... for establishing our ground truth. De-identification Anonymization GPT Llama Large language models Privacy Massively open online course The two main approaches that have been explored for identifying and redacting PII in text are pattern matching and supervised machine learning techniques, each...

... intellectually vulnerable. It is also likely that anonymization, which is often considered the gold standard for protecting privacy, may not sufficiently protect students’ intellectual privacy. Even if students understand that their writing will be anonymized, they may still experience some fear of scrutiny...