Grace Li Zhang

Layer-wise dynamic rank for compressing large language models

PEL-NAS: Search Space Partitioned Architecture Prompt Co-evolutionary LLM-driven Hardware-Aware Neural Architecture Search

CompressKV: Semantic Retrieval Heads Know What Tokens are Not Important Before Generation

Towards Fast LLM Fine-tuning through Zeroth-Order Optimization with Projected Gradient-Aligned Perturbations

Basis Sharing: Cross-Layer Parameter Sharing for Large Language Model Compression