This three-set volume LNCS 16217-16219 constitutes the refereed proceedings of 27th International Conference on Information and Communications Security, ICICS 2025, held in Nanjing, China, during October 2931, 2025.

The 91 full papers presented in this book were carefully selected and reviewed from 357 submissions. The papers are organized in the following topical sections:

Part I: Cryptography; Post-quantum Cryptography; Anonymity and Privacy; Authentication and Authorization.

Part II: Blockchain and Cryptocurrencies, System and Network Security, Security and Privacy of AI, Machine Learning for Security.

Part III: Attack and Defense; Vulnerability Analysis; Anomaly Detection; Traffic Classification; Steganography and Watermarking.

Klappentext

.- Blockchain and Cryptocurrencies. .- EquinoxBFT: BFT Consensus for Blockchain Emergency Governance. .- fFuzz: A State-aware Function-level Fuzzing Framework for Smart Contract Vulnerabilities Detection. .- TraceBFT: Backtracking-based Pipelined Asynchronous BFT Consensus for High-Throughput Distributed Systems. .- RADIAL: Robust Adversarial Discrepancy-aware Framework for Early Detection of Illicit Cryptocurrency Accounts. .- Enhancing Private Signing Key Protection in Digital Currency Transactions Using Obfuscation. .- AnsBridge: Towards Secure Cross-Chain Interoperability via Anonymous and Verifiable Validators. .- TrustBlink: A zkSNARK-Powered On-Demand Relay for PoW Cross-Chain Verification With Low Cost. .- R1-MFSol: a Smart Contract Vulnerability Detection Model Based on LLM and Multi-modal Feature Fusion. .- No Place to Hide: An Efficient and Accurate Backdoor Detection Tool for Ethereum ERC-20 Smart Contracts. .- System and Network Security. .- Batch-oriented Element-wise Approximate Activation for Privacy-Preserving Neural Networks. .- Social-Aware and Quality-Driven Incentives for Mobile Crowd-Sensing with Two-Stage Game. .- A Distributed Privacy Protection Method for Crowd Sensing Based on Trust Evaluation. .- DBG-LB: A Trustworthy and Efficient Framework for Data Sharing in the Internet of Vehicles. .- Actions Speak Louder Than Words: Evidence-Based Trust Level Evaluation in Multi-Agent Systems. .- Bridging the Interoperability Gaps Among Trusted Architectures in MCUs. .- Security and Privacy of AI. .- A Dropout-Resilient and Privacy-Preserving Framework for Federated Learning via Lightweight Masking. .- AFedGAN: Adaptive Federated Learning with Generative Adversarial Networks for Non-IID Data. .- OTTER: Optimized Training with Trustworthy Enhanced Replication via Diffusion and Federated VMUNet for Privacy-Aware Medical Segmentatio. .- EAGLE: Ensemble Adaptive Graph Learning for Enhanced Ethereum Fraud Detection. .- BR-CPPFL: A Blockchain-based Robust Clustered Privacy-preserving Federated Learning System. .- Efficient Semi-asynchronous Federated Learning with Guided Selective Participation and Adaptive Aggregation. .- Improving Byzantine-resilience in Federated Learning via Diverse Aggregation and Adaptive Variance Reduction. .- Hierarchical Recovery of Convolutional Neural Networks via Self-Embedding Watermarking. .- Personalized Federated Learning Algorithm Based on User Grouping and Group Signatures. .- Machine Learning for Security. .- SPCD: A Shot-Based Partial Copy Detection Method. .- Bayesian-Adaptive Graph Neural Network for Anomaly Detection (BAGNN). .- UzPhishNet Model for Phishing Detection. .- CyberNER-LLM: Cyber Threat Intelligence Named Entity Recognition With Large Language Model. .- Provenance-Based Intrusion Detection via Multi-Scale Graph Representation Learning. .- SADGA: A Self Attention GAN-Based Adversarial DGA with High Anti-Detection Ability.

Inhalt

.- Blockchain and Cryptocurrencies.

.- EquinoxBFT: BFT Consensus for Blockchain Emergency Governance.

.- fFuzz: A State-aware Function-level Fuzzing Framework for Smart Contract Vulnerabilities Detection.

.- TraceBFT: Backtracking-based Pipelined Asynchronous BFT Consensus for High-Throughput Distributed Systems.

.- RADIAL: Robust Adversarial Discrepancy-aware Framework for Early Detection of Illicit Cryptocurrency Accounts.

.- Enhancing Private Signing Key Protection in Digital Currency Transactions Using Obfuscation.

.- AnsBridge: Towards Secure Cross-Chain Interoperability via Anonymous and Verifiable Validators.

.- TrustBlink: A zkSNARK-Powered On-Demand Relay for PoW Cross-Chain Verification With Low Cost.

.- R1-MFSol: a Smart Contract Vulnerability Detection Model Based on LLM and Multi-modal Feature Fusion.

.- No Place to Hide: An Efficient and Accurate Backdoor Detection Tool for Ethereum ERC-20 Smart Contracts.

.- System and Network Security.

.- Batch-oriented Element-wise Approximate Activation for Privacy-Preserving Neural Networks.

.- Social-Aware and Quality-Driven Incentives for Mobile Crowd-Sensing with Two-Stage Game.

.- A Distributed Privacy Protection Method for Crowd Sensing Based on Trust Evaluation.

.- DBG-LB: A Trustworthy and Efficient Framework for Data Sharing in the Internet of Vehicles.

.- Actions Speak Louder Than Words: Evidence-Based Trust Level Evaluation in Multi-Agent Systems.

.- Bridging the Interoperability Gaps Among Trusted Architectures in MCUs.

.- Security and Privacy of AI.

.- A Dropout-Resilient and Privacy-Preserving Framework for Federated Learning via Lightweight Masking.

.- AFedGAN: Adaptive Federated Learning with Generative Adversarial Networks for Non-IID Data.

.- OTTER: Optimized Training with Trustworthy Enhanced Replication via Diffusion and Federated VMUNet for Privacy-Aware Medical Segmentatio.

.- EAGLE: Ensemble Adaptive Graph Learning for Enhanced Ethereum Fraud Detection.

.- BR-CPPFL: A Blockchain-based Robust Clustered Privacy-preserving Federated Learning System.

.- Efficient Semi-asynchronous Federated Learning with Guided Selective Participation and Adaptive Aggregation.

.- Improving Byzantine-resilience in Federated Learning via Diverse Aggregation and Adaptive Variance Reduction.

.- Hierarchical Recovery of Convolutional Neural Networks via Self-Embedding Watermarking.

.- Personalized Federated Learning Algorithm Based on User Grouping and Group Signatures.

.- Machine Learning for Security.

.- SPCD: A Shot-Based Partial Copy Detection Method.

.- Bayesian-Adaptive Graph Neural Network for Anomaly Detection (BAGNN).

.- UzPhishNet Model for Phishing Detection.

.- CyberNER-LLM: Cyber Threat Intelligence Named Entity Recognition With Large Language Model.

.- Provenance-Based Intrusion Detection via Multi-Scale Graph Representation Learning.

.- SADGA: A Self Attention GAN-Based Adversarial DGA with High Anti-Detection Ability.