Pengcheng Zhang
YOU?
Author Swipe
View article: BASFuzz: Towards Robustness Evaluation of LLM-based NLP Software via Automated Fuzz Testing
BASFuzz: Towards Robustness Evaluation of LLM-based NLP Software via Automated Fuzz Testing Open
Fuzzing has shown great success in evaluating the robustness of intelligent natural language processing (NLP) software. As large language model (LLM)-based NLP software is widely deployed in critical industries, existing methods still face…
View article: Soft and direct virtual inertia control method of low-voltage distribution network based on consistency algorithm
Soft and direct virtual inertia control method of low-voltage distribution network based on consistency algorithm Open
Distributed power supply, energy storage devices and multi-source loads of low-voltage distribution networks show non-normal distribution, and the coordinated control among multi-agents is very difficult, resulting in virtual inertia chara…
View article: Multi-objective optimization of capacity configuration for district heating and cooling system based on life cycle cost and annual carbon dioxide emissions
Multi-objective optimization of capacity configuration for district heating and cooling system based on life cycle cost and annual carbon dioxide emissions Open
View article: Research on fault diagnosis method for nuclear power plants rotating machinery based on MoCo Siamese neural network
Research on fault diagnosis method for nuclear power plants rotating machinery based on MoCo Siamese neural network Open
Rotating machinery is a kind of significant equipment that widely used in nuclear power plants (NPPs). The harsh operating environment and long-term continuous operation of the rotating machinery can cause various faults due to wear, vibra…
View article: Unlocking Dual-Focused Empowering Leadership Behaviors Toward Multilevel Creativity: A Qualitative View
Unlocking Dual-Focused Empowering Leadership Behaviors Toward Multilevel Creativity: A Qualitative View Open
Drawing on the Event System Theory, this study develops a comprehensive model for understanding dual-focused empowering behavior in team set-ups toward creativity. The study utilizes a triangular qualitative approach for thematic analysis …
View article: On‐Demand Removal of Rapid Hemostatic Sponge for Non‐Compressible Hemorrhage Through Disrupting Ionic Bonds
On‐Demand Removal of Rapid Hemostatic Sponge for Non‐Compressible Hemorrhage Through Disrupting Ionic Bonds Open
Self‐expanding hemostatic sponge plays an important role in the control of non‐compressible hemorrhage in deep wound. After hemostasis is accomplished, the sponge adheres to the wound via blood clots, posing a considerable challenge in wou…
View article: ABFS: Natural Robustness Testing for LLM-based NLP Software
ABFS: Natural Robustness Testing for LLM-based NLP Software Open
Owing to the exceptional performance of Large Language Models (LLMs) in Natural Language Processing (NLP) tasks, LLM-based NLP software has rapidly gained traction across various domains, such as financial analysis and content moderation. …
View article: Multi-omics analysis reveals the role of cancer-associated fibroblasts and their marker genes in clear cell renal cell carcinoma
Multi-omics analysis reveals the role of cancer-associated fibroblasts and their marker genes in clear cell renal cell carcinoma Open
View article: Combined bamboo-derived biochar and DCD improves the catalytic potentials of ureases in forest and agricultural soils
Combined bamboo-derived biochar and DCD improves the catalytic potentials of ureases in forest and agricultural soils Open
View article: Assessing the Robustness of LLM-based NLP Software via Automated Testing
Assessing the Robustness of LLM-based NLP Software via Automated Testing Open
Benefiting from the advancements in LLMs, NLP software has undergone rapid development. Such software is widely employed in various safety-critical tasks, such as financial sentiment analysis, toxic content moderation, and log generation. …
View article: TrainMover: An Interruption-Resilient and Reliable ML Training Runtime
TrainMover: An Interruption-Resilient and Reliable ML Training Runtime Open
Large-scale ML training jobs are frequently interrupted by hardware and software anomalies, failures, and management events. Existing solutions like checkpointing or runtime reconfiguration suffer from long downtimes, degraded performance,…
View article: The 2-complexity of even positive integers
The 2-complexity of even positive integers Open
The question of integer complexity asks about the minimal number of $1$'s that are needed to express a positive integer using only addition and multiplication (and parentheses). In this paper, we propose the notion of $l$-complexity of mul…
View article: Multivariate Zero-Inflated INAR(1) Model with an Application in Automobile Insurance
Multivariate Zero-Inflated INAR(1) Model with an Application in Automobile Insurance Open
View article: A Novel Four-CpG Prognostic Risk Score Model Predicts Survival and Guides Therapy in Patients with Head and Neck Squamous Cell Carcinoma
A Novel Four-CpG Prognostic Risk Score Model Predicts Survival and Guides Therapy in Patients with Head and Neck Squamous Cell Carcinoma Open
Accurate prognosis prediction of head and neck squamous cell carcinoma (HNSCC) is vital for guiding clinical decision-making. DNA methylation (DNAm) plays important roles in the regulation of gene expression and abnormal DNAm is a hallmark…
View article: Response surface test simulation of locking system reliability considering kinematic pair wear evolution
Response surface test simulation of locking system reliability considering kinematic pair wear evolution Open
The concentrated load endured by the locking system during the weapon mounting and release procedures may lead to mechanical wear in hinges and kinematic pairs, thereby influencing the functional reliability of the weapon rack mechanism sy…
View article: Geospatial learning for large-scale transport infrastructure depth prediction
Geospatial learning for large-scale transport infrastructure depth prediction Open
Transportation infrastructure supports the smooth mobility of humans, commodities, and services. Pavement depth measures the quality of road infrastructure through representing the thickness of road surfaces, and influences various aspects…
View article: Application of Artificial Intelligence (AI) in Recruitment and Selection: The Case of Company A and Company B
Application of Artificial Intelligence (AI) in Recruitment and Selection: The Case of Company A and Company B Open
The study explores the integration of Artificial Intelligence (Al) in recruitment and selection processes at Company A and Company B, reflecting the increasing trend of Al utilization in HR practices. The research aims to investigate HR pr…
View article: Revealing the impact of autophagy-related genes in rheumatoid arthritis: Insights from bioinformatics
Revealing the impact of autophagy-related genes in rheumatoid arthritis: Insights from bioinformatics Open
This study identified marker autophagy-related genes in rheumatoid arthritis and analyzed their impact on the disease, providing new perspectives for understanding the role of autophagy-related genes in rheumatoid arthritis and providing n…
View article: Towards effective person search with deep learning: A survey from systematic perspective
Towards effective person search with deep learning: A survey from systematic perspective Open
View article: Design of a Detachable Surface Garbage Cleaning Robot Suitable for Small Water Areas
Design of a Detachable Surface Garbage Cleaning Robot Suitable for Small Water Areas Open
Surface garbage in small water bodies poses a severe threat to the safety of residents’ water usage. Focusing on the characteristics of surface garbage in small water bodies, a robot capable of collecting light floating debris and animal c…
View article: Geospatial Learning for Large-Scale Transport Infrastructure Depth Prediction
Geospatial Learning for Large-Scale Transport Infrastructure Depth Prediction Open
View article: Cone-Beam Computed Laminography with Anisotropic Adaptive Weighted Total Variation Minimization Based on the Framework of the Chambolle-Pock Algorithm
Cone-Beam Computed Laminography with Anisotropic Adaptive Weighted Total Variation Minimization Based on the Framework of the Chambolle-Pock Algorithm Open
View article: Dynamic Monitoring of Bridge Structures via an Integrated Cloud and Edge Computing System
Dynamic Monitoring of Bridge Structures via an Integrated Cloud and Edge Computing System Open
View article: Construction of the particle simulation model for ginger-soil system using discrete element method
Construction of the particle simulation model for ginger-soil system using discrete element method Open
View article: Mae-Recon: Multi-Dimensional Attention-Enhanced Reconstruction for Sparse-View Cbct
Mae-Recon: Multi-Dimensional Attention-Enhanced Reconstruction for Sparse-View Cbct Open
View article: A robust four CpGs-based prognostic risk score model predicts survival and guides therapy in patients with head and neck squamous cell carcinoma
A robust four CpGs-based prognostic risk score model predicts survival and guides therapy in patients with head and neck squamous cell carcinoma Open
Background Accurate prognostic prediction in Head and neck squamous cell carcinoma (HNSCC) patients remains challenging owing to its heterogeneity and intra- and inter-pathological variability. Numerous prognostic models for HNSCC have bee…
View article: Variation of Ferroptosis-Related Markers in HaCaT Cell Photoaging Models Induced by UVB
Variation of Ferroptosis-Related Markers in HaCaT Cell Photoaging Models Induced by UVB Open
According to the findings, ferroptosis may be linked to UVB-induced skin photoaging, which could be attenuated by inhibition of ferroptosis.
View article: Towards Fully Decoupled End-to-End Person Search
Towards Fully Decoupled End-to-End Person Search Open
End-to-end person search aims to jointly detect and re-identify a target person in raw scene images with a unified model. The detection task unifies all persons while the re-id task discriminates different identities, resulting in conflict…
View article: LEAP: Efficient and Automated Test Method for NLP Software
LEAP: Efficient and Automated Test Method for NLP Software Open
The widespread adoption of DNNs in NLP software has highlighted the need for robustness. Researchers proposed various automatic testing techniques for adversarial test cases. However, existing methods suffer from two limitations: weak erro…
View article: Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound
Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound Open
Deep neural networks (DNNs) have been widely and successfully adopted and deployed in various applications of speech recognition. Recently, a few works revealed that these models are vulnerable to backdoor attacks, where the adversaries ca…