@inproceedings{bouziane-etal-2026-candidate,
title = "Candidate-Aware Retrieval and Reranking for Multiple-Choice Question Answering: {A}rabic as a Case Study",
author = "Bouziane, Yassine and
Moukafih, Youness and
Ghogho, Mounir",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.435/",
pages = "8967--8977",
ISBN = "979-8-89176-395-1",
abstract = "Large language models (LLMs) have recently achieved impressive results on multiple-choice question answering (MCQA), with retrieval-augmented generation (RAG) emerging as an effective strategy for improving the performance of smaller models. However, existing RAG formulations face persistent challenges: retrieving too many passages often introduces noise, and even when relevant content is retrieved, models may still struggle with partially relevant or conflicting information. Moreover, while LLMs perform strongly on English benchmarks, their accuracy declines substantially on Arabic multi-task evaluations, revealing ongoing issues in cross-lingual transfer and domain adaptation. In this paper, we propose a novel approach, using Arabic as a representative case study, that jointly models the relevance of both the question and its candidate answers when selecting contextual passages. The method employs a lightweight reranker trained with a hybrid regression{--}triplet loss objective to identify passages that provide discriminative and reliable evidence. Extensive experiments across multiple model sizes and humanities domains show that our approach consistently outperforms both standard RAG baselines and reranker baselines, delivering two- to threefold improvements while remaining competitive with considerably larger models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bouziane-etal-2026-candidate">
<titleInfo>
<title>Candidate-Aware Retrieval and Reranking for Multiple-Choice Question Answering: Arabic as a Case Study</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yassine</namePart>
<namePart type="family">Bouziane</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Youness</namePart>
<namePart type="family">Moukafih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mounir</namePart>
<namePart type="family">Ghogho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Large language models (LLMs) have recently achieved impressive results on multiple-choice question answering (MCQA), with retrieval-augmented generation (RAG) emerging as an effective strategy for improving the performance of smaller models. However, existing RAG formulations face persistent challenges: retrieving too many passages often introduces noise, and even when relevant content is retrieved, models may still struggle with partially relevant or conflicting information. Moreover, while LLMs perform strongly on English benchmarks, their accuracy declines substantially on Arabic multi-task evaluations, revealing ongoing issues in cross-lingual transfer and domain adaptation. In this paper, we propose a novel approach, using Arabic as a representative case study, that jointly models the relevance of both the question and its candidate answers when selecting contextual passages. The method employs a lightweight reranker trained with a hybrid regression–triplet loss objective to identify passages that provide discriminative and reliable evidence. Extensive experiments across multiple model sizes and humanities domains show that our approach consistently outperforms both standard RAG baselines and reranker baselines, delivering two- to threefold improvements while remaining competitive with considerably larger models.</abstract>
<identifier type="citekey">bouziane-etal-2026-candidate</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.435/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>8967</start>
<end>8977</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Candidate-Aware Retrieval and Reranking for Multiple-Choice Question Answering: Arabic as a Case Study
%A Bouziane, Yassine
%A Moukafih, Youness
%A Ghogho, Mounir
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F bouziane-etal-2026-candidate
%X Large language models (LLMs) have recently achieved impressive results on multiple-choice question answering (MCQA), with retrieval-augmented generation (RAG) emerging as an effective strategy for improving the performance of smaller models. However, existing RAG formulations face persistent challenges: retrieving too many passages often introduces noise, and even when relevant content is retrieved, models may still struggle with partially relevant or conflicting information. Moreover, while LLMs perform strongly on English benchmarks, their accuracy declines substantially on Arabic multi-task evaluations, revealing ongoing issues in cross-lingual transfer and domain adaptation. In this paper, we propose a novel approach, using Arabic as a representative case study, that jointly models the relevance of both the question and its candidate answers when selecting contextual passages. The method employs a lightweight reranker trained with a hybrid regression–triplet loss objective to identify passages that provide discriminative and reliable evidence. Extensive experiments across multiple model sizes and humanities domains show that our approach consistently outperforms both standard RAG baselines and reranker baselines, delivering two- to threefold improvements while remaining competitive with considerably larger models.
%U https://aclanthology.org/2026.findings-acl.435/
%P 8967-8977
Markdown (Informal)
[Candidate-Aware Retrieval and Reranking for Multiple-Choice Question Answering: Arabic as a Case Study](https://aclanthology.org/2026.findings-acl.435/) (Bouziane et al., Findings 2026)
ACL