October 2023 arXiv papers — page 108

Zhiguang Fan, Yuedong Yang, Mingyuan Xu, Hongming Chen

The embedding of Biomedical Knowledge Graphs (BKGs) generates robust representations, valuable for a variety of artificial intelligence applications, including predicting drug combinations and reasoning disease-drug relationships. Meanwhile, contrastive learning (CL) is widely employed to enhance the distinctiveness of these representations. However, constru

Haotian Zhou, Tingkai Liu, Qianli Ma, Yufeng Zhang

We introduce DavIR, a model-based data selection method for post-training Large Language Models. DavIR generalizes Reducible Holdout Loss to core-set selection problem of causal language modeling, and quantifies the learnability of a given datum with respect to a pre-trained LLM based on relative reduction in loss during fine-tuning, a metric we show to be c

Simon Muntwiler, Ognjen Stanojev, Andrea Zanelli, Gabriela Hug

This paper presents a novel model order reduction technique tailored for power systems with a large share of inverter-based energy resources. Such systems exhibit an increased level of dynamic stiffness compared to traditional power systems, posing challenges for time-domain simulations and control design. Our approach involves rotation of the coordinate sys

David Chocholatý, Tomáš Fiedor, Vojtěch Havlena, Lukáš Holík

Mata is a well-engineered automata library written in C++ that offers a unique combination of speed and simplicity. It is meant to serve in applications such as string constraint solving and reasoning about regular expressions, and as a~reference implementation of automata algorithms. Besides basic algorithms for (non)deterministic automata, it implements a

Naoya Hatano, Ryota Kawasumi, Hiroki Saito, Hitoshi Tanaka

Let $H^d$, $0<d<n$, be the dyadic Hausdorff content of the $n$-dimensional Euclidean space ${\mathbb R}^n$. It is shown that $H^d$ counts a~Cantor set of the unit cube $[0, 1)^n$ as $\approx 1$, which implies unboundedness of the sparse operator ${\mathcal A}_{{\mathcal S}}$ on the Choquet space ${\mathcal L}^p(H^d)$, $p>0$. In this paper we verify that the

Yangyang Guo, Guangzhi Wang, Mohan Kankanhalli

Applying a pre-trained large model to downstream tasks is prohibitive under resource-constrained conditions. Recent dominant approaches for addressing efficiency issues involve adding a few learnable parameters to the fixed backbone model. This strategy, however, leads to more challenges in loading large models for downstream fine-tuning with limited resourc

Bodhisattwa Prasad Majumder, Bhavana Dalvi Mishra, Peter Jansen, Oyvind Tafjord

Language agents have shown some ability to interact with an external environment, e.g., a virtual world such as ScienceWorld, to perform complex tasks, e.g., growing a plant, without the startup costs of reinforcement learning. However, despite their zero-shot capabilities, these agents to date do not continually improve over time beyond performance refineme

Ramya Burra, Anshoo Tandon, Srishti Mittal

This paper aims to develop an efficient open-source Secure Multi-Party Computation (SMPC) repository, that addresses the issue of practical and scalable implementation of SMPC protocol on machines with moderate computational resources, while aiming to reduce the execution time. We implement the ABY2.0 protocol for SMPC, providing developers with effective to

Chen-Huan Wu

Starting from a Hermitian operator with two distinct eigenvalues, we construct a non-Hermitian bipartite system in Gaussian orthogonal ensemble according to random matrix theory, where we introduce the off-diagonal fluctuations through random eigenkets and realizing the bipartite configuration consisting of two $D\times D$ subsystems (with $D$ the Hilbert sp

Seonok Kim

The increasing popularity of exercises including yoga and Pilates has created a greater demand for professional exercise video datasets in the realm of artificial intelligence. In this study, we developed 3DYoga901, which is organized within a three-level label hierarchy. We have expanded the number of poses from an existing state-of-the-art dataset, increas

Ri He, Haowei Xu, Peijun Yang, Kai Chang

Ferroelastic twin walls in nonpolar materials can give rise to a spontaneous polarization due to symmetry breaking. Nevertheless, the bi-stable polarity of twin walls and its reversal have not yet been demonstrated. Here, we report that the polarity of SrTiO$_3$ twin walls can be switched by ultra-low strain gradient. Using first-principles-based machine-lea

Ognian Kassabov, Velichka Milousheva

We study minimal timelike surfaces in $\mathbb R^3_1$ using a special Weierstrass-type formula in terms of holomorphic functions defined in the algebra of the double (split-complex) numbers. We present a method of obtaining an equation of a minimal timelike surface in terms of canonical parameters, which play a role similar to the role of the natural paramet

Wei Lu, Matthieu Nicoul, Uladzimir Shymanovich, Alexander Tarasevitch

We present a table-top setup for femtosecond time-resolved X-ray diffraction based on a Cu K{\alpha} (8.05 keV) laser driven plasma X-ray source. Due to its modular design it provides high accessibility to its individual components (e.g. X-ray optics and sample environment). The K{\alpha}-yield of the source is optimized using a pre-pulse scheme. A magnifyin

Nodoka Yamanaka

The atomic, nuclear, and nucleon electric dipole moments (EDMs) have significant sensitivity to the CP violation of elementary particle physics, but its quantification has for long been obstructed by the nonperturbative physics of quantum chromodynamics. Quite recently, there were significant progresses in this field, notably the quantification of hadron lev

Koushik Biswas, Meghana Karri, Ulaş Bağcı

Activation functions are crucial in deep learning models since they introduce non-linearity into the networks, allowing them to learn from errors and make adjustments, which is essential for learning complex patterns. The essential purpose of activation functions is to transform unprocessed input signals into significant output activations, promoting informa

Xiang Wang, Shiwei Zhang, Hangjie Yuan, Yingya Zhang

Transferring vision-language knowledge from pretrained multimodal foundation models to various downstream tasks is a promising direction. However, most current few-shot action recognition methods are still limited to a single visual modality input due to the high cost of annotating additional textual descriptions. In this paper, we develop an effective plug-

Joann Qiongna Chen, Xinlei He, Zheng Li, Yang Zhang

Training a machine learning model with data following a meaningful order, i.e., from easy to hard, has been proven to be effective in accelerating the training process and achieving better model performance. The key enabling technique is curriculum learning (CL), which has seen great success and has been deployed in areas like image and text classification.

Yitong Jiang, Zhaoyang Zhang, Tianfan Xue, Jinwei Gu

We present AutoDIR, an innovative all-in-one image restoration system incorporating latent diffusion. AutoDIR excels in its ability to automatically identify and restore images suffering from a range of unknown degradations. AutoDIR offers intuitive open-vocabulary image editing, empowering users to customize and enhance images according to their preferences

Omar Mustafa

We use two correlated metric functions and transform/deform the pointlike global monopole (PGM) spacetime metric into a quasi-PGM (QPGM) spacetime one. We study Klein-Gordon (KG) particles (manifestly introduced by the non-minimal coupling form of the operator $% \tilde{D}_{\mu }=D_{\mu }+\mathcal{F}_{\mu }$, with $\mathcal{F}_{\mu }$ $% \in %TCIMACRO{\U{211

Andi Han, Dai Shi, Lequan Lin, Junbin Gao

Graph neural networks (GNNs) have demonstrated significant promise in modelling relational data and have been widely applied in various fields of interest. The key mechanism behind GNNs is the so-called message passing where information is being iteratively aggregated to central nodes from their neighbourhood. Such a scheme has been found to be intrinsically

Luca Brandolini, Leonardo Colzani, Giancarlo Travaglini

We prove several results concerning the discrepancy, tested on balls in the $d$-dimensional torus $\mathbb{T}^{d}$, between absolutely continuous measures and finite atomic measures.

Fatma Çiçek, Steven M. Gonek

On the assumption of the Riemann hypothesis and a spacing hypothesis for the nontrivial zeros $\frac12+i\gamma$ of the Riemann zeta function, we show that the sequence \[ \Gamma_{[a, b]} =\Bigg\{ \gamma : \gamma>0 \quad \mbox{and} \quad \frac{ \log\big(| \zeta^{(m_{\gamma})} (\frac12+ i\gamma) | / (\log{\gamma})^{m_{\gamma}}\big)}{\sqrt{\frac12\log\log{\gamm

Arthur Amalvy, Vincent Labatut, Richard Dufour

While recent pre-trained transformer-based models can perform named entity recognition (NER) with great accuracy, their limited range remains an issue when applied to long documents such as whole novels. To alleviate this issue, a solution is to retrieve relevant context at the document level. Unfortunately, the lack of supervision for such a task means one

Chuan He, Le Peng, Ju Sun

In practice, many machine learning (ML) problems come with constraints, and their applied domains involve distributed sensitive data that cannot be shared with others, e.g., in healthcare. Collaborative learning in such practical scenarios entails federated learning (FL) for ML problems with constraints, or FL with constraints for short. Despite the extensiv

Blandine Bolteau, Frédéric Gelebart, Jérémie Teisseire, Etienne Barthel

Interfaces between a water droplet and a network of pillars produce eventually superhydrophobic, self-cleaning properties. Considering the surface fraction of the surface in interaction with water, it is possible to tune precisely the contact angle hysteresis (CAH) to low values, which is at the origin of the poor adhesion of water droplets, inducing their h

Luca Castelli, Clément Marteau, Irène Gannaz

This paper investigates some theoretical properties of the Partial Least Square (PLS) method. We focus our attention on the single component case, that provides a useful framework to understand the underlying mechanism. We provide a non-asymptotic upper bound on the quadratic loss in prediction with high probability in a high dimensional regression context.

Eunho Koo, Tongseok Lim

In the node classification task, it is natural to presume that densely connected nodes tend to exhibit similar attributes. Given this, it is crucial to first define what constitutes a dense connection and to develop a reliable mathematical tool for assessing node cohesiveness. In this paper, we propose a probability-based objective function for semi-supervis

Melissa Lee, Tomasz Popiel

The prime graph of a finite group $G$ is the labelled graph $\Gamma(G)$ with vertices the prime divisors of $|G|$ and edges the pairs $\{p,q\}$ for which $G$ contains an element of order $pq$. A group $G$ is recognisable by its prime graph if every group $H$ with $\Gamma(H)=\Gamma(G)$ is isomorphic to $G$. Cameron and Maslova have shown that every group that

Georges Gras

We consider, for real abelian fields K, the Birch--Tate formula linking the tame kernel \#K\_2(Z\_K) to $\zeta$\_K(-1); we compare, for quadratic and cyclic cubic fields with p=2,3, \#K\_2(\BZ\_K)[p^$\infty$] to the order of the torsion group T\_{K, p} of abelian p-ramification theory given by the residue of $\zeta$\_{K, p}(s) at s=1. This is done via the ``

Parul Sharma, Brijesh Kumar, Nihar Ranjan Sahoo, Anshuman Kumar

The heightened sensitivity observed in non-Hermitian systems at exceptional points (EPs) has garnered significant attention. Typical EP sensor implementations rely on precise measurements of spectra and importantly, for real time sensing measurements, the EP condition ceases to hold as the perturbation increases over time, thereby preventing the use of high

Dominique Colnet, Benoît Sonntag

A widespread practice to implement a flexible array is to consider the storage area into two parts: the used area, which is already available for read/write operations, and the supply area, which is used in case of enlargement of the array. The main purpose of the supply area is to avoid as much as possible the reallocation of the whole storage area in case

Olivier Biquard, Tristan Ozuch

We prove the instability of conformally K\"ahler, compact or ALF Einstein 4-manifolds with nonnegative scalar curvature which are not half conformally flat. This applies to all the known examples of gravitational instantons which are not hyperK\"ahler and to the Chen-Lebrun-Weber metric in particular.

An Zhang, Yuxin Chen, Leheng Sheng, Xiang Wang

Recommender systems are the cornerstone of today's information dissemination, yet a disconnect between offline metrics and online performance greatly hinders their development. Addressing this challenge, we envision a recommendation simulator, capitalizing on recent breakthroughs in human-level intelligence exhibited by Large Language Models (LLMs). We propo

Dengwang Tang, Dongze Ye, Rahul Jain, Ashutosh Nayyar

Learning in POMDPs is known to be significantly harder than in MDPs. In this paper, we consider the online learning problem for episodic POMDPs with unknown transition and observation models. We propose a Posterior Sampling-based reinforcement learning algorithm for POMDPs (PS4POMDPs), which is much simpler and more implementable compared to state-of-the-art

Can Cui, Imran Ahamad Sheikh, Mostafa Sadeghi, Emmanuel Vincent

We present an end-to-end multichannel speaker-attributed automatic speech recognition (MC-SA-ASR) system that combines a Conformer-based encoder with multi-frame crosschannel attention and a speaker-attributed Transformer-based decoder. To the best of our knowledge, this is the first model that efficiently integrates ASR and speaker identification modules in

Sergei Kuksin

This work is a review with proofs of a group of results on the stochastic Burgers equation with small viscosity, obtained during the last two decades. These results jointly show that the equation makes a surprisingly good model of hydrodynamical turbulence. The model provides natural and rigorously justified analogies of a number of key predictions of the th

Raziyeh Zaregonbadi, Nasim Saba, Mehrdad Farhoudi

We investigate cosmological solutions of the chameleon model with a non-minimal coupling between the matter and the scalar field through a conformal factor with gravitational strength. By considering the spatially flat FLRW metric and the matter density as a non-relativistic perfect fluid, we focus on the matter-dominated phase and the late-time accelerated

Dhruv Shah, Michael Equi, Blazej Osinski, Fei Xia

Navigation in unfamiliar environments presents a major challenge for robots: while mapping and planning techniques can be used to build up a representation of the world, quickly discovering a path to a desired goal in unfamiliar settings with such methods often requires lengthy mapping and exploration. Humans can rapidly navigate new environments, particular

Truong Thao Nguyen, Balazs Gerofi, Edgar Josafat Martinez-Noriega, François Trahay

This paper proposes a method for hiding the least-important samples during the training of deep neural networks to increase efficiency, i.e., to reduce the cost of training. Using information about the loss and prediction confidence during training, we adaptively find samples to exclude in a given epoch based on their contribution to the overall learning pro

Yu Pan, Ye Yuan, Yichun Yin, Zenglin Xu

Training large models from scratch usually costs a substantial amount of resources. Towards this problem, recent studies such as bert2BERT and LiGO have reused small pretrained models to initialize a large model (termed the ``target model''), leading to a considerable acceleration in training. Despite the successes of these previous studies, they grew pretra

Calum MacRury, Will Ma

We introduce a new approach for designing Random-order Contention Resolution Schemes (RCRS) via exact solution in continuous time. Given a function $c(y):[0,1] \rightarrow [0,1]$, we show how to select each element which arrives at time $y \in [0,1]$ with probability exactly $c(y)$. We provide a rigorous algorithmic framework for achieving this, which discre

The LHAASO Collaboration, Z. Cao, F. Aharonian, Q. An

We report the detection of a $\gamma$-ray bubble spanning at least 100$\rm deg^2$ in ultra high energy (UHE) up to a few PeV in the direction of the star-forming region Cygnus X, implying the presence Super PeVatron(s) accelerating protons to at least 10 PeV. A log-parabola form with the photon index $\Gamma (E) = (2.71 \pm 0.02) + (0.11 \pm 0.02) \times \lo

Yakov I. Korepanov

The investigation of the processes of mineral deposit formation and their history is a fundamental task. Solving this task can increase mining efficiency and make a significant contribution to understanding the formation of the Earth's crust. The main approach to solving this task is determining the values of thermodynamic functions for phases present in dep

Anand Brahmbhatt, Rishi Saket, Aravindan Raghuveer

Learning from label proportions (LLP) is a generalization of supervised learning in which the training data is available as sets or bags of feature-vectors (instances) along with the average instance-label of each bag. The goal is to train a good instance classifier. While most previous works on LLP have focused on training models on such training data, comp

Alexander Iksanov, Vitali Wachtel

Let $\eta_1$, $\eta_2,\ldots$ be independent copies of a random variable $\eta$ with zero mean and finite variance which is bounded from the right, that is, $\eta\leq b$ almost surely for some $b>0$. Considering different types of the asymptotic behaviour of the probability $\mathbb{P}\{\eta\in[b-x,b]\}$ as $x\to 0+$ we derive precise tail asymptotics of the

Anand Brahmbhatt, Mohith Pokala, Rishi Saket, Aravindan Raghuveer

In the task of Learning from Label Proportions (LLP), a model is trained on groups (a.k.a bags) of instances and their corresponding label proportions to predict labels for individual instances. LLP has been applied pre-dominantly on two types of datasets - image and tabular. In image LLP, bags of fixed size are created by randomly sampling instances from an

Yangfan Ni, Duo Zhang, Gege Ma, Lijun Lu

Accurate reorientation and segmentation of the left ventricular (LV) is essential for the quantitative analysis of myocardial perfusion imaging (MPI), in which one critical step is to reorient the reconstructed transaxial nuclear cardiac images into standard short-axis slices for subsequent image processing. Small-scale LV myocardium (LV-MY) region detection

Yao Xiao, Lu Xu, Jiaxi Li, Wei Lu

While prompt tuning approaches have achieved competitive performance with high efficiency, we observe that they invariably employ the same initialization process, wherein the soft prompt is either randomly initialized or derived from an existing embedding vocabulary. In contrast to these conventional methods, this study aims to investigate an alternative way

Shuhei Yoshida

Computer-generated holography (CGH) can be used to display three-dimensional (3D) images and has a special feature that no other technology possesses: it can reconstruct arbitrary object wavefronts. In this study, we investigated a high-speed full-color reconstruction method for improving the realism of 3D images produced using CGH. The proposed method uses

Anand Brahmbhatt, Rishi Saket, Shreyas Havaldar, Anshul Nasery

In many real-world applications, due to recent developments in the privacy landscape, training data may be aggregated to preserve the privacy of sensitive training labels. In the learning from label proportions (LLP) framework, the dataset is partitioned into bags of feature-vectors which are available only with the sum of the labels per bag. A further restr

Tatsuo Kobayashi, Takaaki Nomura, Hiroshi Okada, Hajime Otsuka

We propose an interesting assignment of positive modular wights for fields in a modular non-Abelian discrete flavor symmetry. By this assignment, we can construct inverse seesaw and linear seesaw models without any additional symmetries which possess good testability in current experiments. At first, we discuss probabilities for positive modular wights from

Yanbiao Ma, Licheng Jiao, Fang Liu, Shuyuan Yang

In scenarios with long-tailed distributions, the model's ability to identify tail classes is limited due to the under-representation of tail samples. Class rebalancing, information augmentation, and other techniques have been proposed to facilitate models to learn the potential distribution of tail classes. The disadvantage is that these methods generally pu

Jingyang Zhu, Yuanming Shi, Yong Zhou, Chunxiao Jiang

Federated learning (FL), as an emerging distributed machine learning paradigm, allows a mass of edge devices to collaboratively train a global model while preserving privacy. In this tutorial, we focus on FL via over-the-air computation (AirComp), which is proposed to reduce the communication overhead for FL over wireless networks at the cost of compromising

Hyemi Jang, Junsung Park, Dahuin Jung, Jaihyun Lew

Although supervised image denoising networks have shown remarkable performance on synthesized noisy images, they often fail in practice due to the difference between real and synthesized noise. Since clean-noisy image pairs from the real world are extremely costly to gather, self-supervised learning, which utilizes noisy input itself as a target, has been st

Mario Eudave-Muñoz, Masakazu Teragaito

We show an infinite family of hyperbolic knots that have an exceptional surgery producing a graph manifold containing five disjoint, and non parallel incompressible tori.

Irfan Siddique, Uzma Tabassam

We study the space-time evolution of electromagnetic fields along with the azimuthal fluctuations of these fields and their correlation with the initial matter geometry specified by the participant plane in the presence of finite electric $\left(\sigma\right)$ and chiral magnetic $\left(\sigma_{\chi}\right)$ conductivities in Ru+Ru and Zr+Zr collisions at $\

Aniket Nargundkar, Madhav Rawal, Aryaman Patel, Anand J Kulkarni

Recently, various Artificial Intelligence (AI) based optimization metaheuristics are proposed and applied for a variety of problems. Cohort Intelligence (CI) algorithm is a socio inspired optimization technique which is successfully applied for solving several unconstrained & constrained real-world problems from the domains such as design, manufacturing, sup

Yucheng Fu, Jie Bao, Rajesh Kumar Singh, Chao Wang

Liquid holdup and mass transfer area are critical parameters for packed column design and CO2 capture efficiency prediction. In this paper, a framework was established for modeling the liquid-gas countercurrent flow hydrodynamics in a random packed column with pall rings. Besides the column-averaged information, the radial pall ring distribution, velocity, a

Hayato Morimura

For the stopped Weinstein sector associated with any fanifold recently introduced by Gammage--Shende, we construct a Weinstein sectorial cover which allows us to describe homological mirror symmetry over the fanifold as an isomorphism of cosheaves of categories. In a special case, our Weinstein sectorial cover gives a lift of the open cover for the global sk

Issey Sukeda, Masahiro Suzuki, Hiroki Sakaji, Satoshi Kodera

In the ongoing wave of impact driven by large language models (LLMs) like ChatGPT, the adaptation of LLMs to medical domain has emerged as a crucial research frontier. Since mainstream LLMs tend to be designed for general-purpose applications, constructing a medical LLM through domain adaptation is a huge challenge. While instruction-tuning is used to fine-t

Tianjiao Li, Guanghui Lan

Line search (or backtracking) procedures have been widely employed into first-order methods for solving convex optimization problems, especially those with unknown problem parameters (e.g., Lipschitz constant). In this paper, we show that line search is superfluous in attaining the optimal rate of convergence for solving a convex optimization problem whose p

Nilakantha Meher, Tomáš Opatrný, Gershon Kurizki

We put forth the concept of quantum noise sensing in nonlinear two-mode interferometers coupled to mechanical oscillators. These autonomous machines are capable of sensing quantum nonlinear correlations of two-mode noisy fields via their thermodynamic variable of extractable work, alias work capacity or ergotropy. The fields are formed by thermal noise input

Qianli Ma, Haotian Zhou, Tingkai Liu, Jianbo Yuan

Recent years have seen considerable advancements in multi-step reasoning with Large Language Models (LLMs). The previous studies have elucidated the merits of integrating feedback or search mechanisms during model inference to improve the reasoning accuracy. The Process-Supervised Reward Model (PRM), typically furnishes LLMs with step-by-step feedback during

Deok-Kyeong Jang, Yuting Ye, Jungdam Won, Sung-Hee Lee

Transforming neutral, characterless input motions to embody the distinct style of a notable character in real time is highly compelling for character animation. This paper introduces MOCHA, a novel online motion characterization framework that transfers both motion styles and body proportions from a target character to an input source motion. MOCHA begins by

Leilee Chojnacki, Rico Pohle, Han Yan, Yutaka Akagi

Many large-scale phenomena in our Universe, such as gravitational waves, are challenging to reproduce in laboratory settings. However, parallels with condensed matter systems can provide alternative routes for experimental accessibility. Here we show how spin nematic phases provide a low-energy avenue for accessing the physics of linearized gravity, and in p

Shuyu Jiang, Xingshu Chen, Rui Tang

Recently, Large language models (LLMs) with powerful general capabilities have been increasingly integrated into various Web applications, while undergoing alignment training to ensure that the generated content aligns with user intent and ethics. Unfortunately, they remain the risk of generating harmful content like hate speech and criminal activities in pr

Keita Saito, Akifumi Wachi, Koki Wataoka, Youhei Akimoto

In recent years, Large Language Models (LLMs) have witnessed a remarkable surge in prevalence, altering the landscape of natural language processing and machine learning. One key factor in improving the performance of LLMs is alignment with humans achieved with Reinforcement Learning from Human Feedback (RLHF), as for many LLMs such as GPT-4, Bard, etc. In a

Zhiwu Lin, Yucong Wang, Wenpei Wu

Magneto-rotational instability (MRI) is an important instability mechanism for rotating flows with magnetic fields. In particular, when the strength of the magnetic field tends to zero, the stability criterion for rotating flows is generally different from the classical Rayleigh criterion for rotating flows without a magnetic field. MRI has wide applications

Taesik Gong, Yewon Kim, Taeckyung Lee, Sorn Chottananurak

Test-time adaptation (TTA) aims to address distributional shifts between training and testing data using only unlabeled test data streams for continual model adaptation. However, most TTA methods assume benign test streams, while test samples could be unexpectedly diverse in the wild. For instance, an unseen object or noise could appear in autonomous driving

Taewoong Kang, Jeongsik Oh, Jaeseong Lee, Sunghyun Park

Despite the remarkable advancements in head reenactment, the existing methods face challenges in cross-domain head reenactment, which aims to transfer human motions to domains outside the human, including cartoon characters. It is still difficult to extract motion from out-of-domain images due to the distinct appearances, such as large eyes. Recently, previo

Yingwei Ma, Yue Yu, Shanshan Li, Yu Jiang

Large language models (LLMs) have showcased remarkable prowess in code generation. However, automated code generation is still challenging since it requires a high-level semantic mapping between natural language requirements and codes. Most existing LLMs-based approaches for code generation rely on decoder-only causal language models often treate codes merel

Ehsan Latif, Xiaoming Zhai

This study highlights the potential of fine-tuned ChatGPT (GPT-3.5) for automatically scoring student written constructed responses using example assessment tasks in science education. Recent studies on OpenAI's generative model GPT-3.5 proved its superiority in predicting the natural language with high accuracy and human-like responses. GPT-3.5 has been tra

Yutong Kou, Jin Gao, Bing Li, Gang Wang

Recently, the transformer has enabled the speed-oriented trackers to approach state-of-the-art (SOTA) performance with high-speed thanks to the smaller input size or the lighter feature extraction backbone, though they still substantially lag behind their corresponding performance-oriented versions. In this paper, we demonstrate that it is possible to narrow

Jiaming Liang, Yuwan Xue, Haowei Liu, Zhenqi Dai

In existing splicing forgery datasets, the insufficient semantic variety of spliced regions causes trained detection models to overfit semantic features rather than learn genuine splicing traces. Meanwhile, the lack of a reasonable benchmark dataset has led to inconsistent experimental settings across existing detection methods. To address these issues, we p

S. Suetsugu, T. Asaba, Y. Kasahara, Y. Kohsaka

The two-dimensional (2D) spin-1/2 kagome Heisenberg antiferromagnet is believed to host quantum spin liquid (QSL) states with no magnetic order, but its ground state remains largely elusive. An important outstanding question concerns the presence or absence of the 1/9 magnetization plateau, where exotic quantum states, including topological ones, are expecte

Junjie Li, Guanshuo Wang, Yichao Yan, Fufu Yu

Person search is a challenging task that involves detecting and retrieving individuals from a large set of un-cropped scene images. Existing person search applications are mostly trained and deployed in the same-origin scenarios. However, collecting and annotating training samples for each scene is often difficult due to the limitation of resources and the l

Athul Pradeepkumar Girija

Aerocapture uses atmospheric drag to decelerate spacecraft and achieve orbit insertion. One of the significant risks associated with aerocapture is the uncertainty in the atmospheric density, particularly for outer planets. The paper performs a comparative study of the atmospheric uncertainties and provides design rules for aerocapture missions. The atmosphe

Kakeru Oshikiri, Masaomi Tanaka, Nozomu Tominaga, Tomoki Morokuma

The population of optical transients evolving within a time-scale of a few hours or a day (so-called fast optical transients, FOTs) has recently been debated extensively. In particular, our understanding of extragalactic FOTs and their rates is limited. We present a search for extragalactic FOTs with the Tomo-e Gozen high-cadence survey. Using the data taken

Qin Wang, Guangsheng Yu, Shiping Chen

In this paper, we design, implement, and (partially-) evaluate a lightweight bridge (as a type of middleware) to connect the Bitcoin and Ethereum networks that were heterogeneously uncontactable before. Inspired by the recently introduced Bitcoin Request Comment (BRC-20) standard, we leverage the flexibility of Bitcoin inscriptions by embedding editable oper

Junjie Xu, Enyan Dai, Dongsheng Luo, Xiang Zhang

Spectral Graph Neural Networks (GNNs) are gaining attention for their ability to surpass the limitations of message-passing GNNs. They rely on supervision from downstream tasks to learn spectral filters that capture the graph signal's useful frequency information. However, some works empirically show that the preferred graph frequency is related to the graph

Shahida Perveen, Qurat-ul-Ain, Sarosh Iqbal, Sheeba Wajid

To investigate1,1-Diphenyl-2-picrylhydrazyl (DPPH) and superoxide radical (SOR) 17 scavenging activities of 2-oxo-1,2,3,4-tetrahydropyrimidines derivatives. Free radicals are 18 highly unstable and reactive molecules/atoms. In the body, free radicals form during 19 normal and abnormal metabolism in the body and cause serious damage to other 20 biomolecules t

Alon Jacovi, Avi Caciularu, Jonathan Herzig, Roee Aharoni

A growing area of research investigates augmenting language models with tools (e.g., search engines, calculators) to overcome their shortcomings (e.g., missing or incorrect knowledge, incorrect logical inferences). Various few-shot tool-usage strategies have been proposed. However, there is no systematic and fair comparison across different strategies, or be

Rachel F. Heaton

The following is a dissertation aimed at understanding what the various phenomena in visual search teach us about the nature of human visual representations and processes. I first review some of the major empirical findings in the study of visual search. I next present a theory of visual search in terms of what I believe these findings suggest about the repr

Zijun Gao, Haibao Liu, Lingbo Li

Data Augmentation (DA) has become a critical approach in Time Series Classification (TSC), primarily for its capacity to expand training datasets, enhance model robustness, introduce diversity, and reduce overfitting. However, the current landscape of DA in TSC is plagued with fragmented literature reviews, nebulous methodological taxonomies, inadequate eval

Lei Wang, Piotr Koniusz

Various research studies indicate that action recognition performance highly depends on the types of motions being extracted and how accurate the human actions are represented. In this paper, we investigate different optical flow, and features extracted from these optical flow that capturing both short-term and long-term motion dynamics. We perform power nor

Athena Wang

In this paper, we explore two different methods of finding the degrees of degeneracy for lattice model systems, specifically constructing one with a Fibonacci degree of degeneracy. We also calculate the number of ground states per site as the golden ratio $(\phi)$ for the system that we constructed and extend our results to systems with $k-$Step Fibonacci de

Toshihiro Nakanishi

In this study, we develop a time-varying metasurface based on the bound states in the continuum (BIC) with variable conductors, to store electromagnetic waves. The storage and retrieval of electromagnetic waves are demonstrated numerically through dynamic switching between quasi-BIC and BIC states by modulating the variable conductors. The storage efficiency

Han Qi, Xinyang Geng, Stefano Rando, Iku Ohama

In computational chemistry, crystal structure prediction (CSP) is an optimization problem that involves discovering the lowest energy stable crystal structure for a given chemical formula. This problem is challenging as it requires discovering globally optimal designs with the lowest energies on complex manifolds. One approach to tackle this problem involves

Motoki Asano, Hiroshi Yamaguchi, Hajime Okamoto

We demonstrate a fiber-type optomechanical array consisting of elastically interconnected silica microbottle resonators with high-Q optical and mechanical modes. In total, fifty optomechanical resonators fabricated by fine glass processing are uniformly arrayed on a silica fiber. Evanescent coupling of a tapered optical fiber to an arbitrary resonator allows

Jongwoo Ko, Seungjoon Park, Yujin Kim, Sumyeong Ahn

Structured pruning methods have proven effective in reducing the model size and accelerating inference speed in various network architectures such as Transformers. Despite the versatility of encoder-decoder models in numerous NLP tasks, the structured pruning methods on such models are relatively less explored compared to encoder-only models. In this study,

Jie Tang, Bin He, Junkai Xu, Tian Tan

Falls represent a significant cause of injury among the elderly population. Extensive research has been devoted to the utilization of wearable IMU sensors in conjunction with machine learning techniques for fall detection. To address the challenge of acquiring costly training data, this paper presents a novel method that generates a substantial volume of syn

Shervin Vakili, Mobin Vaziri, Amirhossein Zarei, J. M. Pierre Langlois

Multipliers are widely-used arithmetic operators in digital signal processing and machine learning circuits. Due to their relatively high complexity, they can have high latency and be a significant source of power consumption. One strategy to alleviate these limitations is to use approximate computing. This paper thus introduces an original FPGA-based approx

Jiaqing Zhu, Lin Wang, Fasheng Sun

Hierarchical data analysis is crucial in various fields for making discoveries. The linear mixed model is often used for training hierarchical data, but its parameter estimation is computationally expensive, especially with big data. Subsampling techniques have been developed to address this challenge. However, most existing subsampling methods assume homoge

An Zhang, Wenchang Ma, Jingnan Zheng, Xiang Wang

In leading collaborative filtering (CF) models, representations of users and items are prone to learn popularity bias in the training data as shortcuts. The popularity shortcut tricks are good for in-distribution (ID) performance but poorly generalized to out-of-distribution (OOD) data, i.e., when popularity distribution of test data shifts w.r.t. the traini

Tom Bryan, Jacob Carlson, Abhishek Arora, Melissa Dell

Billions of public domain documents remain trapped in hard copy or lack an accurate digitization. Modern natural language processing methods cannot be used to index, retrieve, and summarize their texts; conduct computational textual analyses; or extract information for statistical analyses, and these texts cannot be incorporated into language model training.

Tao Fan, Yan Kang, Guoqiang Ma, Weijing Chen

Large Language Models (LLMs), such as ChatGPT, LLaMA, GLM, and PaLM, have exhibited remarkable performances across various tasks in recent years. However, LLMs face two main challenges in real-world applications. One challenge is that training LLMs consumes vast computing resources, preventing LLMs from being adopted by small and medium-sized enterprises wit

Ge Zhao, Yanyuan Ma, Huazhen Lin, Yi Li

Kidney transplantation is the most effective renal replacement therapy for end stage renal disease patients. With the severe shortage of kidney supplies and for the clinical effectiveness of transplantation, patient's life expectancy post transplantation is used to prioritize patients for transplantation; however, severe comorbidity conditions and old age ar

Yixin Liu, Avi Singh, C. Daniel Freeman, John D. Co-Reyes

Despite their success in many natural language tasks, solving math problems remains a significant challenge for large language models (LLMs). A large gap exists between LLMs' pass-at-one and pass-at-N performance in solving math problems, suggesting LLMs might be close to finding correct solutions, motivating our exploration of fine-tuning methods to unlock

Baodong Wu, Lei Xia, Qingping Li, Kangyu Li

Large language models (LLMs) with hundreds of billions or trillions of parameters, represented by chatGPT, have achieved profound impact on various fields. However, training LLMs with super-large-scale parameters requires large high-performance GPU clusters and long training periods lasting for months. Due to the inevitable hardware and software failures in

Heng Zhang, Danilo Vasconcellos Vargas

Recently, SyncMap pioneered an approach to learn complex structures from sequences as well as adapt to any changes in underlying structures. This is achieved by using only nonlinear dynamical equations inspired by neuron group behaviors, i.e., without loss functions. Here we propose Symmetrical SyncMap that goes beyond the original work to show how to create