English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep Neural Network (DNN) models have been shown to have high empiricalprivacy leakages. Clinical language models (CLMs) trained on clinical data havebeen used to improve performance in biomedical natural language processingtasks. In this work, we investigate the risks of training-data leakage throughwhite-box or black-box access to CLMs. We design and employ membershipinference attacks to estimate the empirical privacy leaks for modelarchitectures like BERT and GPT2. We show that membership inference attacks onCLMs lead to non-trivial privacy leakages of up to 7%. Our results show thatsmaller models have lower empirical privacy leakages than larger ones, andmasked LMs have lower leakages than auto-regressive LMs. We further show thatdifferentially private CLMs can have improved model utility on clinical domainwhile ensuring low empirical privacy leakage. Lastly, we also study the effectsof group-level membership inference and disease rarity on CLM privacy leakages.

Membership Inference Attack Susceptibility of Clinical Language Models