Zhibo Wang

Attack-Resistant Watermarking for AIGC Image Forensics via Diffusion-based Semantic Deflection

Mitigating Over-Refusal in Adversarial Tuning via Subspace-guided Sample Selection

Towards Real-world Debiasing: Rethinking Evaluation, Challenge, and Solution

CARP: Causal Alignment of Reward Models via Response-to-Prompt Prediction

CollabMask: Explainable Neuron Collaboration with Gradient Masks for LLM Fine-Tuning

Textual Unlearning Gives a False Sense of Unlearning

ICLScan: Detecting Backdoors in Black-Box Large Language Models via Targeted In-context Illumination

Towards Real World Debiasing: A Fine-grained Analysis On Spurious Correlation