Hui Xiong

10 papers Hong Kong University of Science and Technology (Guangzhou)

• BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model
• GVPO: Group Variance Policy Optimization for Large Language Model Post-Training
• Learning to Think: Information-Theoretic Reinforcement Fine-Tuning for LLMs
• Logic-in-Frames: Dynamic Keyframe Search via Visual Semantic-Logical Verification for Long Video Understanding
• Multimodal 3D Genome Pre-training
• Optimizing the Unknown: Black Box Bayesian Optimization with Energy-Based Model and Reinforcement Learning
• Robust Explanations of Graph Neural Networks via Graph Curvatures
• ST$^2$360D: Spatial-to-Temporal Consistency for Training-free 360 Monocular Depth Estimation
• See&Trek: Training-Free Spatial Prompting for Multimodal Large Language Model
• Unveiling the Learning Mind of Language Models: A Cognitive Framework and Empirical Study