Alwin Peng — accepted AI-conference papers

2026

Constitutional Classifiers++: Production-Grade Defenses against Universal Jailbreaks

Hoagy Cunningham, Jerry Wei, Zihan Wang, Andrew Persic, Alwin Peng, Jordan Abderrachid +21

ICLR 2026poster

We introduce enhanced Constitutional Classifiers that deliver production-grade jailbreak robustness with dramatically reduced computational costs and refusal rates compared to previous-generation defenses. We first identify vulnerabilities in existing systems that evaluate model outputs without rega…

Cited by 0 Source ↗Scholar ↗

2026

Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning

Bilgehan Sel, Xuanli He, Alwin Peng, Ming Jin, Jerry Wei

ICML 2026spotlight

Fine-tuning APIs offered by major AI providers create new attack surfaces where adversaries can bypass safety measures through targeted fine-tuning. We introduce **Trojan-Speak**, an adversarial fine-tuning method that bypasses Anthropic's Constitutional Classifiers. Our approach uses curriculum lea…

Cited by 0 Source ↗Scholar ↗