<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2025-demystify">
    <titleInfo>
        <title>Demystify Verbosity Compensation Behavior of Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yusen</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sarkar</namePart>
        <namePart type="family">Snigdha Sarathi Das</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Rui</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2nd Workshop on Uncertainty-Aware NLP (UncertaiNLP 2025)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Noidea</namePart>
            <namePart type="family">Noidea</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-349-4</identifier>
    </relatedItem>
    <abstract>Recent work has revealed Large Language Models (LLMs) often exhibit undesirable behaviors, such as hallucination and toxicity, limiting their reliability and broader adoption. In this paper, we discover an understudied type of undesirable behavior of LLMs, which we term Verbosity Compensation (VC). VC is similar to the hesitation behavior of humans under uncertainty, compensating with excessive words such as repeating questions, introducing ambiguity, or providing excessive enumeration. We present the first work that analyzes Verbosity Compensation, explores its causes, and proposes a simple mitigating approach. Our experiments on five datasets of knowledge and reasoning-based QA tasks with 14 LLMs, reveal three conclusions. 1) A pervasive presence of VC across all models and all datasets. 2) The large performance gap between verbose and concise responses. We also demonstrate that this difference does not naturally diminish as LLM capability increases. 3) Higher uncertainty exhibited by VC responses across all five datasets, suggesting a strong connection between verbosity and model uncertainty. We propose a simple yet effective cascade algorithm that replaces the verbose responses with the other model-generated responses, alleviating the VC of the Mistral model from 63.81% to 16.16% on the Qasper dataset.</abstract>
    <identifier type="citekey">zhang-etal-2025-demystify</identifier>
    <location>
        <url>https://aclanthology.org/2025.uncertainlp-main.14/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>160</start>
            <end>178</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Demystify Verbosity Compensation Behavior of Large Language Models
%A Zhang, Yusen
%A Snigdha Sarathi Das, Sarkar
%A Zhang, Rui
%Y Noidea, Noidea
%S Proceedings of the 2nd Workshop on Uncertainty-Aware NLP (UncertaiNLP 2025)
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-349-4
%F zhang-etal-2025-demystify
%X Recent work has revealed Large Language Models (LLMs) often exhibit undesirable behaviors, such as hallucination and toxicity, limiting their reliability and broader adoption. In this paper, we discover an understudied type of undesirable behavior of LLMs, which we term Verbosity Compensation (VC). VC is similar to the hesitation behavior of humans under uncertainty, compensating with excessive words such as repeating questions, introducing ambiguity, or providing excessive enumeration. We present the first work that analyzes Verbosity Compensation, explores its causes, and proposes a simple mitigating approach. Our experiments on five datasets of knowledge and reasoning-based QA tasks with 14 LLMs, reveal three conclusions. 1) A pervasive presence of VC across all models and all datasets. 2) The large performance gap between verbose and concise responses. We also demonstrate that this difference does not naturally diminish as LLM capability increases. 3) Higher uncertainty exhibited by VC responses across all five datasets, suggesting a strong connection between verbosity and model uncertainty. We propose a simple yet effective cascade algorithm that replaces the verbose responses with the other model-generated responses, alleviating the VC of the Mistral model from 63.81% to 16.16% on the Qasper dataset.
%U https://aclanthology.org/2025.uncertainlp-main.14/
%P 160-178

Markdown (Informal)

[Demystify Verbosity Compensation Behavior of Large Language Models](https://aclanthology.org/2025.uncertainlp-main.14/) (Zhang et al., UncertaiNLP 2025)

• Demystify Verbosity Compensation Behavior of Large Language Models (Zhang et al., UncertaiNLP 2025)

ACL

• Yusen Zhang, Sarkar Snigdha Sarathi Das, and Rui Zhang. 2025. Demystify Verbosity Compensation Behavior of Large Language Models. In Proceedings of the 2nd Workshop on Uncertainty-Aware NLP (UncertaiNLP 2025), pages 160–178, Suzhou, China. Association for Computational Linguistics.