Jinshan Zeng
YOU?
Author Swipe
View article: Drug-Coated Balloons Versus Drug-Eluting Stents for Large Vessel Coronary Artery Disease: A Meta-Analysis
Drug-Coated Balloons Versus Drug-Eluting Stents for Large Vessel Coronary Artery Disease: A Meta-Analysis Open
Objective: We aimed to conduct a meta-analysis of treatments for large vessel coronary artery disease between drug-coated balloons and drug-eluting stents. Method: We searched databases including PubMed, Web of Science, Cochrane, CNKI, and…
View article: FedAPM: Federated Learning via ADMM with Partial Model Personalization
FedAPM: Federated Learning via ADMM with Partial Model Personalization Open
View article: FedAPM: Federated Learning via ADMM with Partial Model Personalization
FedAPM: Federated Learning via ADMM with Partial Model Personalization Open
In federated learning (FL), the assumption that datasets from different devices are independent and identically distributed (i.i.d.) often does not hold due to user differences, and the presence of various data modalities across clients ma…
View article: Self-Supervised Collaborative Information Bottleneck for Text Readability Assessment
Self-Supervised Collaborative Information Bottleneck for Text Readability Assessment Open
Text readability assessment involves categorizing texts based on readers' comprehension levels. Hybrid automatic readability assessment (ARA) models, combining deep and linguistic features, have recently attracted rising attention due to t…
View article: A robust economic dispatch method for coordinating inter-provincial and intra-provincial markets based on an inexact column-and-constraint generation method
A robust economic dispatch method for coordinating inter-provincial and intra-provincial markets based on an inexact column-and-constraint generation method Open
This paper addresses challenges in economic dispatch for multi-regional electricity markets, including low transmission efficiency, high rescheduling costs, and computational complexity. To mitigate these issues, a two-stage robust optimiz…
View article: Serum Lipid Biomarkers and the Risk of Gastrointestinal Cancers in a Chinese Population: The Kailuan Prospective Study
Serum Lipid Biomarkers and the Risk of Gastrointestinal Cancers in a Chinese Population: The Kailuan Prospective Study Open
Background Current evidence on relationships between serum lipid biomarkers and the risk of gastrointestinal cancers remains controversial, with no consensus reached. Methods We conducted a prospective cohort study within the Kailuan Cohor…
View article: Application of GBDT Integrated Algorithm in Financial Data Security Risk Prediction
Application of GBDT Integrated Algorithm in Financial Data Security Risk Prediction Open
View article: EDG‐CDM: A New Encoder‐Guided Conditional Diffusion Model‐Based Image Synthesis Method for Limited Data
EDG‐CDM: A New Encoder‐Guided Conditional Diffusion Model‐Based Image Synthesis Method for Limited Data Open
The Diffusion Probabilistic Model (DM) has emerged as a powerful generative model in the field of image synthesis, capable of producing high‐quality and realistic images. However, training DM requires a large and diverse dataset, which can…
View article: Skeleton-Style Transfer and Structural Contrastive Learning for Few-Shot Font Generation
Skeleton-Style Transfer and Structural Contrastive Learning for Few-Shot Font Generation Open
View article: Few-Shot Font Generation Via Stroke Prompt and Hierarchical Representation Learning
Few-Shot Font Generation Via Stroke Prompt and Hierarchical Representation Learning Open
View article: A Study on the Impact of Obstacle Size on Training Models Based on DQN and DDQN
A Study on the Impact of Obstacle Size on Training Models Based on DQN and DDQN Open
This paper presents a comparative analysis of Deep Q-Network (DQN) and Double Deep Q-Network (DDQN) algorithms in a simulated car racing environment, focusing on how variations in obstacle density affect each algorithm’s learning and traje…
View article: A Simple and Fast Way to Handle Semantic Errors in Transactions
A Simple and Fast Way to Handle Semantic Errors in Transactions Open
Many computer systems are now being redesigned to incorporate LLM-powered agents, enabling natural language input and more flexible operations. This paper focuses on handling database transactions created by large language models (LLMs). T…
View article: Efficient k-means with Individual Fairness via Exponential Tilting
Efficient k-means with Individual Fairness via Exponential Tilting Open
In location-based resource allocation scenarios, the distances between each individual and the facility are desired to be approximately equal, thereby ensuring fairness. Individually fair clustering is often employed to achieve the princip…
View article: Sequential Manipulation Against Rank Aggregation: Theory and Algorithm
Sequential Manipulation Against Rank Aggregation: Theory and Algorithm Open
Rank aggregation with pairwise comparisons is widely encountered in sociology, politics, economics, psychology, sports, etc. Given the enormous social impact and the consequent incentives, the potential adversary has a strong motivation to…
View article: Revealing the Unseen: AI Chain on LLMs for Predicting Implicit Dataflows to Generate Dataflow Graphs in Dynamically Typed Code
Revealing the Unseen: AI Chain on LLMs for Predicting Implicit Dataflows to Generate Dataflow Graphs in Dynamically Typed Code Open
Dataflow graphs (DFGs) capture definitions (defs) and uses across program blocks, which is a fundamental program representation for program analysis, testing and maintenance. However, dynamically typed programming languages like Python pre…
View article: GuessGas:Tell me Fine-Grained Gas Consumption of My Smart Contract and Why
GuessGas:Tell me Fine-Grained Gas Consumption of My Smart Contract and Why Open
Smart contracts with excessive gas consumption can cause economic losses, such as black hole contracts. Acutal gas consumption depends on runtime information and has a probability distribution under different runtime situations. However, e…
View article: InterpretARA: Enhancing Hybrid Automatic Readability Assessment with Linguistic Feature Interpreter and Contrastive Learning
InterpretARA: Enhancing Hybrid Automatic Readability Assessment with Linguistic Feature Interpreter and Contrastive Learning Open
The hybrid automatic readability assessment (ARA) models that combine deep and linguistic features have recently received rising attention due to their impressive performance. However, the utilization of linguistic features is not fully re…
View article: Personalized Federated Learning via ADMM with Moreau Envelope
Personalized Federated Learning via ADMM with Moreau Envelope Open
Personalized federated learning (PFL) is an approach proposed to address the issue of poor convergence on heterogeneous data. However, most existing PFL frameworks require strong assumptions for convergence. In this paper, we propose an al…
View article: GuessGas:Tell me Fine-Grained Gas Consumption of My Smart Contract and Why
GuessGas:Tell me Fine-Grained Gas Consumption of My Smart Contract and Why Open
In this paper, we present the development of an explainable Gas Estimator model EGE, utilizing big data to mine potential distribution of runtime information. Our approach overcomes the limitations of static gas analysis tools by labeling …
View article: GuessGas:Tell me Fine-Grained Gas Consumption of My Smart Contract and Why
GuessGas:Tell me Fine-Grained Gas Consumption of My Smart Contract and Why Open
In this paper, we present the development of an explainable Gas Estimator model EGE, utilizing big data to mine potential distribution of runtime information. Our approach overcomes the limitations of static gas analysis tools by labeling …
View article: PromptARA: Improving Deep Representation in Hybrid Automatic Readability Assessment with Prompt and Orthogonal Projection
PromptARA: Improving Deep Representation in Hybrid Automatic Readability Assessment with Prompt and Orthogonal Projection Open
Readability assessment aims to automatically classify texts based on readers’ reading levels. The hybrid automatic readability assessment (ARA) models using both deep and linguistic features have attracted rising attention in recent years …
View article: Enhancing Zero-Shot Chinese Character Recognition with Stroke-Radical Attention and Ensemble Mechanisms
Enhancing Zero-Shot Chinese Character Recognition with Stroke-Radical Attention and Ensemble Mechanisms Open
View article: SGCE-Font: Skeleton Guided Channel Expansion for Chinese Font Generation
SGCE-Font: Skeleton Guided Channel Expansion for Chinese Font Generation Open
The automatic generation of Chinese fonts is an important problem involved in many applications. The predominated methods for the Chinese font generation are based on the deep generative models, especially the generative adversarial networ…
View article: StrokeGAN+: Few-Shot Semi-Supervised Chinese Font Generation with Stroke Encoding
StrokeGAN+: Few-Shot Semi-Supervised Chinese Font Generation with Stroke Encoding Open
The generation of Chinese fonts has a wide range of applications. The currently predominated methods are mainly based on deep generative models, especially the generative adversarial networks (GANs). However, existing GAN-based models usua…
View article: STAR: Zero-Shot Chinese Character Recognition with Stroke- and Radical-Level Decompositions
STAR: Zero-Shot Chinese Character Recognition with Stroke- and Radical-Level Decompositions Open
Zero-shot Chinese character recognition has attracted rising attention in recent years. Existing methods for this problem are mainly based on either certain low-level stroke-based decomposition or medium-level radical-based decomposition. …
View article: Reducing Capacity Gap in Knowledge Distillation with Review Mechanism for Crowd Counting
Reducing Capacity Gap in Knowledge Distillation with Review Mechanism for Crowd Counting Open
The lightweight crowd counting models, in particular knowledge distillation (KD) based models, have attracted rising attention in recent years due to their superiority on computational efficiency and hardware requirement. However, existing…
View article: CodeGen-Test: An Automatic Code Generation Model Integrating Program Test Information
CodeGen-Test: An Automatic Code Generation Model Integrating Program Test Information Open
Automatic code generation is to generate the program code according to the given natural language description. The current mainstream approach uses neural networks to encode natural language descriptions, and output abstract syntax trees (…
View article: A Tale of HodgeRank and Spectral Method: Target Attack Against Rank Aggregation Is the Fixed Point of Adversarial Game
A Tale of HodgeRank and Spectral Method: Target Attack Against Rank Aggregation Is the Fixed Point of Adversarial Game Open
Rank aggregation with pairwise comparisons has shown promising results in elections, sports competitions, recommendations, and information retrieval. However, little attention has been paid to the security issue of such algorithms, in cont…
View article: Enhancing Automatic Readability Assessment with Pre-training and Soft Labels for Ordinal Regression
Enhancing Automatic Readability Assessment with Pre-training and Soft Labels for Ordinal Regression Open
The readability assessment task aims to assign a difficulty grade to a text. While neural models have recently demonstrated impressive performance, most do not exploit the ordinal nature of the difficulty grades, and make little effort for…
View article: Poisoning Attack Against Estimating From Pairwise Comparisons
Poisoning Attack Against Estimating From Pairwise Comparisons Open
As pairwise ranking becomes broadly employed for elections, sports competitions, recommendation, information retrieval and so on, attackers have strong motivation and incentives to manipulate or disrupt the ranking list. They could inject …