5 papers
Differentially Private Preference Data Synthesis for Large Language Model Alignment
Fengyu Gao, Jing Yang
Preference alignment is a crucial post-training step for large language models (LLMs) to ensure their outputs align with human values. However, post-training on real human preferen…
HeteroFedSyn: Differentially Private Tabular Data Synthesis for Heterogeneous Federated Settings
Xiaochen Li, Fengyu Gao, Xizixiang Wei +3
Traditional Differential Privacy (DP) mechanisms are typically tailored to specific analysis tasks, which limits the reusability of protected data. DP tabular data synthesis overco…
Data-adaptive Differentially Private Prompt Synthesis for In-Context Learning
Fengyu Gao, Ruida Zhou, Tianhao Wang +2
Large Language Models (LLMs) rely on the contextual information embedded in examples/demonstrations to perform in-context learning (ICL). To mitigate the risk of LLMs potentially l…
Federated Online Prediction from Experts with Differential Privacy: Separations and Regret Speed-ups
Fengyu Gao, Ruiquan Huang, Jing Yang
We study the problems of differentially private federated online prediction from experts against both stochastic adversaries and oblivious adversaries. We aim to minimize the avera…
Federated Q-Learning: Linear Regret Speedup with Low Communication Cost
Zhong Zheng, Fengyu Gao, Lingzhou Xue +1
In this paper, we consider federated reinforcement learning for tabular episodic Markov Decision Processes (MDP) where, under the coordination of a central server, multiple agents…