AI safety researcher

Berk Sunar

Google Scholar dblp Homepage

Last 12 months2 papers

AI safety papers from the last year

On this site: 10 papers

May 2026

Revisiting JBShield: Breaking and Rebuilding Representation-Level Jailbreak Defenses
May 4arXiv ↗

December 2025

Super Suffixes: Bypassing Text Generation Alignment and Guard Models Simultaneously
Dec 12arXiv ↗

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.