Full Description
This four-volume set LNISCT 687-690 constitutes the proceedings of the 21st EAI International Conference on Security and Privacy in Communication Networks, SecureComm 2025, held in Xiangtan, China, during July 4 - 6, 2025.
The 119 full papers included in these volumes were carefully reviewed and selected from 341 submissions. They are organized in the following topical sections:
Part I: Distributed and Network Security; ML/AI Security.
Part II: ML/AI Security; CyberSecurity.
Part III: CyberSecurity; Cryptography and Authentication.
Part IV: Cryptography and Authentication; Security and Optimization.
Contents
.- ML/AI Security.
.- SVD-Based Efficient Communication Scheme for Heterogeneous Federated Learning.
.- IncreTTP: An Incremental TTPs Classification Model.
.- Login, Logout, Reset: Measuring Security and Privacy Issues in Real-World Web Logins.
.- Location Privacy Protection for VANETs based on Dynamically Adjustable Security Coefficient.
.- BPGT: A Novel Privacy-Preserving K-Means Clustering Framework to Guarantee Local $d_{\chi}$-privacy.
.- Hamk-CFI: A Hardware-assisted Kernel Control Flow Protection Framework For Cloud Environments.
.- PEARL: A Reinforcement Learning-Based Attack Analysis Approach for Security and Robustness Assessment of Smart Home Systems.
.- The RPKI and Its Hidden Guardians: An Empirical Analysis of Their Relationship and Security Implications.
.- FedCC: Robust Federated Learning against Model Poisoning Attacks.
.- FedCLF: Contrastive Learning-driven Framework for Mitigating Data Heterogeneity in Federated Learning.
.- FedSND:Federated Learning with Symmetric Noise Decentralized Orthogonal Encryption.
.- Group-Based Parallel Split Federated Learning.
.- A Verifiable Federated Learning Aggregation Scheme Based on Homomorphic Hashing.
.- A Lightweight Image Steganography Scheme Based on Invertible Neural Network Architecture with Progressive Channel Attention.
.- Defending Against Malicious Clients in Robust Heterogeneous Federated Learning.
.- CyberSecurity.
.- A Few-Shot-Based Model-Agnostic Meta-Learning for Intrusion Detection in Secure of In-vehicle Network.
.- A Robustness Optimization Mechanism for Intrusion Detection Models Based on Dynamic Ensemble Learning.
.- BFDet: A Method for Detecting Malicious Traffic with Ultra-Low False Positive Rate Based on TLS Behavior Flow.
.- DARD: Dice Adversarial Robustness Distillation Against Adversarial Attacks.
.- EvasionEval: A Benchmark for LLMs in Evaluating Advanced Defense Evasion Techniques.
.- Flow Dissector: A Flow Slicing Representation with a Pre-trained Model for Varied Malicious Traffic Classification.
.- From Dark Network to Honeypot: Analysis and Traceability of Multi-prefix IPv6 Network Attacks.
.- Insights into Ransomware Detection based on Semantic Understanding.
.- Log Anomaly Detection based on Time-Delta Sequential Feature.
.- Log Semantic Parsing based on Sequence Annotation in Security Operations.
.- MOT-Fuzz: A Novel Directed Greybox Fuzzing with Multiple Ordered Target Basic Blocks for Multistep Vulnerabilities.
.- MFRWF: Enhance Website Fingerprinting Robustness Against website content updates and background noise traffic.
.- One Trace is Possible: A Method for Small-Sample Profiling Side-Channel Analysis in Communication Environments.
.- Task-Driven GAN for Class-Imbalanced Intrusion Detection.
.- Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text.
