TamperBench: A Systematic Framework to Stress-Test LLM Safety Under Fine-Tuning and Tampering

Saad Hossain, Tom Tseng, Punya Syon Pandey, Samanvay Vajpayee, Matthew Kowal, Nayeema Nonta, Samuel Simko, Stephen Casper, Zhijing Jin, Kellin Pelrine, Sirisha Rambhatla

Published: 28 Feb 2026, Last Modified: 28 Feb 2026CAOEveryoneRevisionsBibTeXCC BY 4.0
Track: Tiny Paper Track, Up to 4 pages, excluding references
Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.
Attendance: Maybe, the presenting author may attend in person, depending on factors such as funding, visa approval, or travel constraints.
Presenter: ~Saad_Hossain1
Serve As Reviewer: ~Saad_Hossain1
Submission Number: 99