Hybrid DP with LLM or NER preprocessing significantly improves the privacy-utility trade-off for Dutch clinical note de-identification compared to standalone DP.
arXiv preprint arXiv:2209.09631 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Authors introduce MLM and CLM specialization methods that avoid memorizing identifiers in sensitive training data while aiming for a privacy-utility tradeoff on medical datasets.
citing papers explorer
-
Differentially Private De-identification of Dutch Clinical Notes: A Comparative Evaluation
Hybrid DP with LLM or NER preprocessing significantly improves the privacy-utility trade-off for Dutch clinical note de-identification compared to standalone DP.
-
Towards the Anonymization of the Language Modeling
Authors introduce MLM and CLM specialization methods that avoid memorizing identifiers in sensitive training data while aiming for a privacy-utility tradeoff on medical datasets.