<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sultan-astudillo-2026-confidence">
    <titleInfo>
        <title>Confidence-Weighted Token Set Cover for Early Hypothesis Pruning in Self-Consistency</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Md</namePart>
        <namePart type="given">Arafat</namePart>
        <namePart type="family">Sultan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ramón</namePart>
        <namePart type="given">Fernandez</namePart>
        <namePart type="family">Astudillo</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>Despite its simplicity and efficacy, the high token expenditure of self-consistency can limit its practical utility. We investigate whether early hypothesis pruning can improve the token efficiency of self-consistency for long chain-of-thought reasoning tasks, while preserving its parallelism. Concretely, we generate all solutions in parallel but periodically prune intermediate hypotheses based on two lightweight indicators: (a) the model’s confidence in each hypothesis, and (b) the lexical coverage of all current hypotheses by candidate subsets. We design a fast weighted set cover algorithm that utilizes the two indicators; evaluation of five LLMs on three math benchmarks shows that our method improves token efficiency in most cases, with reductions of 10-35% in many.</abstract>
    <identifier type="citekey">sultan-astudillo-2026-confidence</identifier>
    <identifier type="doi">10.18653/v1/2026.findings-acl.2046</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-acl.2046/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>41148</start>
            <end>41155</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Confidence-Weighted Token Set Cover for Early Hypothesis Pruning in Self-Consistency
%A Sultan, Md Arafat
%A Astudillo, Ramón Fernandez
%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 sultan-astudillo-2026-confidence
%X Despite its simplicity and efficacy, the high token expenditure of self-consistency can limit its practical utility. We investigate whether early hypothesis pruning can improve the token efficiency of self-consistency for long chain-of-thought reasoning tasks, while preserving its parallelism. Concretely, we generate all solutions in parallel but periodically prune intermediate hypotheses based on two lightweight indicators: (a) the model’s confidence in each hypothesis, and (b) the lexical coverage of all current hypotheses by candidate subsets. We design a fast weighted set cover algorithm that utilizes the two indicators; evaluation of five LLMs on three math benchmarks shows that our method improves token efficiency in most cases, with reductions of 10-35% in many.
%R 10.18653/v1/2026.findings-acl.2046
%U https://aclanthology.org/2026.findings-acl.2046/
%U https://doi.org/10.18653/v1/2026.findings-acl.2046
%P 41148-41155

Markdown (Informal)

[Confidence-Weighted Token Set Cover for Early Hypothesis Pruning in Self-Consistency](https://aclanthology.org/2026.findings-acl.2046/) (Sultan & Astudillo, Findings 2026)

• Confidence-Weighted Token Set Cover for Early Hypothesis Pruning in Self-Consistency (Sultan & Astudillo, Findings 2026)

ACL

• Md Arafat Sultan and Ramón Fernandez Astudillo. 2026. Confidence-Weighted Token Set Cover for Early Hypothesis Pruning in Self-Consistency. In Findings of the Association for Computational Linguistics: ACL 2026, pages 41148–41155, San Diego, California, United States. Association for Computational Linguistics.