1+1>2: A Synergistic Sparse and Low-Rank Compression Method for Large Language Models

Open in new window