AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sheshadri, Abhay, Ewart, Aidan, Fronsdal, Kai, Gupta, Isha, Bowman, Samuel R., Price, Sara, Marks, Samuel, Wang, Rowan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!