@inproceedings{hemmat-etal-2025-vague,
title = "{VAGUE}{-}{G}ate: {P}lug{-}and{-}{P}lay {L}ocal{-}{P}rivacy Shield for {R}etrieval{-}{A}ugmented Generation",
author = "Hemmat, Arshia and
Moqadas, Matin and
Mamanpoosh, Ali and
Rismanchian, Amirmasoud and
Fatemi, Afsaneh",
editor = "Inui, Kentaro and
Sakti, Sakriani and
Wang, Haofen and
Wong, Derek F. and
Bhattacharyya, Pushpak and
Banerjee, Biplab and
Ekbal, Asif and
Chakraborty, Tanmoy and
Singh, Dhirendra Pratap",
booktitle = "Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics",
month = dec,
year = "2025",
address = "Mumbai, India",
publisher = "The Asian Federation of Natural Language Processing and The Association for Computational Linguistics",
url = "https://aclanthology.org/2025.ijcnlp-long.194/",
pages = "3715--3730",
ISBN = "979-8-89176-298-5",
abstract = "Retrieval-augmented generation (RAG) still *forwards* raw passages to large-language models, so private facts slip through. Prior defenses are either (i) **heavyweight**{---}full DP training that is impractical for today{'}s 70B-parameter models{---}or (ii) **over-zealous**{---}blanket redaction of every named entity, which slashes answer quality.We introduce **VAGUE-Gate**, a lightweight, *locally* differentially-private gate deployable in front of *any* RAG system. A precision pass drops low-utility tokens under a user budget {\ensuremath{\varepsilon}}, then up to k({\ensuremath{\varepsilon}}) high-temperature paraphrase passes further cloud residual cues; post-processing guarantees preserve the same {\ensuremath{\varepsilon}}-LDP bound.To measure both privacy and utility, we release **BlendPriv** (3k blended-sensitivity QA pairs) and two new metrics: a lexical Information-Leakage Score and an LLM-as-Judge score. Across eight pipelines and four SOTA LLMs, **VAGUE-Gate** at {\ensuremath{\varepsilon}} = 0.3 lowers lexical leakage by **70{\%}** and semantic leakage by **1.8** points (1{--}5 scale) while retaining **91{\%}** of Plain-RAG faithfulness with only a **240 ms** latency overhead.All code, data, and prompts are publicly released:- Code: {\ensuremath{<}} https://github.com/arshiahemmat/LDP{\_}RAG {\ensuremath{>}} - Dataset: {\ensuremath{<}}https://huggingface.co/datasets/AliMnp/BlendPriv{\ensuremath{>}}"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hemmat-etal-2025-vague">
<titleInfo>
<title>VAGUE-Gate: Plug-and-Play Local-Privacy Shield for Retrieval-Augmented Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Arshia</namePart>
<namePart type="family">Hemmat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matin</namePart>
<namePart type="family">Moqadas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ali</namePart>
<namePart type="family">Mamanpoosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amirmasoud</namePart>
<namePart type="family">Rismanchian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Afsaneh</namePart>
<namePart type="family">Fatemi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haofen</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Derek</namePart>
<namePart type="given">F</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biplab</namePart>
<namePart type="family">Banerjee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asif</namePart>
<namePart type="family">Ekbal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhirendra</namePart>
<namePart type="given">Pratap</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>The Asian Federation of Natural Language Processing and The Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mumbai, India</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-298-5</identifier>
</relatedItem>
<abstract>Retrieval-augmented generation (RAG) still *forwards* raw passages to large-language models, so private facts slip through. Prior defenses are either (i) **heavyweight**—full DP training that is impractical for today’s 70B-parameter models—or (ii) **over-zealous**—blanket redaction of every named entity, which slashes answer quality.We introduce **VAGUE-Gate**, a lightweight, *locally* differentially-private gate deployable in front of *any* RAG system. A precision pass drops low-utility tokens under a user budget \ensuremathǎrepsilon, then up to k(\ensuremathǎrepsilon) high-temperature paraphrase passes further cloud residual cues; post-processing guarantees preserve the same \ensuremathǎrepsilon-LDP bound.To measure both privacy and utility, we release **BlendPriv** (3k blended-sensitivity QA pairs) and two new metrics: a lexical Information-Leakage Score and an LLM-as-Judge score. Across eight pipelines and four SOTA LLMs, **VAGUE-Gate** at \ensuremathǎrepsilon = 0.3 lowers lexical leakage by **70%** and semantic leakage by **1.8** points (1–5 scale) while retaining **91%** of Plain-RAG faithfulness with only a **240 ms** latency overhead.All code, data, and prompts are publicly released:- Code: \ensuremath< https://github.com/arshiahemmat/LDP_RAG \ensuremath> - Dataset: \ensuremath<https://huggingface.co/datasets/AliMnp/BlendPriv\ensuremath></abstract>
<identifier type="citekey">hemmat-etal-2025-vague</identifier>
<location>
<url>https://aclanthology.org/2025.ijcnlp-long.194/</url>
</location>
<part>
<date>2025-12</date>
<extent unit="page">
<start>3715</start>
<end>3730</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T VAGUE-Gate: Plug-and-Play Local-Privacy Shield for Retrieval-Augmented Generation
%A Hemmat, Arshia
%A Moqadas, Matin
%A Mamanpoosh, Ali
%A Rismanchian, Amirmasoud
%A Fatemi, Afsaneh
%Y Inui, Kentaro
%Y Sakti, Sakriani
%Y Wang, Haofen
%Y Wong, Derek F.
%Y Bhattacharyya, Pushpak
%Y Banerjee, Biplab
%Y Ekbal, Asif
%Y Chakraborty, Tanmoy
%Y Singh, Dhirendra Pratap
%S Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics
%D 2025
%8 December
%I The Asian Federation of Natural Language Processing and The Association for Computational Linguistics
%C Mumbai, India
%@ 979-8-89176-298-5
%F hemmat-etal-2025-vague
%X Retrieval-augmented generation (RAG) still *forwards* raw passages to large-language models, so private facts slip through. Prior defenses are either (i) **heavyweight**—full DP training that is impractical for today’s 70B-parameter models—or (ii) **over-zealous**—blanket redaction of every named entity, which slashes answer quality.We introduce **VAGUE-Gate**, a lightweight, *locally* differentially-private gate deployable in front of *any* RAG system. A precision pass drops low-utility tokens under a user budget \ensuremathǎrepsilon, then up to k(\ensuremathǎrepsilon) high-temperature paraphrase passes further cloud residual cues; post-processing guarantees preserve the same \ensuremathǎrepsilon-LDP bound.To measure both privacy and utility, we release **BlendPriv** (3k blended-sensitivity QA pairs) and two new metrics: a lexical Information-Leakage Score and an LLM-as-Judge score. Across eight pipelines and four SOTA LLMs, **VAGUE-Gate** at \ensuremathǎrepsilon = 0.3 lowers lexical leakage by **70%** and semantic leakage by **1.8** points (1–5 scale) while retaining **91%** of Plain-RAG faithfulness with only a **240 ms** latency overhead.All code, data, and prompts are publicly released:- Code: \ensuremath< https://github.com/arshiahemmat/LDP_RAG \ensuremath> - Dataset: \ensuremath<https://huggingface.co/datasets/AliMnp/BlendPriv\ensuremath>
%U https://aclanthology.org/2025.ijcnlp-long.194/
%P 3715-3730
Markdown (Informal)
[VAGUE‐Gate: Plug‐and‐Play Local‐Privacy Shield for Retrieval‐Augmented Generation](https://aclanthology.org/2025.ijcnlp-long.194/) (Hemmat et al., IJCNLP-AACL 2025)
ACL
- Arshia Hemmat, Matin Moqadas, Ali Mamanpoosh, Amirmasoud Rismanchian, and Afsaneh Fatemi. 2025. VAGUE‐Gate: Plug‐and‐Play Local‐Privacy Shield for Retrieval‐Augmented Generation. In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 3715–3730, Mumbai, India. The Asian Federation of Natural Language Processing and The Association for Computational Linguistics.