Guodong DU

Multi-objective Large Language Model Alignment with Hierarchical Experts

Skill Weaving: Efficient Self-Improvement of LLMs via Modular Skillpacks

Language Bias in LVLMs: From In-Depth Analysis to Simple and Effective Mitigation

Knowledge Fusion of Large Language Models via Modular SkillPacks

Vocabulary Fixation Reveals Visual Attention Sink for Hallucination Mitigation in LVLMs

NeurIPT: Foundation Model for Neural Interfaces

Multi-modality Expansion and Retention for LLMs through Parameter Merging and Decoupling