GLiNER PII

NVIDIA GLiNER PII is a named entity recognition model specialized in detecting Personally Identifiable Information (PII) in text. Built on the GLiNER (Generalist and Lightweight Model for NER) architecture, it identifies entities like names, addresses, phone numbers, email addresses, social security numbers, and other sensitive personal data.

The model uses a zero-shot approach, meaning it can detect PII categories it wasn't explicitly trained on, adapting to new types of personal information through its generalized understanding of entity patterns. This makes it robust against evolving data formats and new types of PII.

GLiNER PII is essential for data privacy compliance, automated redaction, data anonymization pipelines, and content moderation systems that need to protect personal information.