影响力指数

52.49/100

前 7.3%

全站排名 #4,710

发表论文9 篇

平均评分5.6

年均产出3.0 篇/年

Xiaoyu Wu

PhD student@Rice University·美国·OpenReview

研究方向

Data Extraction · Membership Inference Attack · Adversarial Attack and Defense · Diffusion Model · Trustworthness · AI for social good

FedMomentum: Preserving LoRA Training Momentum in Federated Fine-Tuning

ICLR 2026Rejected

POLAR: Policy-based Layerwise Reinforcement Learning for Stealthy Backdoor Attacks in Federated Learning

ICLR 2026Rejected

Targeted Attack Improves Protection against Unauthorized Diffusion Customization

ICLR 2025Spotlight

Leveraging Model Guidance to Extract Training Data from Personalized Diffusion Models

ICML 2025Poster

Unlearned but Not Forgotten: Data Extraction after Exact Unlearning in LLM

NeurIPS 2025Poster

Revealing the Unseen: Guiding Personalized Diffusion Models to Expose Training Data

ICLR 2025Rejected

Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks

ICLR 2025Rejected

合作者 (19)