@inproceedings{fan-etal-2025-camieval,
title = "{CAMIE}val: Enhancing {NLG} Evaluation through Multidimensional Comparative Instruction-Following Analysis",
author = "Fan, Ziyue and
He, Junliang and
Xiaoqing, Li and
Kuang, Shaohui and
Song, Kai and
Zhou, Yaqian and
Qiu, Xipeng",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.438/",
doi = "10.18653/v1/2025.naacl-long.438",
pages = "8708--8733",
ISBN = "979-8-89176-189-6",
abstract = "With the rapid development of large language models (LLMs), due to their strong performance across various fields, LLM-based evaluation methods (LLM-as-a-Judge) have become widely used in natural language generation (NLG) evaluation. However, these methods encounter the following challenges: (1) distinguishing instruction-following ability, (2) being applicable across diverse NLG tasks, and (3) identifying low-quality outputs. To address these issues, we propose CAMIEval, a multidimensional comparative evaluation method based on instruction-following. Specifically, we define three fundamental dimensions of instruction-following: relevance, factuality, and adherence. Subsequently, we introduce a concrete Chain-of-Thoughts (ConcreteCoT) process to enhance the accuracy of evaluations. In addition, we trained a ``regrettable model'' RegretLM to generate low-quality outputs, which helps the evaluator better identify the potential shortcomings of the candidate output by comparing low-quality outputs with reference outputs. Through this comparison, the evaluator can generate instruction-specific dimensions that complement the fundamental dimensions, forming a more comprehensive evaluation metric system. Experiments on two NLG evaluation benchmarks demonstrate that CAMIEval consistently outperforms existing methods in terms of correlation with human evaluations, providing a general and accurate framework for evaluating the outputs of LLMs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fan-etal-2025-camieval">
<titleInfo>
<title>CAMIEval: Enhancing NLG Evaluation through Multidimensional Comparative Instruction-Following Analysis</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ziyue</namePart>
<namePart type="family">Fan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junliang</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Li</namePart>
<namePart type="family">Xiaoqing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shaohui</namePart>
<namePart type="family">Kuang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">Song</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>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>With the rapid development of large language models (LLMs), due to their strong performance across various fields, LLM-based evaluation methods (LLM-as-a-Judge) have become widely used in natural language generation (NLG) evaluation. However, these methods encounter the following challenges: (1) distinguishing instruction-following ability, (2) being applicable across diverse NLG tasks, and (3) identifying low-quality outputs. To address these issues, we propose CAMIEval, a multidimensional comparative evaluation method based on instruction-following. Specifically, we define three fundamental dimensions of instruction-following: relevance, factuality, and adherence. Subsequently, we introduce a concrete Chain-of-Thoughts (ConcreteCoT) process to enhance the accuracy of evaluations. In addition, we trained a “regrettable model” RegretLM to generate low-quality outputs, which helps the evaluator better identify the potential shortcomings of the candidate output by comparing low-quality outputs with reference outputs. Through this comparison, the evaluator can generate instruction-specific dimensions that complement the fundamental dimensions, forming a more comprehensive evaluation metric system. Experiments on two NLG evaluation benchmarks demonstrate that CAMIEval consistently outperforms existing methods in terms of correlation with human evaluations, providing a general and accurate framework for evaluating the outputs of LLMs.</abstract>
<identifier type="citekey">fan-etal-2025-camieval</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.438</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.438/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>8708</start>
<end>8733</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CAMIEval: Enhancing NLG Evaluation through Multidimensional Comparative Instruction-Following Analysis
%A Fan, Ziyue
%A He, Junliang
%A Xiaoqing, Li
%A Kuang, Shaohui
%A Song, Kai
%A Zhou, Yaqian
%A Qiu, Xipeng
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F fan-etal-2025-camieval
%X With the rapid development of large language models (LLMs), due to their strong performance across various fields, LLM-based evaluation methods (LLM-as-a-Judge) have become widely used in natural language generation (NLG) evaluation. However, these methods encounter the following challenges: (1) distinguishing instruction-following ability, (2) being applicable across diverse NLG tasks, and (3) identifying low-quality outputs. To address these issues, we propose CAMIEval, a multidimensional comparative evaluation method based on instruction-following. Specifically, we define three fundamental dimensions of instruction-following: relevance, factuality, and adherence. Subsequently, we introduce a concrete Chain-of-Thoughts (ConcreteCoT) process to enhance the accuracy of evaluations. In addition, we trained a “regrettable model” RegretLM to generate low-quality outputs, which helps the evaluator better identify the potential shortcomings of the candidate output by comparing low-quality outputs with reference outputs. Through this comparison, the evaluator can generate instruction-specific dimensions that complement the fundamental dimensions, forming a more comprehensive evaluation metric system. Experiments on two NLG evaluation benchmarks demonstrate that CAMIEval consistently outperforms existing methods in terms of correlation with human evaluations, providing a general and accurate framework for evaluating the outputs of LLMs.
%R 10.18653/v1/2025.naacl-long.438
%U https://aclanthology.org/2025.naacl-long.438/
%U https://doi.org/10.18653/v1/2025.naacl-long.438
%P 8708-8733
Markdown (Informal)
[CAMIEval: Enhancing NLG Evaluation through Multidimensional Comparative Instruction-Following Analysis](https://aclanthology.org/2025.naacl-long.438/) (Fan et al., NAACL 2025)
ACL