Pan Zhou

7 papers Singapore Management University

• BadVLA: Towards Backdoor Attacks on Vision-Language-Action Models via Objective-Decoupled Optimization
• Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
• EAP-GP: Mitigating Saturation Effect in Gradient-based Automated Circuit Identification
• Fit the Distribution: Cross-Image/Prompt Adversarial Attacks on Multimodal Large Language Models
• GRIFFIN: Effective Token Alignment for Faster Speculative Decoding
• SoPo: Text-to-Motion Generation Using Semi-Online Preference Optimization
• Towards Building Model/Prompt-Transferable Attackers against Large Vision-Language Models