Improving Privacy Benefits of Redaction

Vaibhav Gusain 1 and Douglas Leith 1 1- Trinity College Dublin - School of Computer science and statistics Dublin - Ireland · 2025

Read Improving Privacy Benefits of Redaction by Vaibhav Gusain 1 and Douglas Leith 1 1- Trinity College Dublin - School of Computer science and statistics Dublin - Ireland (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-17762.txt.

Description

Vaibhav Gusain and Douglas Leith propose a smarter way to scrub sensitive details from text before sharing it. Redaction is the standard defense for sanitizing natural language data. Heavy redaction protects privacy but guts the text's usefulness. Light redaction leaves dangerous clues behind. Their new methodology claims a better trade on both sides at once.

Against the best existing techniques, they report stronger privacy benefits at lower redaction levels. Less text removed means more value survives for analysis and sharing. The paper presents the technique as a practical upgrade for data custodians. Hospitals, firms, and agencies that release text corpora face exactly this tension daily. A method that hides more while deleting less would ease many release decisions. The abstract-level claim is bold; the paper's task is to back it with method and measurement.

Catalog record

CREATORVaibhav Gusain 1 and Douglas Leith 1 1- Trinity College Dublin - School of Computer science and statistics Dublin - Ireland
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE12 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-17762

Browse the full library · About the Tower of Babel