影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
37.59/100
前 16.3%
全站排名 #10,475
发表论文6 篇
平均评分
年均产出3.0 篇/年
Guangdong Bai
研究方向
Large language model · Privacy compliance · Machine learning privacy · Mobile security
20
Catch-Only-One: Non-Transferable Examples for Model-Specific Authorization
ICLR 2026Desk Rejected
通讯21
ReTrace: Reinforcement Learning-Guided Reconstruction Attacks on Machine Unlearning
ICLR 2026Poster
通讯12
Evading Safety Alignment in Open-source LLM Deployment by Embedding Semantic Shift
ICLR 2026Rejected
20
GOLD: Graph Out-of-Distribution Detection via Implicit Adversarial Latent Generation
ICLR 2025Spotlight
三作25
FracFace: Breaking The Visual Clues—Fractal-Based Privacy-Preserving Face Recognition
NeurIPS 2025Poster
35
MAA: Meticulous Adversarial Attack against Vision-Language Pre-trained Models
ICLR 2025Rejected
三作