Cybersecurity at the intersection of AI covers both how machine learning strengthens defense and how it opens new attack surfaces. We report on adversarial robustness, model and data security, and the practical risks that come with deploying AI in production, grounded in current research.
Researchers from Jilin University and the University of Technology Sydney built a federated continual recommendation framework that solves two problems nobody had tackled together: forgetting who you used to be, and not knowing enough about people like you — all…
FCUCR: Federated Continual Recommendation That Remembers You Without Storing Your Data Read More »
Researchers at Zhengzhou University discovered that the most dangerous attacks on autonomous vehicle prediction systems don’t look like attacks at all—they look like perfectly normal driving maneuvers, reconstructed from scratch by a generative model that understands motion better than the…
A novel training paradigm that integrates adversarial perturbations with instance-sensitive optimization to enhance robustness and generality in graph neural network-based collaborative filtering, achieving up to 9.2% NDCG improvement over state-of-the-art methods.
ACCF: Adversarial Contrastive Collaborative Filtering Read More »
A novel defense framework that integrates Deep Q-Network algorithms with Mahalanobis distance-based maliciousness evaluation to dynamically select benign users, achieving up to 9% accuracy improvement over state-of-the-art approaches in non-IID federated learning environments.
A novel PUF-based framework embeds unclonable hardware signatures into teacher models during knowledge distillation, enabling post-theft traceability without architectural changes or runtime overhead.
PDF: PUF-based DNN Fingerprinting for Knowledge Distillation Traceability Read More »
Skin cancer remains one of the most pervasive and life-threatening health conditions globally, with over 5 million new cases reported annually in the United States alone. Among the various types, malignant melanoma stands out as particularly alarming—accounting for approximately 4%…
A segmentation model has a quiet advantage that a plain image classifier does not. When it sees a person standing next to a bicycle, it uses that relationship to sharpen its guess about where the bicycle actually ends and the…
SegTrans Explained, Why Segmentation Models Cannot Hide Behind Context Read More »
Generative watermarking has become the default answer to a question every LLM provider eventually faces. If someone copies my model’s outputs, how do I prove it. Methods like KGW nudge token probabilities toward a hidden green list of words, so…
Knowledge Distillation Can Forge and Erase LLM Watermarks Read More »