@inproceedings{liu-etal-2024-proficient,
title = "How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering",
author = "Liu, Jinxin and
Cao, Shulin and
Shi, Jiaxin and
Zhang, Tingjian and
Nie, Lunyiu and
Hu, Linmei and
Hou, Lei and
Li, Juanzi",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand and virtual meeting",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.45",
doi = "10.18653/v1/2024.findings-acl.45",
pages = "792--815",
abstract = "Knowledge Base Question Answering (KBQA) aims to answer natural language questions based on facts in knowledge bases. A typical approach to KBQA is semantic parsing, which translates a question into an executable logical form in a formal language. Recent works leverage the capabilities of large language models (LLMs) for logical form generation to improve performance. However, although it is validated that LLMs are capable of solving some KBQA problems, there has been little discussion on the differences in LLMs{'} proficiency in formal languages used in semantic parsing. In this work, we propose to evaluate the understanding and generation ability of LLMs to deal with differently structured logical forms by examining the inter-conversion of natural and formal language through in-context learning of LLMs. Extensive experiments with models of different sizes show that state-of-the-art LLMs can understand formal languages as well as humans, but generating correct logical forms given a few examples remains a challenge. Most importantly, our results also indicate that LLMs exhibit considerable sensitivity. In general, the formal language with a lower formalization level, i.e., the more similar it is to natural language, is more friendly to LLMs. Code and data can be found at https://github.com/Matthewlliu/structure{\_}probe.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2024-proficient">
<titleInfo>
<title>How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jinxin</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shulin</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaxin</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tingjian</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lunyiu</namePart>
<namePart type="family">Nie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linmei</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Hou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juanzi</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand and virtual meeting</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Knowledge Base Question Answering (KBQA) aims to answer natural language questions based on facts in knowledge bases. A typical approach to KBQA is semantic parsing, which translates a question into an executable logical form in a formal language. Recent works leverage the capabilities of large language models (LLMs) for logical form generation to improve performance. However, although it is validated that LLMs are capable of solving some KBQA problems, there has been little discussion on the differences in LLMs’ proficiency in formal languages used in semantic parsing. In this work, we propose to evaluate the understanding and generation ability of LLMs to deal with differently structured logical forms by examining the inter-conversion of natural and formal language through in-context learning of LLMs. Extensive experiments with models of different sizes show that state-of-the-art LLMs can understand formal languages as well as humans, but generating correct logical forms given a few examples remains a challenge. Most importantly, our results also indicate that LLMs exhibit considerable sensitivity. In general, the formal language with a lower formalization level, i.e., the more similar it is to natural language, is more friendly to LLMs. Code and data can be found at https://github.com/Matthewlliu/structure_probe.</abstract>
<identifier type="citekey">liu-etal-2024-proficient</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.45</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.45</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>792</start>
<end>815</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering
%A Liu, Jinxin
%A Cao, Shulin
%A Shi, Jiaxin
%A Zhang, Tingjian
%A Nie, Lunyiu
%A Hu, Linmei
%A Hou, Lei
%A Li, Juanzi
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand and virtual meeting
%F liu-etal-2024-proficient
%X Knowledge Base Question Answering (KBQA) aims to answer natural language questions based on facts in knowledge bases. A typical approach to KBQA is semantic parsing, which translates a question into an executable logical form in a formal language. Recent works leverage the capabilities of large language models (LLMs) for logical form generation to improve performance. However, although it is validated that LLMs are capable of solving some KBQA problems, there has been little discussion on the differences in LLMs’ proficiency in formal languages used in semantic parsing. In this work, we propose to evaluate the understanding and generation ability of LLMs to deal with differently structured logical forms by examining the inter-conversion of natural and formal language through in-context learning of LLMs. Extensive experiments with models of different sizes show that state-of-the-art LLMs can understand formal languages as well as humans, but generating correct logical forms given a few examples remains a challenge. Most importantly, our results also indicate that LLMs exhibit considerable sensitivity. In general, the formal language with a lower formalization level, i.e., the more similar it is to natural language, is more friendly to LLMs. Code and data can be found at https://github.com/Matthewlliu/structure_probe.
%R 10.18653/v1/2024.findings-acl.45
%U https://aclanthology.org/2024.findings-acl.45
%U https://doi.org/10.18653/v1/2024.findings-acl.45
%P 792-815
Markdown (Informal)
[How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering](https://aclanthology.org/2024.findings-acl.45) (Liu et al., Findings 2024)
ACL
- Jinxin Liu, Shulin Cao, Jiaxin Shi, Tingjian Zhang, Lunyiu Nie, Linmei Hu, Lei Hou, and Juanzi Li. 2024. How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering. In Findings of the Association for Computational Linguistics ACL 2024, pages 792–815, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.