<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rahmani-etal-2026-self">
    <titleInfo>
        <title>Self-Correcting Large Language Models: Generation vs. Multiple Choice</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hossein</namePart>
        <namePart type="given">A</namePart>
        <namePart type="family">Rahmani</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Satyapriya</namePart>
        <namePart type="family">Krishna</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xi</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mohammadmehdi</namePart>
        <namePart type="family">Naghiaei</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Emine</namePart>
        <namePart type="family">Yilmaz</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 Second Workshop for REsearch on Agent Language Models (REALM 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Ehsan</namePart>
            <namePart type="family">Kamalloo</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nicolas</namePart>
            <namePart type="family">Gontier</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nikolai</namePart>
            <namePart type="family">Rozanov</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Maryam</namePart>
            <namePart type="family">Hashemzadeh</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Akhil</namePart>
            <namePart type="family">Arora</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">979-8-89176-485-9</identifier>
    </relatedItem>
    <abstract>Large language models have recently demonstrated remarkable abilities to self-correct their responses through iterative refinement, often referred to as self-consistency or self-reflection. However, the dynamics of this self-correction mechanism may differ substantially depending on whether the model is tasked with open-ended text generation or with selecting the most appropriate response from multiple predefined options. In this paper, we conduct a systematic investigation of these two paradigms by comparing performance trends and error-correction behaviors across question answering tasks, covering language models of different scales and families. Our experimental results reveal distinct patterns of improvement and failure modes:\par While open-ended generation often benefits from the flexibility of re-interpretation and compositional refinement, multiple-choice selection can leverage clearer solution boundaries but may be limited by the provided options. This contrast also reflects the dual demands faced by emerging agentic LLM applications: effective agents must not only generate and refine open-ended plans or explanations, but also make reliable discrete choices when operating within constrained action spaces. Our findings, therefore, highlight that the design of self-correction mechanisms should take into account the interaction between task structure and output space, with implications for both knowledge-intensive reasoning and decision-oriented applications of LLMs.</abstract>
    <identifier type="citekey">rahmani-etal-2026-self</identifier>
    <location>
        <url>https://aclanthology.org/2026.realm-1.4/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>38</start>
            <end>56</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Self-Correcting Large Language Models: Generation vs. Multiple Choice
%A Rahmani, Hossein A.
%A Krishna, Satyapriya
%A Wang, Xi
%A Naghiaei, Mohammadmehdi
%A Yilmaz, Emine
%Y Kamalloo, Ehsan
%Y Gontier, Nicolas
%Y Rozanov, Nikolai
%Y Hashemzadeh, Maryam
%Y Arora, Akhil
%S Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-485-9
%F rahmani-etal-2026-self
%X Large language models have recently demonstrated remarkable abilities to self-correct their responses through iterative refinement, often referred to as self-consistency or self-reflection. However, the dynamics of this self-correction mechanism may differ substantially depending on whether the model is tasked with open-ended text generation or with selecting the most appropriate response from multiple predefined options. In this paper, we conduct a systematic investigation of these two paradigms by comparing performance trends and error-correction behaviors across question answering tasks, covering language models of different scales and families. Our experimental results reveal distinct patterns of improvement and failure modes:\par While open-ended generation often benefits from the flexibility of re-interpretation and compositional refinement, multiple-choice selection can leverage clearer solution boundaries but may be limited by the provided options. This contrast also reflects the dual demands faced by emerging agentic LLM applications: effective agents must not only generate and refine open-ended plans or explanations, but also make reliable discrete choices when operating within constrained action spaces. Our findings, therefore, highlight that the design of self-correction mechanisms should take into account the interaction between task structure and output space, with implications for both knowledge-intensive reasoning and decision-oriented applications of LLMs.
%U https://aclanthology.org/2026.realm-1.4/
%P 38-56

Markdown (Informal)

[Self-Correcting Large Language Models: Generation vs. Multiple Choice](https://aclanthology.org/2026.realm-1.4/) (Rahmani et al., REALM 2026)

• Self-Correcting Large Language Models: Generation vs. Multiple Choice (Rahmani et al., REALM 2026)

ACL

• Hossein A. Rahmani, Satyapriya Krishna, Xi Wang, Mohammadmehdi Naghiaei, and Emine Yilmaz. 2026. Self-Correcting Large Language Models: Generation vs. Multiple Choice. In Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026), pages 38–56, Budapest, Hungary. Association for Computational Linguistics.