Label Privacy in Split Learning for Large Models with Parameter-Efficient Training

Open in new window