The privacy-utility gap in retrieval-augmented generation: An evaluation of embedding inversion defenses against adaptive attackers
High-Confidence Computing · 2026-11 · 100457
DOI: 10.1016/j.hcc.2026.100457 ↗Kurz gefasst
Fachartikel in der Zeitschrift High-Confidence Computing, 2026-11. Verfasst von Srikar Reddy Gunupati und 2 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Kein Abstract verfügbar.
Zitieren
APA
Gunupati, S. R., Varghese, A. K., & Jiang, J. (2026). The privacy-utility gap in retrieval-augmented generation: An evaluation of embedding inversion defenses against adaptive attackers. High-Confidence Computing, 100457. https://doi.org/10.1016/j.hcc.2026.100457
Vancouver
Gunupati SR, Varghese AK, Jiang J. The privacy-utility gap in retrieval-augmented generation: An evaluation of embedding inversion defenses against adaptive attackers. High-Confidence Computing. 2026:100457. doi:10.1016/j.hcc.2026.100457
BibTeX
@article{Gunupati2026,
author = {Gunupati, Srikar Reddy and Varghese, Alen K. and Jiang, Ji-Han},
title = {The privacy-utility gap in retrieval-augmented generation: An evaluation of embedding inversion defenses against adaptive attackers},
journal = {High-Confidence Computing},
year = {2026},
pages = {100457},
doi = {10.1016/j.hcc.2026.100457}
}
EUVIMED