March 2024 arXiv papers — page 144

Jiun-Man Chen, Yu-Hsuan Chao, Yu-Jie Wang, Ming-Der Shieh

Transformer-based models have gained widespread popularity in both the computer vision (CV) and natural language processing (NLP) fields. However, significant challenges arise during post-training linear quantization, leading to noticeable reductions in inference accuracy. Our study focuses on uncovering the underlying causes of these accuracy drops and prop

Floor Bontje, Arkady Zgonnikov

When a person makes a decision, it is automatically accompanied by a subjective probability judgment of the decision being correct, in other words, a confidence judgment. A better understanding of the mechanisms responsible for these confidence judgments could provide novel insights into human behavior. However, so far confidence judgments have been mostly s

Jiawen Zhu, Guansong Pang

This paper explores the problem of Generalist Anomaly Detection (GAD), aiming to train one single detection model that can generalize to detect anomalies in diverse datasets from different application domains without any further training on the target data. Some recent studies have shown that large pre-trained Visual-Language Models (VLMs) like CLIP have str

Juan Zou, Han Chu, Yizhang Xia, Junwen Xu

The effectiveness of Evolutionary Neural Architecture Search (ENAS) is influenced by the design of the search space. Nevertheless, common methods including the global search space, scalable search space and hierarchical search space have certain limitations. Specifically, the global search space requires a significant amount of computational resources and ti

Toru Araki

A subset $S$ of vertices in a graph $G$ is a secure dominating set of $G$ if $S$ is a dominating set of $G$ and, for each vertex $u \not\in S$, there is a vertex $v \in S$ such that $uv$ is an edge and $(S \setminus \{v\}) \cup \{u\}$ is also a dominating set of $G$. The secure domination number of $G$, denoted by $\gamma_{s}(G)$, is the cardinality of a sma

Pham Truong Xuan, Tran Van Thuy, Nguyen Thi Van

In this article we investigate the existence, uniqueness and exponential decay of asymptotically almost periodic solutions of the parabolic-elliptic Keller-Segel system on a real hyperbolic manifold. We prove the existence and uniqueness of such solutions in the linear equation case by using the dispersive and smoothing estimates of the heat semigroup. Then

Xiaomin Cui, Shaojie Hu, Yohei Hidaka, Satoshi Yakata

We investigate the gyrotropic motion of the magnetic vortex core in a chain of a few micron-sized Permalloy disks by electrical resistance measurement with amplitude-modulated magnetic field. We observe a distinctive splitting of the resistance peak due to the resonant vortex-core motion under heightened radio frequency (RF) magnetic field excitation. Our mi

Bernold Fiedler, Hannes Stuke

In parabolic or hyperbolic PDEs, solutions which remain uniformly bounded for all real times $t=r\in\mathbb{R}$ are often called PDE entire or eternal. For example, consider the quadratic parabolic PDE \begin{equation*} \label{*} w_t=w_{xx}+6w^2-\lambda, \tag{*} \end{equation*} for $0<x<\tfrac{1}{2}$, under Neumann boundary conditions. By its gradient-like s

Dingyuan Zhu, Daixin Wang, Zhiqiang Zhang, Kun Kuang

Uplift modeling aims to measure the incremental effect, which we call uplift, of a strategy or action on the users from randomized experiments or observational data. Most existing uplift methods only use individual data, which are usually not informative enough to capture the unobserved and complex hidden factors regarding the uplift. Furthermore, uplift mod

Runmin Cong, Hang Xiong, Jinpeng Chen, Wei Zhang

Previous Few-Shot Segmentation (FSS) approaches exclusively utilize support features for prototype generation, neglecting the specific requirements of the query. To address this, we present the Query-guided Prototype Evolution Network (QPENet), a new method that integrates query features into the generation process of foreground and background prototypes, th

Koji Inoue, Bing'er Jiang, Erik Ekstedt, Tatsuya Kawahara

This paper investigates the application of voice activity projection (VAP), a predictive turn-taking model for spoken dialogue, on multilingual data, encompassing English, Mandarin, and Japanese. The VAP model continuously predicts the upcoming voice activities of participants in dyadic dialogue, leveraging a cross-attention Transformer to capture the dynami

Carlo Silvestri, Massimo Brambilla, Paolo Bardella, Lorenzo Luigi Columbo

We present a unified model to describe the dynamics of optical frequency combs (OFCs) in quantum cascade lasers (QCLs), incorporating both ring and Fabry-P\'erot (FP) cavity configurations. The model derives a modified complex Ginzburg-Landau equation (CGLE), leveraging an order parameter approach and is capable of capturing the dynamics of both configuratio

Jinxi Kuang, Jinyang Liu, Junjie Huang, Renyi Zhong

Due to the scale and complexity of cloud systems, a system failure would trigger an "alert storm", i.e., massive correlated alerts. Although these alerts can be traced back to a few root causes, the overwhelming number makes it infeasible for manual handling. Alert aggregation is thus critical to help engineers concentrate on the root cause and facilitate fa

Paris Avgeriou, Ipek Ozkaya, Alexander Chatzigeorgiou, Marcus Ciolkowski

Technical Debt, considered by many to be the 'silent killer' of software projects, has undeniably become part of the everyday vocabulary of software engineers. We know it compromises the internal quality of a system, either deliberately or inadvertently. We understand Technical Debt is not all derogatory, often serving the purpose of expediency. But, it is a

Yongchuan Tang, Rongfei Li

Negation is an important perspective of knowledge representation. Existing negation methods are mainly applied in probability theory, evidence theory and complex evidence theory. As a generalization of evidence theory, random permutation sets theory may represent information more precisely. However, how to apply the concept of negation to random permutation

Daixin Wang, Zhiqiang Zhang, Yeyu Zhao, Kai Huang

User financial default prediction plays a critical role in credit risk forecasting and management. It aims at predicting the probability that the user will fail to make the repayments in the future. Previous methods mainly extract a set of user individual features regarding his own profiles and behaviors and build a binary-classification model to make defaul

Garry Goldstein

In this work we write down a classical (not quantum) action for the surface height for the Kardar-Parisi-Zhang (KPZ) equation for surface growth. We do so starting with the regular Martin-Siggia-Rose (MSR) action (which is quantum - contains the constraint field) and integrate out the quantum constraint field exactly. We analyze the classical action, we ther

Rostislav Grigorchuk, Ville Salo

We study symbolic dynamical representations of actions of the first Grigorchuk group $G$, namely its action on the boundary of the infinite rooted binary tree, its representation in the topological full group of a minimal substitutive $\mathbb{Z}$-shift, and its representation as a minimal system of Schreier graphs. We show that the first system admits an SF

Jiaxin Guo, Jiangliu Wang, Zhaoshuo Li, Tongyu Jia

Soft tissue tracking is crucial for computer-assisted interventions. Existing approaches mainly rely on extracting discriminative features from the template and videos to recover corresponding matches. However, it is difficult to adopt these techniques in surgical scenes, where tissues are changing in shape and appearance throughout the surgery. To address t

Hidetsugu Sakaguchi, Keito Yamasaki

We study two simple mathematical models of the epidemic. At first, we study the repetitive infection spreading in a simplified SIRS model including the effect of the decay of the acquired immune. The model is an intermediate model of the SIRS model including the recruitment and death terms and the SIR model in which the recovered population is assumed to be

Seongyeon Park, Junguk Hong, Jaeyong Song, Hajin Kim

With the advance in genome sequencing technology, the lengths of deoxyribonucleic acid (DNA) sequencing results are rapidly increasing at lower prices than ever. However, the longer lengths come at the cost of a heavy computational burden on aligning them. For example, aligning sequences to a human reference genome can take tens or even hundreds of hours. Th

Arup Majumdar, P. Sam Johnson, Ram N. Mohapatra

In this paper, we discuss the Hyers-Ulam stability of closable (unbounded) operators with several interesting examples. We also present results pertaining to the Hyers-Ulam stability of the sum and product of closable operators to have the Hyers-Ulam stability and the necessary and sufficient conditions of the Schur complement and the quadratic complement of

Zhuo Wang, Xiaobo He, Fangjun Lu, Hai Zeng

Motivated by the recent observations of electronic correlation effect [M. Corasaniti \textit{et al}., Phys. Rev. B \textbf{104}, L121112 (2021)] and topology-stabilized magnetic fluctuations [N. Drucker \textit{et al}., Nat. Commun. \textbf{14}, 5182 (2023)] in the noncentrosymmetric magnetic Weyl semimetal candidate CeAlGe, we performed systematic studies o

Lixia Yuan, Ji Yang, Xuepeng Chen, Yang Su

Velocity fields of molecular clouds (MCs) can provide crucial information on the merger and split between clouds, as well as their internal kinematics and maintenance, energy injection and redistribution, even star formation within clouds. Using the CO spectral lines data from the Milky Way Imaging Scroll Painting (MWISP) survey, we measure the relative velo

Yinyan Liu, Yi Wang, Jin Ma

Non-Intrusive Load Monitoring (NILM) is pivotal in today's energy landscape, offering vital solutions for energy conservation and efficient management. Its growing importance in enhancing energy savings and understanding consumer behavior makes it a pivotal technology for addressing global energy challenges. This paper delivers an in-depth review of NILM, hi

I. A. Shaposhnikov, A. M. Cherepashchuk, A. V. Dodin, K. A. Postnov

We aim at searching for secular evolution of the orbital period in the short-period binary system WR 127 (WN3b+O9.5V, $P = 9.555^d$). We performed new low-resolution spectroscopic observations of WR 127 on 2.5-m CMO SAI telescope to construct the radial velocity curves of the components suggesting the component masses $M_\mathrm{WR}\sin^3(i) = 11.8\pm1.4$ $M

Tomohiro Tamaya, Takeo Kato, Takahiro Misawa

Spin current, a key concept in spintronics that carries spin angular momentum, has a non-unique definition due to the non-conservation of spins in solids. While two primary definitions exist -- conventional spin current and conserved spin current -- their validity has not been quantitatively examined. Here, we examine the validity of these definitions of spi

Jun-Young Oh, In-Gyu Lee, Tae-Eui Kam, Ji-Hoon Jeong

In the cutting-edge domain of medical artificial intelligence (AI), remarkable advances have been achieved in areas such as diagnosis, prediction, and therapeutic interventions. Despite these advances, the technology for image segmentation faces the significant barrier of having to produce extensively annotated datasets. To address this challenge, few-shot s

Bo Li, Yi-ke Li, Zhi-fen He, Bin Liu

3D-consistent image generation from a single 2D semantic label is an important and challenging research topic in computer graphics and computer vision. Although some related works have made great progress in this field, most of the existing methods suffer from poor disentanglement performance of shape and appearance, and lack multi-modal control. In this pap

Chao-Qiang Geng, Chia-Wei Liu, Sheng-Lin Liu

We study the nonleptonic three-body charmed baryon weak decays of $\mathbf{B}_{c}\rightarrow\mathbf{B}_{n}PP^{\prime}$ under the $SU(3)_{F}$ flavor symmetry, where $\mathbf{B}_{c}$ denotes the anti-triplet charmed baryon, comprising $(\Xi^{0}_{c},-\Xi^{+}_{c},\Lambda^{+}_{c})$, and $\mathbf{B}_{n}$ and $P(P^{\prime})$ represent octet baryon and pseudoscalar

Velmurugan S

The algebra of symmetric functions contains several interesting families of symmetric functions indexed by integer partitions or skew partitions. Given a sequence $\{u_n\}$ of symmetric functions taken from one of these families such that $u_n$ is homogeneous of degree $n$, we provide necessary and sufficient conditions for the sequence to form a system of a

Jiuming Liu, Ruiji Yu, Yian Wang, Yu Zheng

Recently, state space model (SSM) has gained great attention due to its promising performance, linear complexity, and long sequence modeling ability in both language and image domains. However, it is non-trivial to extend SSM to the point cloud field, because of the causality requirement of SSM and the disorder and irregularity nature of point clouds. In thi

Yingzhuo Liu

Multiple Line Bus Scheduling Problem (MLBSP) is vital to save operational cost of bus company and guarantee service quality for passengers. Existing approaches typically generate a bus scheduling scheme in an offline manner and then schedule buses according to the scheme. In practice, uncertain events such as traffic congestion occur frequently, which may ma

Jianxun Lian, Yuxuan Lei, Xu Huang, Jing Yao

This paper introduces RecAI, a practical toolkit designed to augment or even revolutionize recommender systems with the advanced capabilities of Large Language Models (LLMs). RecAI provides a suite of tools, including Recommender AI Agent, Recommendation-oriented Language Models, Knowledge Plugin, RecExplainer, and Evaluator, to facilitate the integration of

Noureddine Igbida

This work introduces a new class of cross-diffusion systems for studying overcrowding dispersal of two species. The approach, based on proximal minimization energy through a minimum flow process, offers a potential generalization of existing segregation models. Unlike prior methods using PDEs or $W_2$-Wasserstein flows, it establishes a well-posed PDE framew

Xiaolei Wang, Chen Yang, Yuzhen Feng, Luohan Hu

For on-demand dynamic ride-pooling services, e.g., Uber Pool and Didi Pinche, a well-designed vehicle dispatching strategy is crucial for platform profitability and passenger experience. Most existing dispatching strategies overlook incoming pairing opportunities, therefore suffer from short-sighted limitations. In this paper, we propose a forward-looking ve

Xiaoyang Wang, Huihui Bai, Limin Yu, Yao Zhao

Semi-supervised semantic segmentation allows model to mine effective supervision from unlabeled data to complement label-guided training. Recent research has primarily focused on consistency regularization techniques, exploring perturbation-invariant training at both the image and feature levels. In this work, we proposed a novel feature-level consistency le

Haozhi Cao, Yuecong Xu, Pengyu Yin, Xingyu Ji

Multi-modal test-time adaptation (MM-TTA) adapts models to an unlabeled target domain by leveraging the complementary multi-modal inputs in an online manner. While previous MM-TTA methods for 3D segmentation offer a promising solution by leveraging self-refinement per frame, they suffer from two major limitations: 1) unstable frame-wise predictions caused by

Han Yan, Hua Chen, Wei Liu, Songjie Yang

Reconfigurable Intelligent Surfaces (RIS) show great promise in the realm of 6th generation (6G) wireless systems, particularly in the areas of localization and communication. Their cost-effectiveness and energy efficiency enable the integration of numerous passive and reflective elements, enabling near-field propagation. In this paper, we tackle the challen

Yuxiang Lai, Xiaoxi Chen, Angtian Wang, Alan Yuille

AI for cancer detection encounters the bottleneck of data scarcity, annotation difficulty, and low prevalence of early tumors. Tumor synthesis seeks to create artificial tumors in medical images, which can greatly diversify the data and annotations for AI training. However, current tumor synthesis approaches are not applicable across different organs due to

Derk Rembold, Bernd Stauss, Stefan Schwarzkopf

Many physical target values in technical processes are error-prone, cumbersome, or expensive to measure automatically. One example of a physical target value is the wort density, which is an important value needed for beer production. This article introduces a system that helps the brewer measure wort density through sensors in order to reduce errors in manu

Haoru Tan, Chuang Wang, Sitong Wu, Xu-Yao Zhang

Graph matching is a commonly used technique in computer vision and pattern recognition. Recent data-driven approaches have improved the graph matching accuracy remarkably, whereas some traditional algorithm-based methods are more robust to feature noises, outlier nodes, and global transformation (e.g.~rotation). In this paper, we propose a graph neural netwo

Abdullah Al-Mamun, Hao Wu, Qiyang He, Jianguo Wang

A recent research trend involves treating database index structures as Machine Learning (ML) models. In this domain, single or multiple ML models are trained to learn the mapping from keys to positions inside a data set. This class of indexes is known as "Learned Indexes." Learned indexes have demonstrated improved search performance and reduced space requir

Y. K. Wang, P. W. Zhao, J. Meng

The neutrinoless double-$\beta$ ($0\nu\beta\beta$) decay and the double Gamow-Teller (DGT) transition are investigated with the state-of-the-art Relativistic Configuration-interaction Density functional theory. A strong linear correlation between the nuclear matrix elements (NMEs) of the $0\nu\beta\beta$ decay and the DGT transition is demonstrated. This lin

Bingqiao Luo

Utilizing graph analytics and learning has proven to be an effective method for exploring aspects of crypto economics such as network effects, decentralization, tokenomics, and fraud detection. However, the majority of existing research predominantly focuses on leading cryptocurrencies, namely Bitcoin (BTC) and Ethereum (ETH), overlooking the vast diversity

Yuki Tatsukawa, I-Chao Shen, Anran Qi, Yuki Koyama

Acquiring the desired font for various design tasks can be challenging and requires professional typographic knowledge. While previous font retrieval or generation works have alleviated some of these difficulties, they often lack support for multiple languages and semantic attributes beyond the training data domains. To solve this problem, we present FontCLI

Guangyang Wu, Xiaohong Liu, Jun Jia, Xuehao Cui

In the digital era, QR codes serve as a linchpin connecting virtual and physical realms. Their pervasive integration across various applications highlights the demand for aesthetically pleasing codes without compromised scannability. However, prevailing methods grapple with the intrinsic challenge of balancing customization and scannability. Notably, stable-

E. Génetay Johansen, C. Vale, T. Simula

The theory of topological quantum computation is underpinned by two important classes of models. One is based on non-abelian Chern-Simons theory, which yields the so-called $\rm{SU}(2)_k$ anyon models that often appear in the context of electrically charged quantum fluids. The physics of the other is captured by symmetry broken Yang-Mills theory in the absen

Dong-Xu Liu, Wei Xu, Xue-Feng Zhang

In the quantum Monte Carlo (QMC) method, the Pseudo-Random Number Generator (PRNG) plays a crucial role in determining the computation time. However, the hidden structure of the PRNG may lead to serious issues such as the breakdown of the Markov process. Here, we systematically analyze the performance of the different PRNGs on the widely used QMC method -- s

Wanwan Zhang

In this paper, we revisit the problem of finite-time blowup for a multi-dimensional nonlocal transport equation studied in [Dong, Adv. Math. 264 (2014) 747-761]. Inspired by a one-dimensional analogous model considered in [Li-Rodrigo, Adv. Math. 374 (2020) 1-26], we establish a new weighted nonlinear inequality implying the blow-up by a completely real varia

Weihang Su, Changyue Wang, Qingyao Ai, Yiran HU

Hallucinations in large language models (LLMs) refer to the phenomenon of LLMs producing responses that are coherent yet factually inaccurate. This issue undermines the effectiveness of LLMs in practical applications, necessitating research into detecting and mitigating hallucinations of LLMs. Previous studies have mainly concentrated on post-processing tech

Junda Wu, Cheng-Chun Chang, Tong Yu, Zhankui He

The long-tail recommendation is a challenging task for traditional recommender systems, due to data sparsity and data imbalance issues. The recent development of large language models (LLMs) has shown their abilities in complex reasoning, which can help to deduce users' preferences based on very few previous interactions. However, since most LLM-based system

Abhisek Bandyopadhyay, S. Lee, D. T. Adroja, G. B. G. Stenning

Frustrated magnets offer a plethora of exotic magnetic ground states, including quantum spin liquids (QSLs), in which enhanced quantum fluctuations prevent a long-range magnetic ordering of the strongly correlated spins down to lowest temperature. Here we have investigated the trimer based mixed valence hexagonal rhodate Ba$_4$NbRh$_3$O$_{12}$ using a combin

Qian Chen, Xingjian Dong, Zhike Peng

Neural networks, with powerful nonlinear mapping and classification capabilities, are widely applied in mechanical fault diagnosis to ensure safety. However, being typical black-box models, their application is limited in high-reliability-required scenarios. To understand the classification logic and explain what typical fault signals look like, the prototyp

Alexander Vasilyev, Vladimir Vasilyev, Abu Bakarr Kamanda Bongay

We consider linear bounded operators acting in Banach spaces with a basis, such operators can be represented by an infinite matrix. We prove that for an invertible operator there exists a sequence of invertible finite-dimensional operators so that the family of norms of their inverses is uniformly bounded. It leads to the fact that solutions of finite-dimens

Guobao Xiao, Jun Yu, Jiayi Ma, Deng-Ping Fan

Estimating reliable geometric model parameters from the data with severe outliers is a fundamental and important task in computer vision. This paper attempts to sample high-quality subsets and select model instances to estimate parameters in the multi-structural data. To address this, we propose an effective method called Latent Semantic Consensus (LSC). The

Yuhan Bao, Lei Sun, Yuqin Ma, Kaiwei Wang

Event cameras, or Dynamic Vision Sensors (DVS) are novel neuromorphic sensors that capture brightness changes as a continuous stream of "events" rather than traditional intensity frames. Converting sparse events to dense intensity frames faithfully has long been an ill-posed problem. Previous methods have primarily focused on converting events to video in dy

Yisen Wang, Liang Huang

As the essential cause of the intrinsic dissipation that limits the quality of graphene nanoresonators, intermodal energy transfer is also a key issue in thermalization dynamics. Typically systems with larger initial energy demand shorter time to be thermalized. However, we find quantitatively that instead of becoming shorter, the equipartition time of the g

S. V. Talalov

In this study, the problem of the energy spectrum of a quantum vortex loop moving in a thin long pipe is solved for the first time. We quantize this dynamic system using a new method, which leads to non-trivial results for circulation $\Gamma$ and energy values $E$. It is shown that the spectrum has a quasi-continuous fractal structure. In the final form, we

Dong Chen, Shuai Zheng, Muhao Xu, Zhenfeng Zhu

In the domain of dynamic graph representation learning (DGRL), the efficient and comprehensive capture of temporal evolution within real-world networks is crucial. Spiking Neural Networks (SNNs), known as their temporal dynamics and low-power characteristic, offer an efficient solution for temporal processing in DGRL task. However, owing to the spike-based i

Ming-Xiao Duan

In this study, we analyze the cross-section data from the $e^+e^-\to \gamma J/\psi\omega$ process to explore both short-distance and long-distance interactions for the radiative transition $Y(4200)\to \gamma X(3872)$. We investigate the short-distance effects through the E1 transition among the $c\bar{c}$ components, and the long-distance effects via hadroni

Qianwan Yang, Ruipeng Guo, Guorong Hu, Yujia Xue

Traditional fluorescence microscopy is constrained by inherent trade-offs among resolution, field-of-view, and system complexity. To navigate these challenges, we introduce a simple and low-cost computational multi-aperture miniature microscope, utilizing a microlens array for single-shot wide-field, high-resolution imaging. Addressing the challenges posed b

Yingbo Hua, Md Saydur Rahman

This paper presents further insights into a recently developed round-trip communication scheme called ``Secret-message Transmission by Echoing Encrypted Probes (STEEP)''. A legitimate wireless channel between a multi-antenna user (Alice) and a single-antenna user (Bob) in the presence of a multi-antenna eavesdropper (Eve) is focused on. STEEP does not requir

Duho Kim, Yun-Kyeong Sheen, Yara L. Jaffé, Kshitija Kelkar

We study the incidence and spatial distribution of galaxies that are currently undergoing gravitational merging (M) or that have signs of a post merger (PM) in six galaxy clusters (A754, A2399, A2670, A3558, A3562, and A3716) within the redshift range, 0.05$\lesssim$$z$$\lesssim$0.08. To this aim, we obtained Dark Energy Camera (DECam) mosaics in $u^{\prime}

Mohammad Khairul Bashar, Abir Hasan, Nikhil Shukla

Probabilistic bit (p-bit)-based compute engines utilize the unique capability of a p-bit to probabilistically switch between two states to solve computationally challenging problems. However, when solving problems that require more than two states (e.g., problems such as Max-3-Cut, verifying if a graph is K-partite (K>2) etc.), additional pre-processing step

Cristian Ramirez Rodriguez

Health experts have suggested that social distancing measures are one of the most effective ways of preventing the spread of Covid-19. Research primarily focused on large Covid filled droplets suggested that these droplets can move further than regulated social distancing guidelines (2 meters apart) in the presence of wind. This project aims to model the pat

Huahang Li, Shuangyin Li, Fei Hao, Chen Jason Zhang

Entity resolution, which involves identifying and merging records that refer to the same real-world entity, is a crucial task in areas like Web data integration. This importance is underscored by the presence of numerous duplicated and multi-version data resources on the Web. However, achieving high-quality entity resolution typically demands significant eff

Konyul Park, Yecheol Kim, Junho Koh, Byungwoo Park

Developing high-performance, real-time architectures for LiDAR-based 3D object detectors is essential for the successful commercialization of autonomous vehicles. Pillar-based methods stand out as a practical choice for onboard deployment due to their computational efficiency. However, despite their efficiency, these methods can sometimes underperform compar

Zhiyang Lu, Qinghan Chen, Ming Cheng

Scene flow prediction is a crucial underlying task in understanding dynamic scenes as it offers fundamental motion information. However, contemporary scene flow methods encounter three major challenges. Firstly, flow estimation solely based on local receptive fields lacks long-dependency matching of point pairs. To address this issue, we propose global atten

Jungwon Choi, Hyungi Lee, Byung-Hoon Kim, Juho Lee

Graph Neural Networks (GNNs) have shown promise in learning dynamic functional connectivity for distinguishing phenotypes from human brain networks. However, obtaining extensive labeled clinical data for training is often resource-intensive, making practical application difficult. Leveraging unlabeled data thus becomes crucial for representation learning in

Ned Cooper, Alex Zafiroglu

Participatory machine learning (ML) encourages the inclusion of end users and people affected by ML systems in design and development processes. We interviewed 18 participation brokers -- individuals who facilitate such inclusion and transform the products of participants' labour into inputs for an ML artefact or system -- across a range of organisational se

Zhenbo Song, Wenhao Gao, Zhenyuan Zhang, Jianfeng Lu

Deep learning-based face restoration models, increasingly prevalent in smart devices, have become targets for sophisticated backdoor attacks. These attacks, through subtle trigger injection into input face images, can lead to unexpected restoration outcomes. Unlike conventional methods focused on classification tasks, our approach introduces a unique degrada

Trung Thanh Nguyen, Kai Hollander, Marius Hoggenmueller, Callum Parker

Recent studies have investigated new approaches for communicating an autonomous vehicle's (AV) intent and awareness to pedestrians. This paper adds to this body of work by presenting the design and evaluation of in-situ projections on the road. Our design combines common traffic light patterns with aesthetic visual elements. We describe the iterative design

Kshitiz Aryal, Maanak Gupta, Mahmoud Abdelsalam, Moustafa Saleh

Windows malware is predominantly available in cyberspace and is a prime target for deliberate adversarial evasion attacks. Although researchers have investigated the adversarial malware attack problem, a multitude of important questions remain unanswered, including (a) Are the existing techniques to inject adversarial perturbations in Windows Portable Execut

David Garfinkle, James Isenberg, Dan Knopf, Haotian Wu

Kr\"oncke has shown that the Fubini-Study metric is an unstable generalized stationary solution of Ricci flow [Kr\"o20]. In this paper, we carry out numerical simulations which indicate that Ricci flow solutions originating at unstable perturbations of the Fubini-Study metric develop local singularities modeled by the blowdown soliton discovered in [FIK03].

Daisy Q. Wang, Zeb Krix, Oleg P. Sushkov, Ian Farrer

In nearly free electron theory the imposition of a periodic electrostatic potential on free electrons creates the bandstructure of a material, determined by the crystal lattice spacing and geometry. Imposing an artificially designed potential to the electrons confined in a GaAs quantum well makes it possible to engineer synthetic two-dimensional band structu

Yazheng Liu, Xi Zhang, Sihong Xie

Graphs are ubiquitous in social networks and biochemistry, where Graph Neural Networks (GNN) are the state-of-the-art models for prediction. Graphs can be evolving and it is vital to formally model and understand how a trained GNN responds to graph evolution. We propose a smooth parameterization of the GNN predicted distributions using axiomatic attribution,

Ziliang Samuel Zhong, Xiang Pan, Qi Lei

Multi-source domain adaptation aims to reduce performance degradation when applying machine learning models to unseen domains. A fundamental challenge is devising the optimal strategy for feature selection. Existing literature is somewhat paradoxical: some advocate for learning invariant features from source domains, while others favor more diverse features.

Zeyang Jia, Kosuke Imai, Michael Lingzhi Li

We introduce the cram method as a general statistical framework for evaluating the final learned policy from a multi-armed contextual bandit algorithm, using the dataset generated by the same bandit algorithm. The proposed on-policy evaluation methodology differs from most existing methods that focus on off-policy performance evaluation of contextual bandit

Guanhua Ding, Jianan Liu, Yuxuan Xia, Tao Huang

Multiple extended target tracking (ETT) has gained increasing attention due to the development of high-precision LiDAR and radar sensors in automotive applications. For LiDAR point cloud-based vehicle tracking, this paper presents a probabilistic measurement-region association (PMRA) ETT model, which can describe the complex measurement distribution by parti

Yu-Bin Wang, Xia Zhou, Abdusattar Kurban, Fa-Yin Wang

Exploring the concept of a massive photon has been an important area in astronomy and physics. If photons have mass, their propagation in nonvacuum space would be affected by both the nonzero mass $m_{\gamma}$ and the presence of a plasma medium. This would lead to a delay time proportional to $m_{\gamma}^2\nu^{-4}$, which deviates from the classical dispers

Weixia Zhang, Chengguang Zhu, Jingnan Gao, Yichao Yan

The rapid advancement of Artificial Intelligence Generated Content (AIGC) technology has propelled audio-driven talking head generation, gaining considerable research attention for practical applications. However, performance evaluation research lags behind the development of talking head generation techniques. Existing literature relies on heuristic quantit

On Tai Wu, Frodo Kin Sun Chan, Zunhao Zhang, Yan Nei Law

Few-shot prompting and step-by-step reasoning have enhanced the capabilities of Large Language Models (LLMs) in tackling complex tasks including code generation. In this paper, we introduce a prompt selection and augmentation algorithm aimed at improving mathematical reasoning and robot arm operations. Our approach incorporates a multi-stage example augmenta

Liangliang Chen, Yutian Lei, Shiyu Jin, Ying Zhang

Reinforcement learning (RL) has demonstrated its capability in solving various tasks but is notorious for its low sample efficiency. In this paper, we propose RLingua, a framework that can leverage the internal knowledge of large language models (LLMs) to reduce the sample complexity of RL in robotic manipulations. To this end, we first present a method for

Yukun Song, Dayuan Cao, Jiali Miao, Shuai Yang

Multi-label feature selection serves as an effective mean for dealing with high-dimensional multi-label data. To achieve satisfactory performance, existing methods for multi-label feature selection often require the centralization of substantial data from multiple sources. However, in Federated setting, centralizing data from all sources and merging them int

Dapeng Wang, Zhaojian Liang, Ziqi Zhang, Mengying Li

Convective heat transfer is crucial for photovoltaic (PV) systems, as the power generation of PV is sensitive to temperature. The configuration of PV arrays have a significant impact on convective heat transfer by influencing turbulent characteristics. Conventional methods of quantifying the configuration effects are either through Computational Fluid Dynami

Shengji Tang, Weihao Lin, Hancheng Ye, Peng Ye

Sparsification-based pruning has been an important category in model compression. Existing methods commonly set sparsity-inducing penalty terms to suppress the importance of dropped weights, which is regarded as the suppressed sparsification paradigm. However, this paradigm inactivates the dropped parts of networks causing capacity damage before pruning, the

Yihan Wang, Bing Zhang

The origin of supermassive black holes (SMBHs) residing in the centers of most galaxies remains a mystery. The Event Horizon Telescope (EHT) provided direct imaging of the SMBH Sgr A* at the Milky Way's center, indicating it likely spins rapidly with its spin axis significantly misaligned relative to the Galactic plane's angular momentum. Through investigati

Martin Kreuzer, Lorenzo Robbiano

Let $K$ be a field and $P=K[x_1,\dots,x_n]$. The technique of elimination by substitution is based on discovering a coherently $Z=(z_1,\dots,z_s)$-separating tuple of polynomials $(f_1,\dots,f_s)$ in an ideal $I$, i.e., on finding polynomials such that $f_i = z_i - h_i$ with $h_i \in K[X \setminus Z]$. Here we elaborate on this technique in the case when $P$

Chengyuan Liu, Yangyang Kang, Fubang Zhao, Kun Kuang

Large language models (LLMs) have demonstrated remarkable capabilities across various NLP tasks. However, their computational costs are prohibitively high. To address this issue, previous research has attempted to distill the knowledge of LLMs into smaller models by generating annotated data. Nonetheless, these works have mainly focused on the direct use of

Ruhan Zhao, Lifang Zhou

We completely characterize $L^p-L^q$ boundedness of two classes of Forelli-Rudin type operators on the unit ball of $\mathbb{C}^n$ for all $(p, q)\in [1, \infty]\times [1, \infty]$. The results are not only a complement to some previous results on Forelli-Rudin type operators by Kures and Zhu in 2006 and the first author in 2015, but also a high dimension ex

Eunsu Kim, Juyoung Suk, Philhoon Oh, Haneul Yoo

Despite the rapid development of large language models (LLMs) for the Korean language, there remains an obvious lack of benchmark datasets that test the requisite Korean cultural and linguistic knowledge. Because many existing Korean benchmark datasets are derived from the English counterparts through translation, they often overlook the different cultural c

Tamara Sopek, Peter Jacobs, Suria-Devi Subiah, Peter Collen

Pitot pressure is the most common measurement in high total enthalpy shock tunnels for test condition verification. Nozzle calculations using multi-temperature non-equilibrium thermochemistry are needed in conjunction with Pitot measurements to quantify freestream properties. Pitot pressure is typically matched by tuning the boundary layer transition locatio

Li Yuan, Yi Cai, Haopeng Ren, Jiexin Wang

Generating coherent and credible explanations remains a significant challenge in the field of AI. In recent years, researchers have delved into the utilization of entailment trees to depict explanations, which exhibit a reasoning process of how a hypothesis is deduced from the supporting facts. However, existing models often overlook the importance of genera

Bryce Allen Bagley, Claudia K Petritsch

The rapid advancement in neurotechnology in recent years has created an emerging critical intersection between neurotechnology and security. Implantable devices, non-invasive monitoring, and non-invasive therapies all carry with them the prospect of violating the privacy and autonomy of individuals' cognition. A growing number of scientists and physicians ha

Bovornpratch Vijarnwannaluk, Masayuki Akiyama, Malte Schramm, Yoshihiro Ueda

We investigated the gas obscuration and host galaxy properties of active galactic nuclei (AGN) during the peak of cosmic accretion growth of supermassive black holes (SMBHs) at redshift 0.8-1.8 using X-ray detected AGN with mid-infrared and far-infrared detection. The sample was classified as type-1 and type-2 AGN using optical spectral and morphological cla

Zhuocheng Gong, Jiahao Liu, Jingang Wang, Xunliang Cai

Quantization has emerged as a promising technique for improving the memory and computational efficiency of large language models (LLMs). Though the trade-off between performance and efficiency is well-known, there is still much to be learned about the relationship between quantization and LLM performance. To shed light on this relationship, we propose a new

Xin Lin, Tianhuang Su, Zhenya Huang, Shangzi Xue

Knowledge-based question answering (KBQA) is a key task in NLP research, and also an approach to access the web data and knowledge, which requires exploiting knowledge graphs (KGs) for reasoning. In the literature, one promising solution for KBQA is to incorporate the pretrained language model (LM) with KGs by generating KG-centered pretraining corpus, which

Jiawei Chen, Yue Jiang, Dingkang Yang, Mingcheng Li

While Large Language Models (LLMs) excel in world knowledge understanding, adapting them to specific subfields requires precise adjustments. Due to the model's vast scale, traditional global fine-tuning methods for large models can be computationally expensive and impact generalization. To address this challenge, a range of innovative Parameters-Efficient Fi

Weixia Zhang, Dingquan Li, Guangtao Zhai, Xiaokang Yang

Contemporary no-reference image quality assessment (NR-IQA) models can effectively quantify perceived image quality, often achieving strong correlations with human perceptual scores on standard IQA benchmarks. Yet, limited efforts have been devoted to treating NR-IQA models as natural image priors for real-world image enhancement, and consequently comparing