How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Jinxin, Cao, Shulin, Shi, Jiaxin, Zhang, Tingjian, Nie, Lunyiu, Hu, Linmei, Hou, Lei, Li, Juanzi
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913390687944704
author Liu, Jinxin
Cao, Shulin
Shi, Jiaxin
Zhang, Tingjian
Nie, Lunyiu
Hu, Linmei
Hou, Lei
Li, Juanzi
author_facet Liu, Jinxin
Cao, Shulin
Shi, Jiaxin
Zhang, Tingjian
Nie, Lunyiu
Hu, Linmei
Hou, Lei
Li, Juanzi
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2401_05777
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering
Liu, Jinxin
Cao, Shulin
Shi, Jiaxin
Zhang, Tingjian
Nie, Lunyiu
Hu, Linmei
Hou, Lei
Li, Juanzi
Computation and Language
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.
title How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering
topic Computation and Language
url https://arxiv.org/abs/2401.05777