security vulnerabilities

4 papers

• Attack via Overfitting: 10-shot Benign Fine-tuning to Jailbreak LLMs
• BadVLA: Towards Backdoor Attacks on Vision-Language-Action Models via Objective-Decoupled Optimization
• From Counterfactuals to Trees: Competitive Analysis of Model Extraction Attacks
• MIP against Agent: Malicious Image Patches Hijacking Multimodal OS Agents