probabilistic constraints. Unlike standard sampling-based methods, our idea consists of partitioning the uncertainty domain in a user-defined number of sets, enabling more flexibility in the trade-off betwe

Jun Liu, Hongxun Liu, Cheng Zhang, Jiandang Xing

An unmanned deformable vehicle is a wheel-legged robot transforming between two configurations: vehicular and humanoid states, with different motion modes and stability characteristics. To address motion stability in multiple configurations, a center-of-mass adjustment mechanism was designed. Further, a motion stability hierarchical control algorithm was pro

Zhuo Li, Guodong Du, Weiyang Guo, Yigeng Zhou

Aligning large language models (LLMs) to simultaneously satisfy multiple objectives remains a significant challenge, especially given the diverse and often conflicting nature of human preferences. Existing alignment methods struggle to balance trade-offs effectively, often requiring costly retraining or yielding suboptimal results across the Pareto frontier

Marius Bock, Maximilian Hopp, Kristof Van Laerhoven, Michael Moeller

While prior work has shown that Federated Learning updates can leak sensitive information, label reconstruction attacks, which aim to recover input labels from shared gradients, have not yet been examined in the context of Human Activity Recognition (HAR). Given the sensitive nature of activity labels, this study evaluates the effectiveness of state-of-the-a

Marius Müller

We study a higher order version of the Alt-Caffarelli problem in two dimensions, where the Dirichlet energy is replaced by an anisotropic bending energy. This extends a previous study of the isotropic case in [41]. It turns out that smooth anisotropies do not affect the optimal $C^{2,1}$-regularity of minimizers. The proof requires an anisotropic version of

Yang Zhang, Xinran Li, Jianing Ye, Shuang Qiu

World models have recently attracted growing interest in Multi-Agent Reinforcement Learning (MARL) due to their ability to improve sample efficiency for policy learning. However, accurately modeling environments in MARL is challenging due to the exponentially large joint action space and highly uncertain dynamics inherent in multi-agent systems. To address t

Injae Na, Keonwoong Noh, Woohwan Jung

LLM providers typically offer multiple LLM tiers, varying in performance and price. As NLP tasks become more complex and modularized, selecting the suitable LLM tier for each subtask is a key challenge to balance between cost and performance. To address the problem, we introduce LLM Automatic Transmission (LLM-AT) framework that automatically selects LLM tie

Qihang Fang, Chengcheng Tang, Bugra Tekin, Shugao Ma

We present HuMoCon, a novel motion-video understanding framework designed for advanced human behavior analysis. The core of our method is a human motion concept discovery framework that efficiently trains multi-modal encoders to extract semantically meaningful and generalizable features. HuMoCon addresses key challenges in motion concept discovery for unders

Amit Kumar Jha

This study is motivated by empirical observations of periodic fluctuations in interest rates, notably long-term economic cycles spanning decades, which the conventional Hull-White short-rate model fails to adequately capture. To address this limitation, we propose an extension that incorporates a sinusoidal, time-varying mean reversion speed, allowing the mo

Rahul Nair, Inge Vejsbjerg, Elizabeth Daly, Christos Varytimidis

Humble AI (Knowles et al., 2023) argues for cautiousness in AI development and deployments through scepticism (accounting for limitations of statistical learning), curiosity (accounting for unexpected outcomes), and commitment (accounting for multifaceted values beyond performance). We present a real-world case study for humble AI in the domain of algorithmi

Nicholas Cartocci, Antonios E. Gkikakis, Natalia Kurvina, Natnael Takele

A key aspect of developing fall prevention systems is the early prediction of a fall before it occurs. This paper presents a statistical overview of results obtained by analyzing 22 activities of daily living to recognize physiological patterns and estimate the risk of an imminent fall. The results demonstrate distinctive patterns between high-intensity and

S. N. Vergeles

The unitarity of the 4D lattice theory of gravity in the case of the Minkowski signature is proved. The proof is valid only for lattices that conserve the number of degrees of freedom during time evolution. The Euclidean signature and the Minkowski signature are related by the deformation of the integration contours of dynamic variables in a discrete lattice

Kyzyl Monteiro, Yuchen Wu, Sauvik Das

Users often struggle to navigate the privacy / publicity boundary in sharing images online: they may lack awareness of image privacy risks and/or the ability to apply effective mitigation strategies. To address this challenge, we introduce and evaluate Imago Obscura, an AI-powered, image-editing copilot that enables users to identify and mitigate privacy ris

Nicolas Bousquet, Laurent Feuilloley, Sébastien Zeitoun

An impressive recent line of work has charted the complexity landscape of distributed graph algorithms. For many settings, it has been determined which time complexities exist, and which do not (in the sense that no local problem could have an optimal algorithm with that complexity). In this paper, we initiate the study of the landscape for space complexity

Anum Fatima, Gesine Reinert

Complex data are often represented as a graph, which in turn can often be viewed as a realisation of a random graph, such as an inhomogeneous random graph model (IRG). For general fast goodness-of-fit tests in high dimensions, kernelised Stein discrepancy (KSD) tests are a powerful tool. Here, we develop a KSD-type test for IRG models that can be carried out

Diksha Kashyap, Devanshi Gupta, Naman Kumar Mehta, Gajendra P. S. Raghava

Addressing the growing need for organized data on tumor homing peptides (THPs), we present TumorHoPe2, a manually curated database offering extensive details on experimentally validated THPs. This represents a significant update to TumorHoPe, originally developed by our group in 2012. TumorHoPe2 now contains 1847 entries, representing 1297 unique tumor homin

Romain de Laage

Fully homomorphic encryption (FHE) and trusted execution environments (TEE) are two approaches to provide confidentiality during data processing. Each approach has its own strengths and weaknesses. In certain scenarios, computations can be carried out in a hybrid environment, using both FHE and TEE. However, processing data in such hybrid settings presents c

Bálint Siklósi, Pushpender K. Sharma, David J. Lusher, István Z. Reguly

The use of reduced and mixed precision computing has gained increasing attention in high-performance computing (HPC) as a means to improve computational efficiency, particularly on modern hardware architectures like GPUs. In this work, we explore the application of mixed precision arithmetic in compressible turbulent flow simulations using explicit finite di

Hang Zeng, Xiangyu Liu, Yong Hu, Chaoyue Niu

Users interacting with large language models (LLMs) under their real identifiers often unknowingly risk disclosing private information. Automatically notifying users whether their queries leak privacy and which phrases leak what private information has therefore become a practical need. Existing privacy detection methods, however, were designed for different

Wei Li, Hebei Li, Yansong Peng, Siying Wu

Diffusion models have significantly advanced text-to-image generation, laying the foundation for the development of personalized generative frameworks. However, existing methods lack precise layout controllability and overlook the potential of dynamic features of reference subjects in improving fidelity. In this work, we propose Layout-Controllable Personali

Wei Meng

This study proposes a systematic non-kinetic deterrence path modeling framework based on strategic rare earth supply cut-off, aiming to assess the strategic effects of China's export control policy against the United States at the military system level. The model adopts a four-layer structure of "policy input -- resource node -- equipment system -- capabilit

Leo Kotipalo, Markus Battarbee, Yann Pfau-Kempf, Vertti Tarvus

Parallelization is a necessity for large-scale simulations due to the amount of data processed. In this article we investigate different load balancing methods using Vlasiator, a global magnetospheric simulation as our case study. The theoretical basis for load balancing is the (hyper)graph partitioning problem, modeling simulation units as vertices and thei

Kanta Kudo, Youichi Yanase

We uncover a geometric mechanism of odd-parity multipole magnetism driven by the quantum metric of Bloch electrons. By analyzing spin and odd-parity multipole susceptibilities in a multi-sublattice model, we demonstrate that the quantum metric directly controls the instability toward odd-parity magnetic multipole order over a wide range of parameters, which

Bingxiang Kang, Jie Zou, Guofa Li, Pengwei Zhang

Visual-inertial simultaneous localization and mapping (SLAM) is a key module of robotics and low-speed autonomous vehicles, which is usually limited by the high computation burden for practical applications. To this end, an innovative strategy-based hybrid framework HS-SLAM is proposed to integrate the advantages of direct and feature-based methods for fast

A. S. Mikhaylov, V. S. Mikhaylov

We consider dynamical systems with boundary control associated with finite Jacobi matrices. Using the method previously developed by the authors, we associate with these systems special Hilbert spaces of analytic functions (de Branges spaces)

Guanghu Xie, Yonglong Zhang, Zhiduo Jiang, Yang Liu

Transparent and reflective objects pose significant challenges for depth sensors, resulting in incomplete depth information that adversely affects downstream robotic perception and manipulation tasks. To address this issue, we propose HTMNet, a novel hybrid model integrating Transformer, CNN, and Mamba architectures. The encoder is based on a dual-branch CNN

Ziming Wang, Zeyu Shi, Haoyi Zhou, Shiqi Gao

Fine-tuned Large Language Models (LLMs) often demonstrate poor calibration, with their confidence scores misaligned with actual performance. While calibration has been extensively studied in models trained from scratch, the impact of LLMs' prior knowledge on calibration during fine-tuning remains understudied. Our research reveals that LLMs' prior knowledge

Ruiying Li, Bin Pan, Lan Ma, Xia Xu

Multitemporal hyperspectral unmixing can capture dynamical evolution of materials. Despite its capability, current methods emphasize variability of endmembers while neglecting dynamics of abundances, which motivates our adoption of neural ordinary differential equations to model abundances temporally. However, this motivation is hindered by two challenges: t

Junhyuk Choi, Minju Kim, Yeseon Hong, Bugeun Kim

As large vision language models(LVLMs) rapidly advance, concerns about their potential to learn and generate social biases and stereotypes are increasing. Previous studies on LVLM's stereotypes face two primary limitations: metrics that overlooked the importance of content words, and datasets that overlooked the effect of color. To address these limitations,

Zhongjin Zhang, Yu Liang, Cong Fu, Yuxuan Zhu

Embedding-based collaborative filtering, often coupled with nearest neighbor search, is widely deployed in large-scale recommender systems for personalized content selection. Modern systems leverage multiple implicit feedback signals (e.g., clicks, add to cart, purchases) to model user preferences comprehensively. However, prevailing approaches adopt a feedb

Jeongsoo Choi, Jaehun Kim, Joon Son Chung

This paper introduces a cross-lingual dubbing system that translates speech from one language to another while preserving key characteristics such as duration, speaker identity, and speaking speed. Despite the strong translation quality of existing speech translation approaches, they often overlook the transfer of speech patterns, leading to mismatches with

Garima Khetawat, Moumita Manna, Tarakanta Nayak

By an independent set in a simple graph $G$, we mean a set of pairwise non-adjacent vertices in $G$. The independence polynomial of $G$ is defined as $I_G(z)=a_0 + a_1 z + a_2 z^2+\cdots+a_\alpha z^{\alpha}$, where $a_i$ is the number of independent sets in $G$ with cardinality $i$ and $\alpha$ is the cardinality of a largest independent set in $G$, known as

Titouan Parcollet, Yuan Tseng, Shucong Zhang, Rogier van Dalen

Automatic speech recognition (ASR) research is driven by the availability of common datasets between industrial researchers and academics, encouraging comparisons and evaluations. LibriSpeech, despite its long success as an ASR benchmark, is now limited by its size and focus on clean, read speech, leading to near-zero word error rates. More recent datasets,

Pingrui Zhang, Yifei Su, Pengyuan Wu, Dong An

Vision-and-Language Navigation (VLN) requires the agent to navigate by following natural instructions under partial observability, making it difficult to align perception with language. Recent methods mitigate this by imagining future scenes, yet they rely on vision-based synthesis, leading to high computational cost and redundant details. To this end, we pr

Haowei Yang, Haotian Lyu, Tianle Zhang, Dingzhou Wang

As e-commerce competition intensifies, balancing creative content with conversion effectiveness becomes critical. Leveraging LLMs' language generation capabilities, we propose a framework that integrates prompt engineering, multi-objective fine-tuning, and post-processing to generate marketing copy that is both engaging and conversion-driven. Our fine-tuning

Yiwei Wu, Atticus Geiger, Raphaël Millière

Variable binding -- the ability to associate variables with values -- is fundamental to symbolic computation and cognition. Although classical architectures typically implement variable binding via addressable memory, it is not well understood how modern neural networks lacking built-in binding operations may acquire this capacity. We investigate this by tra

Fabian Reimers, Müfit Sezer

For modular indecomposable representations of a cyclic group $G$ of prime order $p$ we propose a list of polynomial invariants of degree $\leq 3$ that, together with a simple invariant of degree $p$, separate generic orbits and generate the field of rational invariants. A similar result is proven for decomposable representations of $G$.

Huacan Wang, Ziyi Ni, Shuo Zhang, Shuo Lu

The ultimate goal of code agents is to solve complex tasks autonomously. Although large language models (LLMs) have made substantial progress in code generation, real-world tasks typically demand full-fledged code repositories rather than simple scripts. Building such repositories from scratch remains a major challenge. Fortunately, GitHub hosts a vast, evol

Jeonghwan Cheon, Jaehyuk Bae, Se-Bum Paik

Backpropagation is the cornerstone of deep learning, but its reliance on symmetric weight transport and global synchronization makes it computationally expensive and biologically implausible. Feedback alignment offers a promising alternative by approximating error gradients through fixed random feedback, thereby avoiding symmetric weight transport. However,

Qihao Peng, Qu Luo, Yi Ma, Chuan Heng Foh

This paper characterizes the impacts of channel estimation errors and Rician factors on achievable data rate and investigates the user scheduling strategy, combining scheme, power control, and dynamic bandwidth allocation to maximize the sum data rate in the distributed multiple-input-multiple-output (MIMO)-enabled low earth orbit (LEO) satellite networks. H

Yoojin Kwon, Hongjun Suh, Wooseok Lee, Taesik Gong

Modern on-device neural network applications must operate under resource constraints while adapting to unpredictable domain shifts. However, this combined challenge-model compression and domain adaptation-remains largely unaddressed, as prior work has tackled each issue in isolation: compressed networks prioritize efficiency within a fixed domain, whereas la

Leizhen Wang, Peibo Duan, Cheng Lyu, Zhenliang Ma

Modern navigation systems and shared mobility platforms increasingly rely on personalized route recommendations to improve individual travel experience and operational efficiency. However, a key question remains: can such sequential, personalized routing decisions collectively lead to system-optimal (SO) traffic assignment? This paper addresses this question

Chengyu Wang, Junbing Yan, Wenrui Cai, Yuanhao Yue

In this paper, we present EasyDistill, a comprehensive toolkit designed for effective black-box and white-box knowledge distillation (KD) of large language models (LLMs). Our framework offers versatile functionalities, including data synthesis, supervised fine-tuning, ranking optimization, and reinforcement learning techniques specifically tailored for KD sc

Kaining Wang, Bo Yang, Yusheng Lei, Zhiwen Yu

Reconfigurable intelligent surfaces (RISs) have demonstrated an unparalleled ability to reconfigure wireless environments by dynamically controlling the phase, amplitude, and polarization of impinging waves. However, as nearly passive reflective metasurfaces, RISs may not distinguish between desired and interference signals, which can lead to severe spectrum

Byon N. Jayawiguna, Piyabut Burikham

We investigate the moment of inertia, quadrupole deformation, and tidal deformation within the framework of nonlocal gravity, utilizing the exact modified Tolman-VII (NEMTVII) density model with an isotropic perfect fluid. The Love number~$(k_{2})$ is derived using standard even-parity perturbation theory. Additionally, we explore the observational implicati

Haipeng Luo, Spandan Senapati, Vatsal Sharan

In this paper, we consider the related problems of multicalibration -- a multigroup fairness notion and omniprediction -- a simultaneous loss minimization paradigm, both in the distributional and online settings. The recent work of Garg et al. (2024) raised the open problem of whether it is possible to efficiently achieve $O(\sqrt{T})$ $\ell_{2}$-multicalibr

Weichao Pan, Bohan Xu, Xu Wang, Chengze Lv

Fire detection in dynamic environments faces continuous challenges, including the interference of illumination changes, many false detections or missed detections, and it is difficult to achieve both efficiency and accuracy. To address the problem of feature extraction limitation and information loss in the existing YOLO-based models, this study propose You

Luc Molinet, Tomoyuki Tanaka

We consider the derivative nonlinear Schr\"odinger equation on the real line, with a background function $\psi(t,x)\in L^\infty(\mathbb{R}^2)$ that satisfies suitable conditions. Such a function may, for example, be a non-decaying solution of the equation, such as a dark soliton. By developing the energy method with correction terms, we prove that the Cauchy

Marc Damie, Edwige Cyffers

Decentralized machine learning - where each client keeps its own data locally and uses its own computational resources to collaboratively train a model by exchanging peer-to-peer messages - is increasingly popular, as it enables better scalability and control over the data. A major challenge in this setting is that learning dynamics depend on the topology of

Yiding Shi, Jianan Zhou, Wen Song, Jieyi Bi

Heuristic design with large language models (LLMs) has emerged as a promising approach for tackling combinatorial optimization problems (COPs). However, existing approaches often rely on manually predefined evolutionary computation (EC) heuristic-optimizers and single-task training schemes, which may constrain the exploration of diverse heuristic algorithms

Baraa Hikal, Ahmed Nasreldin, Ali Hamdi

This paper describes our submission for SemEval-2025 Task 3: Mu-SHROOM, the Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes. The task involves detecting hallucinated spans in text generated by instruction-tuned Large Language Models (LLMs) across multiple languages. Our approach combines task-specific prompt engineer

Max Bastian Mertens, Michael Buchholz

Vehicle-to-anything connectivity, especially for autonomous vehicles, promises to increase passenger comfort and safety of road traffic, for example, by sharing perception and driving intention. Cooperative maneuver planning uses connectivity to enhance traffic efficiency, which has, so far, been mainly considered for automated intersection management. In th

C. -Y. Lee, K. -T. Lin, G. -D. Lin, H. H. Jen

We theoretically investigate excitation dynamics in one-dimensional arrays of quantum emitters coupled to a waveguide, focusing on localization and long-time population trapping. By combining time-domain simulations with spectral analysis of an effective non-Hermitian Hamiltonian, we identify two distinct mechanisms that give rise to localization: geometry-i

Zhaoyi Zhang, Chenggang Cui, Ning Yang, Chuanlin Zhang

The widespread adoption of electric vehicles (EVs) has increased the importance of demand response in smart grids. This paper proposes a two-layer demand response optimization framework for EV users and aggregators, leveraging large language models (LLMs) to balance electricity supply and demand and optimize energy utilization during EV charging. The upper-l

Tatsuya Sasayama, Shintaro Ito, Koichi Ito, Takafumi Aoki

In this paper, we propose a stereo radargrammetry method using deep learning from airborne Synthetic Aperture Radar (SAR) images. Deep learning-based methods are considered to suffer less from geometric image modulation, while there is no public SAR image dataset used to train such methods. We create a SAR image dataset and perform fine-tuning of a deep lear

Jiyoung Lee, Seungho Kim, Jieun Han, Jun-Min Lee

Large Language Models (LLMs) are predominantly evaluated on Standard American English (SAE), often overlooking the diversity of global English varieties. This narrow focus may raise fairness concerns as degraded performance on non-standard varieties can lead to unequal benefits for users worldwide. Therefore, it is critical to extensively evaluate the lingui

Sirui Xia, Aili Chen, Xintao Wang, Tinghui Zhu

Recent advances in Large Language Models (LLMs) have demonstrated strong capabilities in tasks such as code and mathematics. However, their potential to internalize structured spatial knowledge remains underexplored. This study investigates whether LLMs, grounded in locally relative human observations, can construct coherent global spatial cognition by integ

Chaeyoung Jung, Youngjoon Jang, Jongmin Choi, Joon Son Chung

The goal of this work is to enhance balanced multimodal understanding in audio-visual large language models (AV-LLMs) by addressing modality bias without additional training. In current AV-LLMs, audio and video features are typically processed jointly in the decoder. While this strategy facilitates unified multimodal understanding, it may introduce modality

Antony Zhao, Alex Proshkin, Fergal Hennessy, Francesco Crivelli

Large transformer models have been shown to be capable of performing in-context learning. By using examples in a prompt as well as a query, they are capable of performing tasks such as few-shot, one-shot, or zero-shot learning to output the corresponding answer to this query. One area of interest to us is that these transformer models have been shown to be c

Xin Sun, Jianan Xie, Zhongqi Chen, Qiang Liu

Large language models (LLMs) augmented with retrieval systems have significantly advanced natural language processing tasks by integrating external knowledge sources, enabling more accurate and contextually rich responses. To improve the robustness of such systems against noisy retrievals, Retrieval-Augmented Fine-Tuning (RAFT) has emerged as a widely adopte

Chenglin Gong, Ziming Wang, Guanxuan Jiang, Xin Wang

In this paper, we tackle the state transformation problem in non-strict full state-constrained systems by introducing an adaptive fixed-time control method, utilizing a one-to-one asymmetric nonlinear mapping auxiliary system. Additionally, we develop a class of multi-threshold event-triggered control strategies that facilitate autonomous controller updates,

Kuo Zhou, Lu Zhang

Large Language Models (LLMs) have demonstrated formidable capabilities in solving mathematical problems, yet they may still commit logical reasoning and computational errors during the problem-solving process. Thus, this paper proposes a framework, MATH-VF, which includes a Formalizer and a Critic, for formally verifying the correctness of the solutions gene

Nam-Gyu Kim, Deok-Hyeon Cho, Seung-Bin Kim, Seong-Whan Lee

Recent advances in expressive text-to-speech (TTS) have introduced diverse methods based on style embedding extracted from reference speech. However, synthesizing high-quality expressive speech remains challenging. We propose Spotlight-TTS, which exclusively emphasizes style via voiced-aware style extraction and style direction adjustment. Voiced-aware style

Sania Asif

This paper explores various algebraic and homotopical aspects of Nijenhuis Lie conformal algebras, including their cohomology theory, $\mathcal{L}_\infty$-structures, non-abelian extensions, and automorphism groups. We define the cohomology of a Nijenhuis Lie conformal algebra and relate it to the deformation theory of such structures. We also introduce $2$-

Lin Mu, Xiaoyu Wang, Li Ni, Yang Li

Low-rank adaptation (LoRA) has been developed as an efficient approach for adapting large language models (LLMs) by fine-tuning two low-rank matrices, thereby reducing the number of trainable parameters. However, prior research indicates that many of the weights in these matrices are redundant, leading to inefficiencies in parameter utilization. To address t

Xinjie Lin, Gang Xiong, Gaopeng Gou, Wenqi Dong

Encrypted traffic classification is highly challenging in network security due to the need for extracting robust features from content-agnostic traffic data. Existing approaches face critical issues: (i) Distribution drift, caused by reliance on the closedworld assumption, limits adaptability to realworld, shifting patterns; (ii) Dependence on labeled data r

Andrea Gentile, Idriss Mazari-Fouquer, Raphaël Prunier

The goal of this paper is to address some optimal control and shape optimisation problems arising from bulk-surface cooperative systems. The basic model under consideration is the following: letting $\Omega$ be a fixed domain, we assume that a population (with density $u$) lives inside $\Omega$ and can access some resources $f$, while a second population (wi

Mahdi Nouraie, Connor Smith, Samuel Muller

The Lasso is a prominent algorithm for variable selection. However, its instability in the presence of correlated variables in the high-dimensional setting is well-documented. Although previous research has attempted to address this issue by modifying the Lasso loss function, this paper introduces an approach that simplifies the data processed by Lasso. We p

Daniel Barta, Darya Martyniuk, Johannes Jung, Adrian Paschke

Quantum computing holds immense potential, yet its practical success depends on multiple factors, including advances in quantum circuit design. In this paper, we introduce a generative approach based on denoising diffusion models (DMs) to synthesize parameterized quantum circuits (PQCs). Extending the recent diffusion model pipeline of F\"urrutter et al. [1]

Chaeyoung Jung, Youngjoon Jang, Joon Son Chung

Hallucination remains a major challenge in multimodal large language models (MLLMs). To address this, various contrastive decoding (CD) methods have been proposed that contrasts original logits with hallucinated logits generated from perturbed inputs. While CD has shown promise in vision-language models (VLMs), it is not well-suited for AV-LLMs, where halluc

Yifeng Ma, Jinwei Qi, Chaonan Ji, Peng Zhang

This paper introduces a new control signal for facial motion generation: timeline control. Compared to audio and text signals, timelines provide more fine-grained control, such as generating specific facial motions with precise timing. Users can specify a multi-track timeline of facial actions arranged in temporal intervals, allowing precise control over the

Qifeng Wu, Zhengzhe Liu, Han Zhu, Yizhou Zhao

This paper aims to retrieve proteins with similar structures and semantics from large-scale protein dataset, facilitating the functional interpretation of protein structures derived by structural determination methods like cryo-Electron Microscopy (cryo-EM). Motivated by the recent progress of vision-language models (VLMs), we propose a CLIP-style framework

Si-Jiang Yang, Shan-Ping Wu, Shao-Wen Wei, Yu-Xiao Liu

Black hole thermodynamics is a crucial and foundational aspect of black hole physics, yet its observational verification remains exceptionally challenging. The photon sphere of a black hole, a manifestation of strong gravitational effects, is intrinsically linked to its shadow, which has been directly captured through observations made by the Event Horizon T

David Ceddia, Howard Bondell, Peter Taylor

We conduct a mathematical optimization of the training impulse profile to maximize performance for two seminal athletic performance models: the Banister et al. (1975) Fitness--Fatigue Impulse Response Model and the Busso (2003) Variable Dose--Response Model. We discuss discrepancies in both the quantitative and qualitative aspects of the optimized training i

Funing Li, Yuan Tian, Ruben Noortwyck, Jifeng Zhou

In modern industrial and logistics environments, the rapid expansion of fast delivery services has heightened the demand for storage systems that combine high efficiency with increased density. Multi-deep autonomous vehicle storage and retrieval systems (AVS/RS) present a viable solution for achieving greater storage density. However, these systems encounter

Jason Chui, Hector Andrade-Loarca, Daniel Cremers

Classical Bundle Adjustment (BA) is fundamentally limited by its reliance on precise metric initialization and prior camera intrinsics. While modern dense matchers offer high-fidelity correspondences, traditional Structure-from-Motion (SfM) pipelines struggle to leverage them, as rigid track-building heuristics fail in the presence of their inherent noise. W

Sun-Sig Byun, Yumi Cho, Seungjin Ryu

We establish an optimal Calder\'{o}n-Zygmund theory for nonuniformly elliptic double phase problems with matrix weights. For $1<p<q<\infty$, $a(\cdot)\in C^{0,\alpha}(\Omega)$ ($0<\alpha\le1$), and a symmetric, almost everywhere positive definite matrix weight $\M$ with $|\M(x)|\,|\M(x)^{-1}|\le\Lambda$ for some constant $\Lambda\ge 1$ and small $|\log \M|_{

Bernardo Almeida, Andreia Mordido, Vasco T. Vasconcelos

We address the problem of local type inference for a language based on System F with context-free session types. We present an algorithm that leverages the bidirectional type checking approach to propagate type information, enabling first class polymorphism while addressing the intricacies brought about by the sequential composition operator and type equival

Xin Zhou, Kisub Kim, Ting Zhang, Martin Weyssow

Large Language Models (LLMs) and other automated techniques have been increasingly used to support software developers by generating software artifacts such as code snippets, patches, and comments. However, accurately assessing the correctness of these generated artifacts remains a significant challenge. On one hand, human evaluation provides high accuracy b

Vu Huu Nhu

This work addresses an optimal control problem for a semilinear elliptic equation in two-dimensional space, characterized by an exponential nonlinearity and a singular source term. The source is modeled as a finite linear combination of Dirac measures concentrated at a fixed set of distinct points. The control variable is a finite-dimensional vector whose co

Robin Riblet, Titien Schehr

We highlight a certain compactness of Sidon sets and $B_2[g]$-sets and provide several applications. Notably, we prove the existence of such sets that maximize certain functions. In particular, we show the existence of a Sidon set whose reciprocal sum is equal to the distinct distance constant. We also improve the best known bounds for this constant.

Shucai Yao, Emil Sekerinski

In the shared variable model of concurrency, guarded atomic actions restrict the possible interference between processes by regions of atomic execution. The guard specifies the condition for entering an atomic region. That is a convenient model for the specification and verification of concurrent programs, but has eschewed efficient execution so far. This ar

Mads Rosendahl, Maja H. Kirkeby

Hardware acceleration of algorithms is an effective method for improving performance in high-demand computational tasks. However, developing hardware designs for such acceleration fundamentally differs from software development, as it requires a deep understanding of the highly parallel nature of the hardware architecture. In this paper, we present a framewo

Luís Caires

CLASS is a proof-of-concept general purpose linear programming language, flexibly supporting realistic concurrent programming idioms, and featuring an expressive linear type system ensuring that programs (1) never misuse or leak stateful resources or memory, (2) never deadlock, and (3) always terminate. The design of CLASS and the strong static guarantees of

Zeming Wu, Lu Liu

This paper presents a novel collision avoidance method for general ellipsoids based on control barrier functions (CBFs) and separating hyperplanes. First, collision-free conditions for general ellipsoids are analytically derived using the concept of dual cones. These conditions are incorporated into the CBF framework by extending the system dynamics of contr

Karen Adam, Clémentine Aguet, Patrick Theurillat, Florent Baty

Sleep apnea (SA) is a chronic sleep-related disorder consisting of repetitive pauses or restrictions in airflow during sleep and is known to be a risk factor for cerebro- and cardiovascular disease. It is generally diagnosed using polysomnography (PSG) recorded overnight in an in-lab setting at the hospital. This includes the measurement of blood oxygen satu

Alexander Bohosian, Andrew K. Hirsch

Concurrent programming often entails meticulous pairing of sends and receives between participants to avoid deadlock. Choreographic programming alleviates this burden by specifying the system as a single program. However, there are more applications than implementations of choreographies, and developing new implementations takes a lot of time and effort. Our

Maverick J. Millican, Vassili G. Matsos, Christophe H. Valahu, Tomas Navickas

We demonstrate an optimal quantum control strategy for the deterministic preparation of entangled harmonic oscillator states in trapped ions. The protocol employs dynamical phase modulation of laser-driven Jaynes-Cummings and anti-Jaynes-Cummings interactions. We prepare Two-Mode Squeezed Vacuum (TMSV) states in the mechanical motions of a trapped ion and ch

Rui Peng, Shibo Fang, Pin Ho, Tong Zhou

Synergizing altermagnetism and other ferroic orders, such as ferroelectric switchable altermagnetism [Phys. Rev. Lett. 134, 106801 (2025) and ibid. 106802 (2025)], offers an effective route to achieve nonvolatile switching of altermagnetic spin splitting. In this work, by synergizing altermagnetism and ferroelasticity, we propose the concept of ferroelastic

Alexander Tarasenkov, Kirill Sokolovsky, Alexandr Dodin, Oxana Chernyshenko

We present the discovery of TCP J07222683$+$6220548, a new ultracompact binary system of the AM CVn type. This system was first identified displaying a $\Delta V = 7.6$ mag outburst on 2025-01-20.9416 UTC by the New Milky Way wide-field survey for transients and later independently detected by ASAS-SN and ZTF. The outburst peaked at $V_{\rm max} = 12.45$ and