arXiv subjects

Rong Wang

Publications and source records attributed to Rong Wang.

At least 91 records · Page 5Linked to original sources

arXiv2024

Automatic Essay Multi-dimensional Scoring with Fine-tuning and Multiple Regression

Automated essay scoring (AES) involves predicting a score that reflects the writing quality of an essay. Most existing AES systems produce only a single overall score. However, users and L2 learners expect scores across different dimensions (e.g., vocabulary, grammar, coherence) for English essays in real-world applications. To address this need, we have developed two models that automatically score English essays across multiple dimensions by employing fine-tuning and other strategies on two large datasets. The results demonstrate that our systems achieve impressive performance in evaluation using three criteria: precision, F1 score, and Quadratic Weighted Kappa. Furthermore, our system outperforms existing methods in overall scoring.

cs.CL↗

arXiv2024

The European Muon Collaboration effect from short-range correlated nucleons in a nucleon swelling model

The relation between the nuclear EMC effect and the nucleon-nucleon short-range correlation is a hot topic in high-energy nuclear physics, ever since a peculiar linear correlation between these two phenomena discovered. In this paper, the contribution to the nuclear EMC effect arising from the short-range correlated nucleons is examined in a nucleon-swelling model. We find that the structure modifications of the N-N SRC nucleons reproduce more or less the measured EMC ratios of light nuclei, while they are not enough to explain the measured EMC ratios of heavy nuclei. We speculate that the hypothesis of causal connection between SRC and the EMC effect is not exact, or the universality of the inner structure of SRC nucleon is violated noticeably from light to heavy nuclei, or there are other origins for the EMC effect.

nucl-th↗

arXiv2024

Classical information entropy of parton distribution functions and an application in searching gluon saturation

Entropy or information is a fundamental quantity contained in a system in statistical mechanics and information theory. In this paper, a definition of classical information entropy of parton distribution functions is suggested. The extensive and supper-additive properties of the defined entropy are discussed. The concavity is also deduced for the defined entropy. As an example, the classical information entropy of the gluon distribution of the proton is presented. There are some particular features of the evolution of the information entropy in the saturating domain, which could be used in finding the signals of gluon saturation.

hep-ph↗

arXiv2024

Gluon distributions and mass decompositions of the pion and kaon

We present the gluon distribution functions of pion and kaon in small-$x$ and large-$x$ regions, and compare them with the results obtained from lattice QCD and continuum Schwinger function methods. Whether in the small-$x$ region or the large-$x$ region, our gluon distribution of pion is consistent with the results of lattice QCD and continuum Schwinger function method. In addition, the first four moments of gluon distributions of pion and kaon at different $Q^{2}$ scales are calculated. Furthermore, we present the mass decompositions of pion and kaon with the dynamical parton distribution functions calculated by the DGLAP equation with parton-parton recombination corrections. The mass structures of pion and kaon are totally different from that of the proton.

hep-ph↗

arXiv2024

Robust Capped lp-Norm Support Vector Ordinal Regression

Ordinal regression is a specialized supervised problem where the labels show an inherent order. The order distinguishes it from normal multi-class problem. Support Vector Ordinal Regression, as an outstanding ordinal regression model, is widely used in many ordinal regression tasks. However, like most supervised learning algorithms, the design of SVOR is based on the assumption that the training data are real and reliable, which is difficult to satisfy in real-world data. In many practical applications, outliers are frequently present in the training set, potentially leading to misguide the learning process, such that the performance is non-optimal. In this paper, we propose a novel capped $\ell_{p}$-norm loss function that is theoretically robust to both light and heavy outliers. The capped $\ell_{p}$-norm loss can help the model detect and eliminate outliers during training process. Adhering to this concept, we introduce a new model, Capped $\ell_{p}$-Norm Support Vector Ordinal Regression(CSVOR), that is robust to outliers. CSVOR uses a weight matrix to detect and eliminate outliers during the training process to improve the robustness to outliers. Moreover, a Re-Weighted algorithm algorithm which is illustrated convergence by our theoretical results is proposed to effectively minimize the corresponding problem. Extensive experimental results demonstrate that our model outperforms state-of-the-art(SOTA) methods, particularly in the presence of outliers.

cs.LG↗

arXiv2024

Dual Model Replacement:invisible Multi-target Backdoor Attack based on Federal Learning

In recent years, the neural network backdoor hidden in the parameters of the federated learning model has been proved to have great security risks. Considering the characteristics of trigger generation, data poisoning and model training in backdoor attack, this paper designs a backdoor attack method based on federated learning. Firstly, aiming at the concealment of the backdoor trigger, a TrojanGan steganography model with encoder-decoder structure is designed. The model can encode specific attack information as invisible noise and attach it to the image as a backdoor trigger, which improves the concealment and data transformations of the backdoor trigger.Secondly, aiming at the problem of single backdoor trigger mode, an image poisoning attack method called combination trigger attack is proposed. This method realizes multi-backdoor triggering by multiplexing combined triggers and improves the robustness of backdoor attacks. Finally, aiming at the problem that the local training mechanism leads to the decrease of the success rate of backdoor attack, a dual model replacement backdoor attack algorithm based on federated learning is designed. This method can improve the success rate of backdoor attack while maintaining the performance of the federated learning aggregation model. Experiments show that the attack strategy in this paper can not only achieve high backdoor concealment and diversification of trigger forms under federated learning, but also achieve good attack success rate in multi-target attacks.door concealment and diversification of trigger forms but also achieve good results in multi-target attacks.

cs.LG↗

arXiv2024

TIMIT Speaker Profiling: A Comparison of Multi-task learning and Single-task learning Approaches

This study employs deep learning techniques to explore four speaker profiling tasks on the TIMIT dataset, namely gender classification, accent classification, age estimation, and speaker identification, highlighting the potential and challenges of multi-task learning versus single-task models. The motivation for this research is twofold: firstly, to empirically assess the advantages and drawbacks of multi-task learning over single-task models in the context of speaker profiling; secondly, to emphasize the undiminished significance of skillful feature engineering for speaker recognition tasks. The findings reveal challenges in accent classification, and multi-task learning is found advantageous for tasks of similar complexity. Non-sequential features are favored for speaker recognition, but sequential ones can serve as starting points for complex models. The study underscores the necessity of meticulous experimentation and parameter tuning for deep learning models.

cs.SD↗

arXiv2024

Computational Sentence-level Metrics Predicting Human Sentence Comprehension

The majority of research in computational psycholinguistics has concentrated on the processing of words. This study introduces innovative methods for computing sentence-level metrics using multilingual large language models. The metrics developed sentence surprisal and sentence relevance and then are tested and compared to validate whether they can predict how humans comprehend sentences as a whole across languages. These metrics offer significant interpretability and achieve high accuracy in predicting human sentence reading speeds. Our results indicate that these computational sentence-level metrics are exceptionally effective at predicting and elucidating the processing difficulties encountered by readers in comprehending sentences as a whole across a variety of languages. Their impressive performance and generalization capabilities provide a promising avenue for future research in integrating LLMs and cognitive science.

cs.CL↗

arXiv2024

Experimental demonstration of improved reference-frame-independent quantum key distribution over 175km

Reference-frame-independent (RFI) quantum key distribution (QKD) presents promising advantages, especially for mobile-platform-based implementations, as it eliminates the need for active reference frame calibration. While RFI-QKD has been explored in various studies, limitations in key rate and distance persist due to finite data collection. In this study, we experimentally demonstrate an improved RFI-QKD protocol proposed by Zhu \textit{et al.} [Opt. Lett. 47, 4219 (2022)], featuring a statistical quantity for bounding information leaked to Eve that exhibits more insensitivity to statistical fluctuations and more robustness to variations in the reference frame. Taking into account finite-size considerations and potential general attacks, RFI-QKD is implemented over a distance of 175 \si{\kilo\meter} in this work. We believe that our study extends the communication distance achievable by RFI-QKD, thereby constituting a notable advancement for its practical application.

quant-ph↗

arXiv2024

Exploring the cloud of feature interaction scores in a Rashomon set

Interactions among features are central to understanding the behavior of machine learning models. Recent research has made significant strides in detecting and quantifying feature interactions in single predictive models. However, we argue that the feature interactions extracted from a single pre-specified model may not be trustworthy since: a well-trained predictive model may not preserve the true feature interactions and there exist multiple well-performing predictive models that differ in feature interaction strengths. Thus, we recommend exploring feature interaction strengths in a model class of approximately equally accurate predictive models. In this work, we introduce the feature interaction score (FIS) in the context of a Rashomon set, representing a collection of models that achieve similar accuracy on a given task. We propose a general and practical algorithm to calculate the FIS in the model class. We demonstrate the properties of the FIS via synthetic data and draw connections to other areas of statistics. Additionally, we introduce a Halo plot for visualizing the feature interaction variance in high-dimensional space and a swarm plot for analyzing FIS in a Rashomon set. Experiments with recidivism prediction and image classification illustrate how feature interactions can vary dramatically in importance for similarly accurate predictive models. Our results suggest that the proposed FIS can provide valuable insights into the nature of feature interactions in machine learning models.

cs.LG↗

arXiv2024

On the local streamline pattern of planar polynomial velocity field with nonzero linear part

The streamline pattern of planar polynomial velocity field is far from fully understood. In the community of fluid mechanics, most studies simply focus on the velocity gradient, or the linear part of the velocity field, but few studies on high-order terms. This paper is concerned with the local streamline pattern (LSP) of velocity field around an isotropic point. In virtue of the concept and method of dynamical systems, where the streamline pattern is equivalent to the phase portrait, we make clear the classification of LSPs of planar velocity fields with nonzero linear part, especially the cases where the determinant of velocity gradient vanishes at the isotropic point.

physics.flu-dyn↗

arXiv2023

DeepSimHO: Stable Pose Estimation for Hand-Object Interaction via Physics Simulation

This paper addresses the task of 3D pose estimation for a hand interacting with an object from a single image observation. When modeling hand-object interaction, previous works mainly exploit proximity cues, while overlooking the dynamical nature that the hand must stably grasp the object to counteract gravity and thus preventing the object from slipping or falling. These works fail to leverage dynamical constraints in the estimation and consequently often produce unstable results. Meanwhile, refining unstable configurations with physics-based reasoning remains challenging, both by the complexity of contact dynamics and by the lack of effective and efficient physics inference in the data-driven learning framework. To address both issues, we present DeepSimHO: a novel deep-learning pipeline that combines forward physics simulation and backward gradient approximation with a neural network. Specifically, for an initial hand-object pose estimated by a base network, we forward it to a physics simulator to evaluate its stability. However, due to non-smooth contact geometry and penetration, existing differentiable simulators can not provide reliable state gradient. To remedy this, we further introduce a deep network to learn the stability evaluation process from the simulator, while smoothly approximating its gradient and thus enabling effective back-propagation. Extensive experiments show that our method noticeably improves the stability of the estimation and achieves superior efficiency over test-time optimization. The code is available at https://github.com/rongakowang/DeepSimHO.

cs.CV↗

arXiv2023

Quantum Random Number Generation with Partial Source Assumptions

Quantum random number generator harnesses the power of quantum mechanics to generate true random numbers, making it valuable for various scientific applications. However, real-world devices often suffer from imperfections that can undermine the integrity and privacy of generated randomness. To combat this issue, we present a novel quantum random number generator and experimentally demonstrate it. Our approach circumvents the need for exhaustive characterization of measurement devices, even in the presence of a quantum side channel. Additionally, we also do not require detailed characterization of the source, relying instead on reasonable assumptions about encoding dimension and noise constraints. Leveraging commercially available all-fiber devices, we achieve a randomness generation rate of 40 kbps.

quant-ph↗

arXiv2023

A Novel Normalized-Cut Solver with Nearest Neighbor Hierarchical Initialization

Normalized-Cut (N-Cut) is a famous model of spectral clustering. The traditional N-Cut solvers are two-stage: 1) calculating the continuous spectral embedding of normalized Laplacian matrix; 2) discretization via $K$-means or spectral rotation. However, this paradigm brings two vital problems: 1) two-stage methods solve a relaxed version of the original problem, so they cannot obtain good solutions for the original N-Cut problem; 2) solving the relaxed problem requires eigenvalue decomposition, which has $\mathcal{O}(n^3)$ time complexity ($n$ is the number of nodes). To address the problems, we propose a novel N-Cut solver designed based on the famous coordinate descent method. Since the vanilla coordinate descent method also has $\mathcal{O}(n^3)$ time complexity, we design various accelerating strategies to reduce the time complexity to $\mathcal{O}(|E|)$ ($|E|$ is the number of edges). To avoid reliance on random initialization which brings uncertainties to clustering, we propose an efficient initialization method that gives deterministic outputs. Extensive experiments on several benchmark datasets demonstrate that the proposed solver can obtain larger objective values of N-Cut, meanwhile achieving better clustering performance compared to traditional solvers.

cs.LG↗

arXiv2023

Joint Feature and Differentiable $ k $-NN Graph Learning using Dirichlet Energy

Feature selection (FS) plays an important role in machine learning, which extracts important features and accelerates the learning process. In this paper, we propose a deep FS method that simultaneously conducts feature selection and differentiable $ k $-NN graph learning based on the Dirichlet Energy. The Dirichlet Energy identifies important features by measuring their smoothness on the graph structure, and facilitates the learning of a new graph that reflects the inherent structure in new feature subspace. We employ Optimal Transport theory to address the non-differentiability issue of learning $ k $-NN graphs in neural networks, which theoretically makes our method applicable to other graph neural networks for dynamic graph learning. Furthermore, the proposed framework is interpretable, since all modules are designed algorithmically. We validate the effectiveness of our model with extensive experiments on both synthetic and real-world datasets.

cs.LG↗

arXiv2023

Exploring the Mass Radius of $^4$He and Implications for Nuclear Structure

In this study, we determine the mass radius of $^4$He, a very light nucleus, by examining the near-threshold $ϕ$-meson photoproduction data of the LEPS Collaboration. To assess the gravitational form factor of $^4$He, we employ multiple models for the mass distribution, including Yukawa-type, exponential, Gaussian, and uniform functions. The mass radius of $^4$He is measured to be $1.70\pm0.14$ fm, which is approximately equal to the charge radius of $^4$He. Surprisingly, in contrast to the findings of the proton, no noticeable discrepancy between the charge radius and the mass radius is noted for the $^4$He nucleus. The proton and neutron distributions within $^4$He are likely to be identical, confirming its regular tetrahedral structure in a new way. We propose exploring the difference between charge and mass radii as a new approach to examine the nuclear structure.

hep-ph↗

arXiv2023

OCU-Net: A Novel U-Net Architecture for Enhanced Oral Cancer Segmentation

Accurate detection of oral cancer is crucial for improving patient outcomes. However, the field faces two key challenges: the scarcity of deep learning-based image segmentation research specifically targeting oral cancer and the lack of annotated data. Our study proposes OCU-Net, a pioneering U-Net image segmentation architecture exclusively designed to detect oral cancer in hematoxylin and eosin (H&E) stained image datasets. OCU-Net incorporates advanced deep learning modules, such as the Channel and Spatial Attention Fusion (CSAF) module, a novel and innovative feature that emphasizes important channel and spatial areas in H&E images while exploring contextual information. In addition, OCU-Net integrates other innovative components such as Squeeze-and-Excite (SE) attention module, Atrous Spatial Pyramid Pooling (ASPP) module, residual blocks, and multi-scale fusion. The incorporation of these modules showed superior performance for oral cancer segmentation for two datasets used in this research. Furthermore, we effectively utilized the efficient ImageNet pre-trained MobileNet-V2 model as a backbone of our OCU-Net to create OCU-Netm, an enhanced version achieving state-of-the-art results. Comprehensive evaluation demonstrates that OCU-Net and OCU-Netm outperformed existing segmentation methods, highlighting their precision in identifying cancer cells in H&E images from OCDC and ORCA datasets.

eess.IV↗

arXiv2023

Production of the newly observed $\bar{T}_{c\bar{s}0}$ by kaon-induced reactions on a proton/neutron target

Recently, a new doubly charged tetraquark $T^{a}_{c\bar{s}0}(2900)^{++}$ and its neutral partner $T^{a}_{c\bar{s}0}(2900)^0$ at the invariant mass spectrum of $πD_s$ were observed by the LHCb Collaboration. According to its properties, such as the mass and decay width, the $T^{a}_{c\bar{s}0}(2900)^{++/0}$ have been suggested to be a compact multi-quark state or a hadron molecule. In order to distinguish the various interpretations of the $T^{a}_{c\bar{s}0}(2900)^{++/0}$, we investigate the possibility to study the $\bar{T}^{a}_{c\bar{s}0}(2900)$ [the antiparticle of $T^{a}_{c\bar{s}0}(2900)$] by kaon-induced reactions on a proton target in an effective Lagrangian approach. The production mechanism is characterized by the $t$-channel $D$ meson exchange. Our theoretical approach is based on the assumption that $\bar{T}^{a}_{c\bar{s}0}(2900)$ can be either a $K^{*}D^{*}-D_s^{*}ρ$ molecule or a compact tetraquark state. Using the coupling constants of the $\bar{T}^{a}_{c\bar{s}0}(2900)$ to $KD$ channel obtained from molecule or compact tetraquark picture of the $\bar{T}^{a}_{c\bar{s}0}(2900)$, we compute the cross-sections for the process $K^{-}n\to{}\bar{T}^{a}_{c\bar{s}0}(2900)^{--}Λ^{+}_c$. The $\bar{K}N$ initial state interaction mediated by Pomeron and Reggeon exchanges is also included, which reduces the production of the $\bar{T}^{a}_{c\bar{s}0}(2900)$. Our calculations show that whether $\bar{T}^{a}_{c\bar{s}0}(2900)$ is a molecule or a compact tetraquark state, the cross-sections for the $K^{-}n\to{}\bar{T}^{a}_{c\bar{s}0}(2900)^{--}Λ^{+}_c$ reaction are of similar magnitude, ranging from approximately 0.075 nb to 0.270 nb.

hep-ph↗