AS TRONG REJECT for Empty Jailbreaks
–Neural Information Processing Systems
We show that existing benchmarks suffer from significant shortcomings and introduce the StrongREJECT benchmark to address these issues.
Neural Information Processing Systems
Oct-10-2025, 19:29:36 GMT
- Genre:
- Research Report > New Finding (1.00)
- Industry:
- Law Enforcement & Public Safety > Crime Prevention & Enforcement (1.00)
- Information Technology > Security & Privacy (1.00)
- Government (1.00)
- Law > Criminal Law (0.93)
- Health & Medicine > Therapeutic Area
- Psychiatry/Psychology > Addiction Disorder (0.46)
- Technology: