@inproceedings{mandiga-etal-2026-paraphrase,
title = "Paraphrase-Induced Retrieval Instability: Systematic Measurement and the Limits of Inference-Time Mitigation",
author = "Mandiga, Aravind and
Li, Guoming and
Lu, Jin",
editor = {Aziz, Wilker and
Berant, Jonathan and
Eikema, Bryan and
Ilia, Evgenia and
de Marneffe, Marie-Catherine and
Plank, Barbara and
Shelmanov, Artem and
Swayamdipta, Swabha and
Tiedemann, J{\"o}rg and
Vazhentsev, Artem and
V{\'a}zquez, Ra{\'u}l and
Zerva, Chrysoula},
booktitle = "The Proceedings of the Third Workshop on Uncertainty-Aware {NLP}",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.uncertainlp-main.34/",
pages = "555--569",
ISBN = "979-8-89176-486-6",
abstract = "Dense retrieval can return different evidence when a query is paraphrased, even when the meaning stays the same. This means that linguistic variation can create uncertainty in the evidence given to a retrieval-augmented generation (RAG) system. We study this problem across HotpotQA, FEVER, and SQuAD 2.0 using five dense encoders spanning 22M to 8B parameters and multiple model families. Paraphrase Jaccard distance remains 32 to 50 percent, showing that retrieval instability persists with stronger and alternative encoders. We then ask how much this problem can be reduced at inference time. Multi-source consensus over paraphrased queries and different retrievers improves retrieval coverage. Fixed source-type weights can help in harder settings. However, adapting those weights to each query using retrieval stability changes Coverage@20 by at most {\ensuremath{\pm}}0.3 percentage points per setting. Retrieval stability is more useful as a label-free routing signal: on HotpotQA/MiniLM, routing only unstable queries through the stability-conditioned consensus configuration reduces retrieval calls by 47 percent for a 0.6 percentage-point Coverage@20 cost compared with always using that configuration. Overall, paraphrase-induced retrieval instability is a persistent source of uncertainty in retrieved evidence, and the query-adaptive stability component provides only a small marginal refinement."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mandiga-etal-2026-paraphrase">
<titleInfo>
<title>Paraphrase-Induced Retrieval Instability: Systematic Measurement and the Limits of Inference-Time Mitigation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aravind</namePart>
<namePart type="family">Mandiga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guoming</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jin</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>The Proceedings of the Third Workshop on Uncertainty-Aware NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wilker</namePart>
<namePart type="family">Aziz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathan</namePart>
<namePart type="family">Berant</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bryan</namePart>
<namePart type="family">Eikema</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Evgenia</namePart>
<namePart type="family">Ilia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Plank</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Shelmanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Swabha</namePart>
<namePart type="family">Swayamdipta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Tiedemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Vazhentsev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raúl</namePart>
<namePart type="family">Vázquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chrysoula</namePart>
<namePart type="family">Zerva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-486-6</identifier>
</relatedItem>
<abstract>Dense retrieval can return different evidence when a query is paraphrased, even when the meaning stays the same. This means that linguistic variation can create uncertainty in the evidence given to a retrieval-augmented generation (RAG) system. We study this problem across HotpotQA, FEVER, and SQuAD 2.0 using five dense encoders spanning 22M to 8B parameters and multiple model families. Paraphrase Jaccard distance remains 32 to 50 percent, showing that retrieval instability persists with stronger and alternative encoders. We then ask how much this problem can be reduced at inference time. Multi-source consensus over paraphrased queries and different retrievers improves retrieval coverage. Fixed source-type weights can help in harder settings. However, adapting those weights to each query using retrieval stability changes Coverage@20 by at most \ensuremath\pm0.3 percentage points per setting. Retrieval stability is more useful as a label-free routing signal: on HotpotQA/MiniLM, routing only unstable queries through the stability-conditioned consensus configuration reduces retrieval calls by 47 percent for a 0.6 percentage-point Coverage@20 cost compared with always using that configuration. Overall, paraphrase-induced retrieval instability is a persistent source of uncertainty in retrieved evidence, and the query-adaptive stability component provides only a small marginal refinement.</abstract>
<identifier type="citekey">mandiga-etal-2026-paraphrase</identifier>
<location>
<url>https://aclanthology.org/2026.uncertainlp-main.34/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>555</start>
<end>569</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Paraphrase-Induced Retrieval Instability: Systematic Measurement and the Limits of Inference-Time Mitigation
%A Mandiga, Aravind
%A Li, Guoming
%A Lu, Jin
%Y Aziz, Wilker
%Y Berant, Jonathan
%Y Eikema, Bryan
%Y Ilia, Evgenia
%Y de Marneffe, Marie-Catherine
%Y Plank, Barbara
%Y Shelmanov, Artem
%Y Swayamdipta, Swabha
%Y Tiedemann, Jörg
%Y Vazhentsev, Artem
%Y Vázquez, Raúl
%Y Zerva, Chrysoula
%S The Proceedings of the Third Workshop on Uncertainty-Aware NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-486-6
%F mandiga-etal-2026-paraphrase
%X Dense retrieval can return different evidence when a query is paraphrased, even when the meaning stays the same. This means that linguistic variation can create uncertainty in the evidence given to a retrieval-augmented generation (RAG) system. We study this problem across HotpotQA, FEVER, and SQuAD 2.0 using five dense encoders spanning 22M to 8B parameters and multiple model families. Paraphrase Jaccard distance remains 32 to 50 percent, showing that retrieval instability persists with stronger and alternative encoders. We then ask how much this problem can be reduced at inference time. Multi-source consensus over paraphrased queries and different retrievers improves retrieval coverage. Fixed source-type weights can help in harder settings. However, adapting those weights to each query using retrieval stability changes Coverage@20 by at most \ensuremath\pm0.3 percentage points per setting. Retrieval stability is more useful as a label-free routing signal: on HotpotQA/MiniLM, routing only unstable queries through the stability-conditioned consensus configuration reduces retrieval calls by 47 percent for a 0.6 percentage-point Coverage@20 cost compared with always using that configuration. Overall, paraphrase-induced retrieval instability is a persistent source of uncertainty in retrieved evidence, and the query-adaptive stability component provides only a small marginal refinement.
%U https://aclanthology.org/2026.uncertainlp-main.34/
%P 555-569
Markdown (Informal)
[Paraphrase-Induced Retrieval Instability: Systematic Measurement and the Limits of Inference-Time Mitigation](https://aclanthology.org/2026.uncertainlp-main.34/) (Mandiga et al., UncertaiNLP 2026)
ACL