21 papers
Where Can We Fool the Model's Mind? Response-Guided Semantic Probing and Adaptive Routing for Transferable Attacks on Black-Box MLLMs
Ahmed Abdelnaby, Mohamed Elmahallawy
Multimodal Large Language Models (MLLMs) ground language in visual evidence, yet their semantic reasoning can be exploited through transferable black-box adversarial attacks. Exist…
Veil-Then-Reveal: Protecting PII Privacy via Reversible Pseudo-Anonymization in Zero-Trust User--LLM Interaction
Mohamed Shaaban, Mohamed Elmahallawy, Thomas E. Carroll
Large language models (LLMs) increasingly process sensitive documents and personal information, yet sending prompts containing personally identifiable information (PII) to remote p…
Prune the Leaky Neurons: Mitigating Private Data Leakage in Fine-Tuned LLMs via Joint-Embedding-Guided Pruning
Mohamed Shaaban, Mohamed Elmahallawy
Fine-tuning large language models (LLMs) on proprietary records enables domain adaptation but introduces two challenges: (i) privacy leakage of memorized private data, such as pers…
Tokenized Key-Gated Adapter Routing: A Secure Access Control Mechanism Against Private Data Leakage in LLMs
Mohamed Shaaban, Mohamed Elmahallawy
Large language models (LLMs) are increasingly deployed in privacy-critical domains (e.g., healthcare, finance, and government), but their propensity to memorize and disclose person…
Label Less, Learn More: Resource-Efficient Active Semi-Supervised Learning for Onboard Satellite Image Annotation
Ahmed Abdelnaby, Mohamed Elmahallawy, Marius Bernahrndt +1
Large-scale pervasive sensing increasingly relies on high-resolution satellite imagery, yet task-specific onboard vision is constrained by costly annotation and limited computation…
Resource-Aware Federated Mixture-of-Experts with Adaptive Pruning for Onboard Learning in LEO Satellite Constellations
Mohamed Shaaban, Mohamed Elmahallawy, Marius Bernahrndt +1
Low-Earth-orbit (LEO) satellites are increasingly expected to perform onboard learning for applications such as disaster response and environmental monitoring. However, conventiona…
Region-Level Black-Box Defense Against Stealthy Embedding-Space Backdoors in CLIP
Ahmed Abdelnaby, Mohamed Elmahallawy
Contrastive Language--Image Pretraining (CLIP) has emerged as a dominant vision backbone due to its strong transferability and zero-shot capabilities. However, recent studies revea…
When Clients Are Orchestrated: Strategic Gradient Manipulation to Defeat Federated Learning Servers with Efficient Defense
Mohamed Shaaban, Ahmed Abdelnaby, Mohamed Elmahallawy
Federated Learning enables decentralized model training by exchanging model updates--rather than raw data--with a central parameter server (PS). While most of the existing defenses…
Beyond Small Patches: Black-Box Detection and Purification of Diverse Backdoor Triggers
Ahmed Abdelnaby, Mohamed Elmahallawy
Deep neural networks (DNNs) are increasingly deployed in real-world vision systems, yet their predictions can be covertly manipulated by backdoor attacks, in which malicious trigge…
Learning What Can Be Picked: Active Reachability Estimation for Efficient Robotic Fruit Harvesting
Nur Afsa Syeda, Mohamed Elmahallawy, Luis Fernando de la Torre +1
Agriculture remains a cornerstone of global health and economic sustainability, yet labor-intensive tasks such as harvesting high-value crops continue to face growing workforce sho…
Prototype Fusion: A Training-Free Multi-Layer Approach to OOD Detection
Shreen Gul, Mohamed Elmahallawy, Ardhendu Tripathy +1
Deep learning models are increasingly deployed in safety-critical applications, where reliable out-of-distribution (OOD) detection is essential to ensure robustness. Existing metho…
SecureGate: Learning When to Reveal PII Safely via Token-Gated Dual-Adapters for Federated LLMs
Mohamed Shaaban, Mohamed Elmahallawy
Federated learning (FL) enables collaborative training across organizational silos without sharing raw data, making it attractive for privacy-sensitive applications. With the rapid…
Explaining the Unseen: Multimodal Vision-Language Reasoning for Situational Awareness in Underground Mining Disasters
Mizanur Rahman Jewel, Mohamed Elmahallawy, Sanjay Madria +1
Underground mining disasters produce pervasive darkness, dust, and collapses that obscure vision and make situational awareness difficult for humans and conventional systems. To ad…
Decentralized Trust for Space AI: Blockchain-Based Federated Learning Across Multi-Vendor LEO Satellite Networks
Mohamed Elmahallawy, Asma Jodeiri Akbarfam
The rise of space AI is reshaping government and industry through applications such as disaster detection, border surveillance, and climate monitoring, powered by massive data from…
Secure and Privacy-Preserving Federated Learning for Next-Generation Underground Mine Safety
Mohamed Elmahallawy, Sanjay Madria, Samuel Frimpong
Underground mining operations depend on sensor networks to monitor critical parameters such as temperature, gas concentration, and miner movement, enabling timely hazard detection…
Detecting Untargeted Attacks and Mitigating Unreliable Updates in Federated Learning for Underground Mining Operations
Md Sazedur Rahman, Mohamed Elmahallawy, Sanjay Madria +1
Underground mining operations rely on distributed sensor networks to collect critical data daily, including mine temperature, toxic gas concentrations, and miner movements for haza…
LPLgrad: Optimizing Active Learning Through Gradient Norm Sample Selection and Auxiliary Model Training
Shreen Gul, Mohamed Elmahallawy, Sanjay Madria +1
Machine learning models are increasingly being utilized across various fields and tasks due to their outstanding performance and strong generalization capabilities. Nonetheless, th…
DIS-Mine: Instance Segmentation for Disaster-Awareness in Poor-Light Condition in Underground Mines
Mizanur Rahman Jewel, Mohamed Elmahallawy, Sanjay Madria +1
Detecting disasters in underground mining, such as explosions and structural damage, has been a persistent challenge over the years. This problem is compounded for first responders…
FisherMask: Enhancing Neural Network Labeling Efficiency in Image Classification Using Fisher Information
Shreen Gul, Mohamed Elmahallawy, Sanjay Madria +1
Deep learning (DL) models are popular across various domains due to their remarkable performance and efficiency. However, their effectiveness relies heavily on large amounts of lab…
CAV-AD: A Robust Framework for Detection of Anomalous Data and Malicious Sensors in CAV Networks
Md Sazedur Rahman, Mohamed Elmahallawy, Sanjay Madria +1
The adoption of connected and automated vehicles (CAVs) has sparked considerable interest across diverse industries, including public transportation, underground mining, and agricu…
FedHAP: Fast Federated Learning for LEO Constellations Using Collaborative HAPs
Mohamed Elmahallawy, Tie Luo
Low Earth Orbit (LEO) satellite constellations have seen a surge in deployment over the past few years by virtue of their ability to provide broadband Internet access as well as to…