AI-secure/aug-pe

[ICML 2024 Spotlight] Differentially Private Synthetic Data via Foundation Model APIs 2: Text

43
/ 100
Emerging

This project helps data analysts and researchers create artificial text data that mimics real data, such as Yelp reviews or scientific abstracts, while protecting the original data's privacy. You provide your sensitive text data, and it generates new, synthetic text that has similar characteristics but can't be traced back to individuals. This is perfect for sharing datasets or developing models without compromising confidentiality.

No commits in the last 6 months.

Use this if you need to generate privacy-preserving synthetic text data from sensitive sources like customer reviews, medical abstracts, or research papers for analysis or model training.

Not ideal if you need to generate synthetic data for images or other non-text data types, or if you require perfect, verbatim copies of your original text.

data-privacy text-generation data-sharing natural-language-processing research-data
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

57

Forks

17

Language

Python

License

Apache-2.0

Last pushed

Jan 11, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/AI-secure/aug-pe"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.