Overview
I am a research fellow at the NSF ACTION Institute (AI Institute for Agent-based Cyber Threat Intelligence and Operation), hosted at UC Santa Barbara for the 2026 summer research program. Working with Dr. Yigitcan Kaya in UCSB's SecLab, I study token entanglement and subliminal learning in large language models, specifically whether hidden system-prompt secrets leak into unrelated outputs through likelihood-based inference.
Key Achievements
- Researching token entanglement in subliminal learning within LLMs under Dr. Yigitcan Kaya at UC Santa Barbara.
- Investigating whether hidden system-prompt secrets leak into unrelated outputs via likelihood-based inference.
- Designed a multi-stage pipeline of logit probes, ablations, cross-model attacks, and prompt ladders across 14+ conditions to characterize when LLM instruction secrets become recoverable.