<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="long-etal-2026-reflection">
    <titleInfo>
        <title>Reflection Is Not Monolithic: Experience-Driven Control of Recheck in LLM Reasoning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Quanyu</namePart>
        <namePart type="family">Long</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiang</namePart>
        <namePart type="given">Kai</namePart>
        <namePart type="family">Jie</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jianda</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Qisheng</namePart>
        <namePart type="family">Hu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xu</namePart>
        <namePart type="family">Guo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Leilei</namePart>
        <namePart type="family">Gan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Wenya</namePart>
        <namePart type="family">Wang</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 Reasoning Models (LRMs) achieve strong performance by generating long reasoning traces with explicit reflection. However, reflection is not functionally uniform: some reflections revise the reasoning strategy, while others verify previously derived intermediate results. In this work, we examine whether local self-verification can be controlled selectively without suppressing useful reasoning. Through a large-scale analysis of mathematical reasoning traces, we distinguish strategy-level rethink from local self-verification recheck, and further separate rechecks into corrective and confirmatory cases. We find that recheck occurs frequently across models and benchmarks, but most observed rechecks are confirmatory, rarely uncovering errors or changing the reasoning trajectory. Motivated by this finding, we propose Experience-Driven Suppression (EDS), a conservative test-time controller that suppresses a recheck only when historical outcomes of similar episodes suggest low utility. Unlike early-exit or trigger-suppression methods that stop reasoning globally or suppress reflection cues broadly, EDS targets local recheck while preserving later rethink. Experiments show that EDS yields moderate reasoning-length reductions with limited accuracy changes, supporting behavior-specific rather than monolithic reflection control.</abstract>
    <identifier type="citekey">long-etal-2026-reflection</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1130/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>23920</start>
            <end>23940</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Reflection Is Not Monolithic: Experience-Driven Control of Recheck in LLM Reasoning
%A Long, Quanyu
%A Jie, Jiang Kai
%A Chen, Jianda
%A Hu, Qisheng
%A Guo, Xu
%A Gan, Leilei
%A Wang, Wenya
%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 long-etal-2026-reflection
%X Large Reasoning Models (LRMs) achieve strong performance by generating long reasoning traces with explicit reflection. However, reflection is not functionally uniform: some reflections revise the reasoning strategy, while others verify previously derived intermediate results. In this work, we examine whether local self-verification can be controlled selectively without suppressing useful reasoning. Through a large-scale analysis of mathematical reasoning traces, we distinguish strategy-level rethink from local self-verification recheck, and further separate rechecks into corrective and confirmatory cases. We find that recheck occurs frequently across models and benchmarks, but most observed rechecks are confirmatory, rarely uncovering errors or changing the reasoning trajectory. Motivated by this finding, we propose Experience-Driven Suppression (EDS), a conservative test-time controller that suppresses a recheck only when historical outcomes of similar episodes suggest low utility. Unlike early-exit or trigger-suppression methods that stop reasoning globally or suppress reflection cues broadly, EDS targets local recheck while preserving later rethink. Experiments show that EDS yields moderate reasoning-length reductions with limited accuracy changes, supporting behavior-specific rather than monolithic reflection control.
%U https://aclanthology.org/2026.findings-emnlp.1130/
%P 23920-23940

Markdown (Informal)

[Reflection Is Not Monolithic: Experience-Driven Control of Recheck in LLM Reasoning](https://aclanthology.org/2026.findings-emnlp.1130/) (Long et al., Findings 2026)

• Reflection Is Not Monolithic: Experience-Driven Control of Recheck in LLM Reasoning (Long et al., Findings 2026)

ACL

• Quanyu Long, Jiang Kai Jie, Jianda Chen, Qisheng Hu, Xu Guo, Leilei Gan, and Wenya Wang. 2026. Reflection Is Not Monolithic: Experience-Driven Control of Recheck in LLM Reasoning. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 23920–23940, Budapest, Hungary. Association for Computational Linguistics.