Presented in the USENIX Security 2026 poster session on 12 August 2026.
AdvPIIBench studies personally identifiable information (PII) detection under adversarial conditions, with resources for evaluating detectors in the context of LLM interactions.
Publication · 2026-08-12
AdvPIIBench - The Adversarial Collapse of PII Detection in LLM Interactions
Roei Arpaly and Yoni Birman investigate how adversarial prompts challenge the detection of personally identifiable information in LLM interactions.
Publication