Xuefeng Du

Unlocking the Pre-Trained Model as a Dual-Alignment Calibrator for Post-Trained LLMs

Limited Preference Data? Learning Better Reward Model with Latent Space Synthesis

Steer LLM Latents for Hallucination Detection

Understanding Multimodal LLMs Under Distribution Shifts: An Information-Theoretic Approach

VLMGuard: Defending VLMs against Malicious Prompts via Unlabeled Data

How Reliable Is Human Feedback For Aligning Large Language Models?