@inproceedings{fatemi-etal-2026-guided,
title = "Guided Permutation Scoring ({GPS}) : Option-Order Permutation for Multiple-Choice Question Answering",
author = "Fatemi, Md Shahrar and
Shi, Zhan and
Mueller, Klaus and
Ma, Tengfei",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1290/",
pages = "27294--27313",
ISBN = "979-8-89176-462-0",
abstract = "Large Language Models (LLMs) are sensitive to the order in which answer choices are presented in Multiple-Choice Question Answering. The same question can yield a different prediction when its options are shuffled. Although exhaustive majority voting over all permutations can mitigate this bias, its factorial cost makes it impractical at inference time. We propose GPS (Guided Permutation Scoring), a two-stage inference-time pipeline that selects option orderings more likely to produce correct predictions from a frozen LLM. A sensitivity classifier first identifies which questions are genuinely vulnerable to positional interference; robust questions are answered directly, while sensitive ones are forwarded to a trained permutation scorer that identifies a more position-stable option ordering, promoting better predictions. Unlike prior approaches that either enumerate all permutations uniformly, or require modifying the LLM{'}s parameters, GPS is the first method to actively select permutations so that the base model can consistently give better answers. GPS steadily improves Accuracy, and Macro F1 across all evaluated model-dataset combinations, with remarkable reductions in positional bias metrics."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fatemi-etal-2026-guided">
<titleInfo>
<title>Guided Permutation Scoring (GPS) : Option-Order Permutation for Multiple-Choice Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="given">Shahrar</namePart>
<namePart type="family">Fatemi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhan</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Klaus</namePart>
<namePart type="family">Mueller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tengfei</namePart>
<namePart type="family">Ma</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) are sensitive to the order in which answer choices are presented in Multiple-Choice Question Answering. The same question can yield a different prediction when its options are shuffled. Although exhaustive majority voting over all permutations can mitigate this bias, its factorial cost makes it impractical at inference time. We propose GPS (Guided Permutation Scoring), a two-stage inference-time pipeline that selects option orderings more likely to produce correct predictions from a frozen LLM. A sensitivity classifier first identifies which questions are genuinely vulnerable to positional interference; robust questions are answered directly, while sensitive ones are forwarded to a trained permutation scorer that identifies a more position-stable option ordering, promoting better predictions. Unlike prior approaches that either enumerate all permutations uniformly, or require modifying the LLM’s parameters, GPS is the first method to actively select permutations so that the base model can consistently give better answers. GPS steadily improves Accuracy, and Macro F1 across all evaluated model-dataset combinations, with remarkable reductions in positional bias metrics.</abstract>
<identifier type="citekey">fatemi-etal-2026-guided</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1290/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>27294</start>
<end>27313</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Guided Permutation Scoring (GPS) : Option-Order Permutation for Multiple-Choice Question Answering
%A Fatemi, Md Shahrar
%A Shi, Zhan
%A Mueller, Klaus
%A Ma, Tengfei
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F fatemi-etal-2026-guided
%X Large Language Models (LLMs) are sensitive to the order in which answer choices are presented in Multiple-Choice Question Answering. The same question can yield a different prediction when its options are shuffled. Although exhaustive majority voting over all permutations can mitigate this bias, its factorial cost makes it impractical at inference time. We propose GPS (Guided Permutation Scoring), a two-stage inference-time pipeline that selects option orderings more likely to produce correct predictions from a frozen LLM. A sensitivity classifier first identifies which questions are genuinely vulnerable to positional interference; robust questions are answered directly, while sensitive ones are forwarded to a trained permutation scorer that identifies a more position-stable option ordering, promoting better predictions. Unlike prior approaches that either enumerate all permutations uniformly, or require modifying the LLM’s parameters, GPS is the first method to actively select permutations so that the base model can consistently give better answers. GPS steadily improves Accuracy, and Macro F1 across all evaluated model-dataset combinations, with remarkable reductions in positional bias metrics.
%U https://aclanthology.org/2026.findings-emnlp.1290/
%P 27294-27313
Markdown (Informal)
[Guided Permutation Scoring (GPS) : Option-Order Permutation for Multiple-Choice Question Answering](https://aclanthology.org/2026.findings-emnlp.1290/) (Fatemi et al., Findings 2026)
ACL