VibeGuard: Secure AI-Generated Code with Pre-Publish Gate
Protect AI-generated code with VibeGuard, a security gate framework that prevents vulnerabilities like source-map exposure and hardcoded secrets.
AI-Driven RAN Slicing: SLA Violations & Attack Recovery
Explore how adversarial attacks impact AI-driven RAN slicing, causing SLA violations and biased resource allocation, and learn about recovery strategies.
Secure LLM Instructions: Automated Framework Against Encoding Attacks
Discover an automated framework to evaluate and harden LLM system instructions, protecting sensitive data from encoding attacks without costly retraining.
Mitigating Popularity Bias in Personalized Recommendations
Discover how SPREE and Popularity Quantile Calibration improve user alignment and reduce popularity bias in recommender systems for better personalization.
Multi-LLM Pipelines: Analyzing Second-Pass Performance Gains
Explore how task type and draft quality impact second-pass gains in multi-LLM pipelines, revealing insights for optimizing AI revision systems.