作者位次 #7/8ICLR 2026Poster· Social Aspects / Trustworthy Machine Learning
When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretraining
Zhihao Li · Gezheng Xu · Jiale Cai · Ruiyi Fang · Di Wu · Qicheng Lao · Charles Ling · Boyu Wang