@inproceedings{hsia-etal-2026-joint,
title = "Joint Constraint Aggregation for Qualifier-Sensitive Retrieval-Augmented {QA}",
author = "Hsia, Jennifer and
Lipton, Zachary Chase and
Singh, Aarti",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2382/",
pages = "54126--54140",
ISBN = "979-8-89176-459-0",
abstract = "Retrieval-augmented QA can fail even when the correct answer is in the retrieved evidence. This can happen when a competing answer satisfies some, but not all, of the question{'}s requirements and is mentioned frequently in the retrieved documents, making it look like a strong choice. To test whether this retrieval skew actually affects the decision, we manipulate it directly: increasing the prevalence of passages that favor the competing answer makes models more likely to select it. This points to a decision problem that candidate-level support alone cannot resolve. We therefore introduce Joint Constraint Aggregation (JCA), which compares candidates against the same question-derived requirements using the retrieved evidence. Across our evaluation settings, JCA improves overall accuracy, and its advantage grows when retrieval favors a competing answer. Together, these results suggest that when partial matches and retrieval prominence pull models toward a competing answer, comparing candidates against the question{'}s full requirements can make selection more reliable."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hsia-etal-2026-joint">
<titleInfo>
<title>Joint Constraint Aggregation for Qualifier-Sensitive Retrieval-Augmented QA</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jennifer</namePart>
<namePart type="family">Hsia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zachary</namePart>
<namePart type="given">Chase</namePart>
<namePart type="family">Lipton</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aarti</namePart>
<namePart type="family">Singh</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Retrieval-augmented QA can fail even when the correct answer is in the retrieved evidence. This can happen when a competing answer satisfies some, but not all, of the question’s requirements and is mentioned frequently in the retrieved documents, making it look like a strong choice. To test whether this retrieval skew actually affects the decision, we manipulate it directly: increasing the prevalence of passages that favor the competing answer makes models more likely to select it. This points to a decision problem that candidate-level support alone cannot resolve. We therefore introduce Joint Constraint Aggregation (JCA), which compares candidates against the same question-derived requirements using the retrieved evidence. Across our evaluation settings, JCA improves overall accuracy, and its advantage grows when retrieval favors a competing answer. Together, these results suggest that when partial matches and retrieval prominence pull models toward a competing answer, comparing candidates against the question’s full requirements can make selection more reliable.</abstract>
<identifier type="citekey">hsia-etal-2026-joint</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2382/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>54126</start>
<end>54140</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Joint Constraint Aggregation for Qualifier-Sensitive Retrieval-Augmented QA
%A Hsia, Jennifer
%A Lipton, Zachary Chase
%A Singh, Aarti
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F hsia-etal-2026-joint
%X Retrieval-augmented QA can fail even when the correct answer is in the retrieved evidence. This can happen when a competing answer satisfies some, but not all, of the question’s requirements and is mentioned frequently in the retrieved documents, making it look like a strong choice. To test whether this retrieval skew actually affects the decision, we manipulate it directly: increasing the prevalence of passages that favor the competing answer makes models more likely to select it. This points to a decision problem that candidate-level support alone cannot resolve. We therefore introduce Joint Constraint Aggregation (JCA), which compares candidates against the same question-derived requirements using the retrieved evidence. Across our evaluation settings, JCA improves overall accuracy, and its advantage grows when retrieval favors a competing answer. Together, these results suggest that when partial matches and retrieval prominence pull models toward a competing answer, comparing candidates against the question’s full requirements can make selection more reliable.
%U https://aclanthology.org/2026.emnlp-main.2382/
%P 54126-54140
Markdown (Informal)
[Joint Constraint Aggregation for Qualifier-Sensitive Retrieval-Augmented QA](https://aclanthology.org/2026.emnlp-main.2382/) (Hsia et al., EMNLP 2026)
ACL