Weihao Luo

Meta-UCF: Unified Task-Conditioned LoRA Generation for Continual Learning in Large Language Models

Synergy over Discrepancy: A Partition-Based Approach to Multi-Domain LLM Fine-Tuning

Whose Instructions Count? Resolving Preference Bias in Instruction Fine-Tuning

Dynamic Cross-Layer Prefix Alignment for Resolving Label Preference Discrepancies in LLMs Fine-Tuning