Zihao Zhao
@zihaozhao.bsky.social
PhD student @jhuclsp.bsky.social| AI safety & privacy
Previous: Undergrad @jhucompsci.bsky.social
Previous: Undergrad @jhucompsci.bsky.social
🚀 Text anonymization is hard; DP often hurts utility.
We use entity-aware control codes + either ICL (with bad-token blocking) or prefix-tuning w/ masking to get strong privacy–utility tradeoffs on legal & clinical data, outperforming DP-SGD in practice (EMNLP 2025).
www.arxiv.org/abs/2509.25729
We use entity-aware control codes + either ICL (with bad-token blocking) or prefix-tuning w/ masking to get strong privacy–utility tradeoffs on legal & clinical data, outperforming DP-SGD in practice (EMNLP 2025).
www.arxiv.org/abs/2509.25729
October 15, 2025 at 8:24 PM
🚀 Text anonymization is hard; DP often hurts utility.
We use entity-aware control codes + either ICL (with bad-token blocking) or prefix-tuning w/ masking to get strong privacy–utility tradeoffs on legal & clinical data, outperforming DP-SGD in practice (EMNLP 2025).
www.arxiv.org/abs/2509.25729
We use entity-aware control codes + either ICL (with bad-token blocking) or prefix-tuning w/ masking to get strong privacy–utility tradeoffs on legal & clinical data, outperforming DP-SGD in practice (EMNLP 2025).
www.arxiv.org/abs/2509.25729