PrivFusion: A Privacy-preserving Multi-Agent Framework for Harmonizing Distributed Datasets
Anisa Halimi, Liubov Nedoshivina, Kieran Fraser, Stefano Braghin · 2026
Read PrivFusion: A Privacy-preserving Multi-Agent Framework for Harmonizing Distributed Datasets by Anisa Halimi, Liubov Nedoshivina, Kieran Fraser, Stefano Braghin (2026) 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-2605-24249.txt.
Description
Halimi and three co-authors present PrivFusion, a private multi-agent frame that aligns spread-out clinic data sets before joint training. Central pooling is often blocked for private health data, and joint learning still stumbles on mismatch across site data sets. PrivFusion agents analyze local data, cluster like-meaning features across sites, and iterate change tips until things line up. Tests across four mixed COVID-19 data sets show good, fast matching with far less manual effort.
Catalog record
| CREATOR | Anisa Halimi, Liubov Nedoshivina, Kieran Fraser, Stefano Braghin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29720 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.24249