@inproceedings{ryu-etal-2023-retrieval,
title = "Retrieval-based Evaluation for {LLM}s: A Case Study in {K}orean Legal {QA}",
author = "Ryu, Cheol and
Lee, Seolhwa and
Pang, Subeen and
Choi, Chanyeol and
Choi, Hojun and
Min, Myeonggee and
Sohn, Jy-Yong",
editor = "Preo{\textcommabelow{t}}iuc-Pietro, Daniel and
Goanta, Catalina and
Chalkidis, Ilias and
Barrett, Leslie and
Spanakis, Gerasimos and
Aletras, Nikolaos",
booktitle = "Proceedings of the Natural Legal Language Processing Workshop 2023",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.nllp-1.13",
doi = "10.18653/v1/2023.nllp-1.13",
pages = "132--137",
abstract = "While large language models (LLMs) have demonstrated significant capabilities in text generation, their utilization in areas requiring domain-specific expertise, such as law, must be approached cautiously. This caution is warranted due to the inherent challenges associated with LLM-generated texts, including the potential presence of factual errors. Motivated by this issue, we propose Eval-RAG, a new evaluation method for LLM-generated texts. Unlike existing methods, Eval-RAG evaluates the validity of generated texts based on the related document that are collected by the retriever. In other words, Eval-RAG adopts the idea of retrieval augmented generation (RAG) for the purpose of evaluation. Our experimental results on Korean Legal Question-Answering (QA) tasks show that conventional LLM-based evaluation methods can be better aligned with Lawyers{'} evaluations, by combining with Eval-RAG. In addition, our qualitative analysis show that Eval-RAG successfully finds the factual errors in LLM-generated texts, while existing evaluation methods cannot.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ryu-etal-2023-retrieval">
<titleInfo>
<title>Retrieval-based Evaluation for LLMs: A Case Study in Korean Legal QA</title>
</titleInfo>
<name type="personal">
<namePart type="given">Cheol</namePart>
<namePart type="family">Ryu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seolhwa</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Subeen</namePart>
<namePart type="family">Pang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chanyeol</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hojun</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Myeonggee</namePart>
<namePart type="family">Min</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jy-Yong</namePart>
<namePart type="family">Sohn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Natural Legal Language Processing Workshop 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Preo\textcommabelowtiuc-Pietro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Catalina</namePart>
<namePart type="family">Goanta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ilias</namePart>
<namePart type="family">Chalkidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leslie</namePart>
<namePart type="family">Barrett</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gerasimos</namePart>
<namePart type="family">Spanakis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolaos</namePart>
<namePart type="family">Aletras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>While large language models (LLMs) have demonstrated significant capabilities in text generation, their utilization in areas requiring domain-specific expertise, such as law, must be approached cautiously. This caution is warranted due to the inherent challenges associated with LLM-generated texts, including the potential presence of factual errors. Motivated by this issue, we propose Eval-RAG, a new evaluation method for LLM-generated texts. Unlike existing methods, Eval-RAG evaluates the validity of generated texts based on the related document that are collected by the retriever. In other words, Eval-RAG adopts the idea of retrieval augmented generation (RAG) for the purpose of evaluation. Our experimental results on Korean Legal Question-Answering (QA) tasks show that conventional LLM-based evaluation methods can be better aligned with Lawyers’ evaluations, by combining with Eval-RAG. In addition, our qualitative analysis show that Eval-RAG successfully finds the factual errors in LLM-generated texts, while existing evaluation methods cannot.</abstract>
<identifier type="citekey">ryu-etal-2023-retrieval</identifier>
<identifier type="doi">10.18653/v1/2023.nllp-1.13</identifier>
<location>
<url>https://aclanthology.org/2023.nllp-1.13</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>132</start>
<end>137</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Retrieval-based Evaluation for LLMs: A Case Study in Korean Legal QA
%A Ryu, Cheol
%A Lee, Seolhwa
%A Pang, Subeen
%A Choi, Chanyeol
%A Choi, Hojun
%A Min, Myeonggee
%A Sohn, Jy-Yong
%Y Preo\textcommabelowtiuc-Pietro, Daniel
%Y Goanta, Catalina
%Y Chalkidis, Ilias
%Y Barrett, Leslie
%Y Spanakis, Gerasimos
%Y Aletras, Nikolaos
%S Proceedings of the Natural Legal Language Processing Workshop 2023
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F ryu-etal-2023-retrieval
%X While large language models (LLMs) have demonstrated significant capabilities in text generation, their utilization in areas requiring domain-specific expertise, such as law, must be approached cautiously. This caution is warranted due to the inherent challenges associated with LLM-generated texts, including the potential presence of factual errors. Motivated by this issue, we propose Eval-RAG, a new evaluation method for LLM-generated texts. Unlike existing methods, Eval-RAG evaluates the validity of generated texts based on the related document that are collected by the retriever. In other words, Eval-RAG adopts the idea of retrieval augmented generation (RAG) for the purpose of evaluation. Our experimental results on Korean Legal Question-Answering (QA) tasks show that conventional LLM-based evaluation methods can be better aligned with Lawyers’ evaluations, by combining with Eval-RAG. In addition, our qualitative analysis show that Eval-RAG successfully finds the factual errors in LLM-generated texts, while existing evaluation methods cannot.
%R 10.18653/v1/2023.nllp-1.13
%U https://aclanthology.org/2023.nllp-1.13
%U https://doi.org/10.18653/v1/2023.nllp-1.13
%P 132-137
Markdown (Informal)
[Retrieval-based Evaluation for LLMs: A Case Study in Korean Legal QA](https://aclanthology.org/2023.nllp-1.13) (Ryu et al., NLLP-WS 2023)
ACL