@inproceedings{zhang-etal-2024-calibrating,
title = "Calibrating the Confidence of Large Language Models by Eliciting Fidelity",
author = "Zhang, Mozhi and
Huang, Mianqiu and
Shi, Rundong and
Guo, Linsen and
Peng, Chong and
Yan, Peng and
Zhou, Yaqian and
Qiu, Xipeng",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.173",
pages = "2959--2979",
abstract = "Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit overconfidence, where the expressed confidence does not accurately calibrate with their correctness rate. In this paper, we decompose the language model confidence into the \textit{Uncertainty} about the question and the \textit{Fidelity} to the answer generated by language models. Then, we propose a plug-and-play method, \textit{UF Calibration}, to estimate the confidence of language models. Our method has shown good calibration performance by conducting experiments with 6 RLHF-LMs on four MCQA datasets. Moreover, we propose two novel metrics, IPR and CE, to evaluate the calibration of the model, and we have conducted a detailed discussion on \textit{Truly Well-Calibrated Confidence} for large language models. Our method could serve as a strong baseline, and we hope that this work will provide some insights into the model confidence calibration.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2024-calibrating">
<titleInfo>
<title>Calibrating the Confidence of Large Language Models by Eliciting Fidelity</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mozhi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mianqiu</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rundong</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linsen</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chong</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peng</namePart>
<namePart type="family">Yan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaqian</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xipeng</namePart>
<namePart type="family">Qiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit overconfidence, where the expressed confidence does not accurately calibrate with their correctness rate. In this paper, we decompose the language model confidence into the Uncertainty about the question and the Fidelity to the answer generated by language models. Then, we propose a plug-and-play method, UF Calibration, to estimate the confidence of language models. Our method has shown good calibration performance by conducting experiments with 6 RLHF-LMs on four MCQA datasets. Moreover, we propose two novel metrics, IPR and CE, to evaluate the calibration of the model, and we have conducted a detailed discussion on Truly Well-Calibrated Confidence for large language models. Our method could serve as a strong baseline, and we hope that this work will provide some insights into the model confidence calibration.</abstract>
<identifier type="citekey">zhang-etal-2024-calibrating</identifier>
<location>
<url>https://aclanthology.org/2024.emnlp-main.173</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>2959</start>
<end>2979</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Calibrating the Confidence of Large Language Models by Eliciting Fidelity
%A Zhang, Mozhi
%A Huang, Mianqiu
%A Shi, Rundong
%A Guo, Linsen
%A Peng, Chong
%A Yan, Peng
%A Zhou, Yaqian
%A Qiu, Xipeng
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F zhang-etal-2024-calibrating
%X Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit overconfidence, where the expressed confidence does not accurately calibrate with their correctness rate. In this paper, we decompose the language model confidence into the Uncertainty about the question and the Fidelity to the answer generated by language models. Then, we propose a plug-and-play method, UF Calibration, to estimate the confidence of language models. Our method has shown good calibration performance by conducting experiments with 6 RLHF-LMs on four MCQA datasets. Moreover, we propose two novel metrics, IPR and CE, to evaluate the calibration of the model, and we have conducted a detailed discussion on Truly Well-Calibrated Confidence for large language models. Our method could serve as a strong baseline, and we hope that this work will provide some insights into the model confidence calibration.
%U https://aclanthology.org/2024.emnlp-main.173
%P 2959-2979
Markdown (Informal)
[Calibrating the Confidence of Large Language Models by Eliciting Fidelity](https://aclanthology.org/2024.emnlp-main.173) (Zhang et al., EMNLP 2024)
ACL
- Mozhi Zhang, Mianqiu Huang, Rundong Shi, Linsen Guo, Chong Peng, Peng Yan, Yaqian Zhou, and Xipeng Qiu. 2024. Calibrating the Confidence of Large Language Models by Eliciting Fidelity. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 2959–2979, Miami, Florida, USA. Association for Computational Linguistics.