ANALOGYKB: Unlocking Analogical Reasoning of Language Models with A Million-scale Knowledge Base

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yuan, Siyu, Chen, Jiangjie, Sun, Changzhi, Liang, Jiaqing, Xiao, Yanghua, Yang, Deqing
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909205622947840
author Yuan, Siyu
Chen, Jiangjie
Sun, Changzhi
Liang, Jiaqing
Xiao, Yanghua
Yang, Deqing
author_facet Yuan, Siyu
Chen, Jiangjie
Sun, Changzhi
Liang, Jiaqing
Xiao, Yanghua
Yang, Deqing
contents Analogical reasoning is a fundamental cognitive ability of humans. However, current language models (LMs) still struggle to achieve human-like performance in analogical reasoning tasks due to a lack of resources for model training. In this work, we address this gap by proposing ANALOGYKB, a million-scale analogy knowledge base (KB) derived from existing knowledge graphs (KGs). ANALOGYKB identifies two types of analogies from the KGs: 1) analogies of the same relations, which can be directly extracted from the KGs, and 2) analogies of analogous relations, which are identified with a selection and filtering pipeline enabled by large language models (LLMs), followed by minor human efforts for data quality control. Evaluations on a series of datasets of two analogical reasoning tasks (analogy recognition and generation) demonstrate that ANALOGYKB successfully enables both smaller LMs and LLMs to gain better analogical reasoning capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2305_05994
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle ANALOGYKB: Unlocking Analogical Reasoning of Language Models with A Million-scale Knowledge Base
Yuan, Siyu
Chen, Jiangjie
Sun, Changzhi
Liang, Jiaqing
Xiao, Yanghua
Yang, Deqing
Computation and Language
Artificial Intelligence
Analogical reasoning is a fundamental cognitive ability of humans. However, current language models (LMs) still struggle to achieve human-like performance in analogical reasoning tasks due to a lack of resources for model training. In this work, we address this gap by proposing ANALOGYKB, a million-scale analogy knowledge base (KB) derived from existing knowledge graphs (KGs). ANALOGYKB identifies two types of analogies from the KGs: 1) analogies of the same relations, which can be directly extracted from the KGs, and 2) analogies of analogous relations, which are identified with a selection and filtering pipeline enabled by large language models (LLMs), followed by minor human efforts for data quality control. Evaluations on a series of datasets of two analogical reasoning tasks (analogy recognition and generation) demonstrate that ANALOGYKB successfully enables both smaller LMs and LLMs to gain better analogical reasoning capabilities.
title ANALOGYKB: Unlocking Analogical Reasoning of Language Models with A Million-scale Knowledge Base
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2305.05994