This three-volume set, CCIS 2637-2639 , constitutes the refereed proceedings of the 7th International Conference on Blockchain and Trustworthy Systems, BlockSys 2025, held in Zhuhai, China, during May 3031, 2025.

The 110 full papers included in this book were carefully reviewed and selected from 254 submissions. They were organized in topical sections as follows:
Part I: Blockchain Security & Attack Mitigation; Identity Authentication & Data Privacy; Blockchain Applications & Industry Solutions;
Part II: Deep Learning Architectures & Optimization; Computer Vision & Medical Image Analysis; Natural Language Processing & Multimodal Learning; Reinforcement Learning & Edge Computing Systems;
part III: Blockchain & Smart Contract Trustworthiness; Federated Learning & Edge Computing **** ; Industrial & IoT System Reliability; Privacy-Preserving & Secure Data Management

Inhalt

.- **Blockchain & Smart Contract Trustworthiness.
** .- Slime: An Intelligent Hybrid Recovery Strategy that Minimizes Recovery Overhead.
.- ASCRec: Recommending Accurate and Secure Codes for Smart Contract Developers.
.- A Smart Contract Vulnerability Detection Method Based on Graph Neural Networks and Zero-Shot Learning.
.- Decentralized Data Trading Solutions with Poisoning Attacks Defense.
.- USCSafe: Identifying Permission Vulnerabilities in Upgradeable Smart Contracts.
.- SAFE: A Blockchain-Based Framework for Secure Health Data Sharing with Symmetric Encryption.
.- The Impact of Cryptographic Algorithm Performance on Smart Contracts in WebAssembly.
.- A Survey of Sharding Blockchains on Performance and Scalability.
.- Investigating Tornado Cash: Empirical Insights into Mixing Service Anonymity.
.- **Federated Learning & Edge Computing Security.
** .-A Spam Detection Model With Federated Learning and Large Language Model.
.- Secure and Efficient Task Offloading with Resource Allocation in Vehicular Edge Computing.
.- Crowdsensing with Federated Trust Management: Adaptive Defense Against Malicious Contributions.
.- CDDSF: A Zero-Shot Domain Adaptation Framework for Secure Solidity Code Search.
.- Cost-aware Makespan Minimization for Workflow Scheduling in Trustworthy Heterogeneous Clouds.
.- A Mutimodal Smart Contract Classification Method Based on Hybrid Convolutional Neural Networks.
.- Computility Planning for Sustainable Supercomputing Centers.
.- Research on Web Vulnerability Mining System Based on Machine Learning.
.- A Survey on Trustworthy Systems in ChatGPT-Like Large-Scale Generative AI Models: Security, Interpretability, Reliability, and Privacy Considerations.
.- **Industrial & IoT System Reliability
** .- HMC-Driven Efficient Prediction of SiC Ion Annealing.
.- IEDD-Net: Industrial Endoscope Defect Detection Network Based on Improved YOLOv12n.
.- Research on Reliable Control Methods for Industrial Robotic Arms.
.- Adaptive Damage Detection Algorithm for Food Packaging Based on Deep Learning.
.- Service Quality Improvement Strategy in Public Services: An NN-SHAP Analysis Model for Satisfaction Survey.
.- Feature-Fusion-Based Adaptive Architectural Crack Detection System.
.- Research on Vegetation Generation Technology for 3D Scenes Based on Multi-Physics Field Coupling.
.- A 2D Engineering Drawing Element Parsing Method Based on Multi-Modal Frequency-Spatial Feature Pyramid Network.
.- Pi 4B-based Visual Navigation System for Low-computing Power Intelligent Vehicles.
.- CIHD-Net: A Cross-modal Interactive Hierarchical Dilated Network for RGB-D Salient Object Detection.
.- A Semi-fragile Reversible Watermarking for 3D Models Using Quantization Interval Division Modulation.
.- An Effective Reliability Prediction Model for Blockchain Services via Hybrid Multi-Layer Graph Attention and Self-attention.
.- DCD: A Semantic Segmentation Model for Fetal Ultrasound Four-Chamber View.
.- Detecting Inconsistent Privacy Statements between Mobile Vehicle Apps and their Third-party Integrations.
.- CSM:Cross-Scan Mamba for Remote Sensing Object Detection.
.- Robust Cross-Modal Semantic Communication for Object Detection.
.- Research on the Application of Pedagogical Agents in Interdisciplinary Teaching.
.- A Dynamic Swarm Reputation-DrivenConsensus Mechanism for TrustworthyEmbodied Intelligent Collaboration.