Jun Luo

More Than Meets the Eye? Uncovering the Reasoning-Planning Disconnect in Training Vision-Language Driving Models

TrojanPraise: Jailbreak LLMs via Benign Fine-Tuning

XPoison: Cross-Class Attacks through Clean-Label Data Poisoning in Fine-Tuning

Attack via Overfitting: 10-shot Benign Fine-tuning to Jailbreak LLMs

Co-Reinforcement Learning for Unified Multimodal Understanding and Generation