一作ICLR 2026Poster
LipNeXt: Scaling up Lipschitz-based Certified Robustness to Billion-parameter Models
Kai Hu · Haoqi Hu · Matt Fredrikson
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Kai Hu · Haoqi Hu · Matt Fredrikson
Kai Hu · Weichen Yu · Li Zhang · Alexander Robey · Andy Zou · Haoqi Hu · Chengming Xu · Matt Fredrikson
Kai Hu · Abhinav Aggarwal · Mehran Khodabandeh · David Zhang · Eric Hsin · Li Chen · Ankit Jain · Matt Fredrikson · Akash Bharadwaj
This paper presents a novel Automated Red Teaming (ART) framework that shifts from example-based to policy-based evaluation, addressing critical limitations in scalability and validity. We define harmful content through abstract safety policies rather than specific static examples. We also introd...