Enabling LLM Knowledge Analysis via Extensive Materialization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hu, Yujia, Nguyen, Tuan-Phong, Ghosh, Shrestha, Razniewski, Simon
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913873591795712
author Hu, Yujia
Nguyen, Tuan-Phong
Ghosh, Shrestha
Razniewski, Simon
author_facet Hu, Yujia
Nguyen, Tuan-Phong
Ghosh, Shrestha
Razniewski, Simon
contents Large language models (LLMs) have majorly advanced NLP and AI, and next to their ability to perform a wide range of procedural tasks, a major success factor is their internalized factual knowledge. Since Petroni et al. (2019), analyzing this knowledge has gained attention. However, most approaches investigate one question at a time via modest-sized pre-defined samples, introducing an ``availability bias'' (Tversky&Kahnemann, 1973) that prevents the analysis of knowledge (or beliefs) of LLMs beyond the experimenter's predisposition. To address this challenge, we propose a novel methodology to comprehensively materialize an LLM's factual knowledge through recursive querying and result consolidation. Our approach is a milestone for LLM research, for the first time providing constructive insights into the scope and structure of LLM knowledge (or beliefs). As a prototype, we build GPTKB, a knowledge base (KB) comprising 101 million relational triples for over 2.9 million entities from GPT-4o-mini. We use GPTKB to exemplarily analyze GPT-4o-mini's factual knowledge in terms of scale, accuracy, bias, cutoff and consistency, at the same time. GPTKB is accessible at https://gptkb.org
format Preprint
id arxiv_https___arxiv_org_abs_2411_04920
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enabling LLM Knowledge Analysis via Extensive Materialization
Hu, Yujia
Nguyen, Tuan-Phong
Ghosh, Shrestha
Razniewski, Simon
Computation and Language
Artificial Intelligence
Databases
Large language models (LLMs) have majorly advanced NLP and AI, and next to their ability to perform a wide range of procedural tasks, a major success factor is their internalized factual knowledge. Since Petroni et al. (2019), analyzing this knowledge has gained attention. However, most approaches investigate one question at a time via modest-sized pre-defined samples, introducing an ``availability bias'' (Tversky&Kahnemann, 1973) that prevents the analysis of knowledge (or beliefs) of LLMs beyond the experimenter's predisposition. To address this challenge, we propose a novel methodology to comprehensively materialize an LLM's factual knowledge through recursive querying and result consolidation. Our approach is a milestone for LLM research, for the first time providing constructive insights into the scope and structure of LLM knowledge (or beliefs). As a prototype, we build GPTKB, a knowledge base (KB) comprising 101 million relational triples for over 2.9 million entities from GPT-4o-mini. We use GPTKB to exemplarily analyze GPT-4o-mini's factual knowledge in terms of scale, accuracy, bias, cutoff and consistency, at the same time. GPTKB is accessible at https://gptkb.org
title Enabling LLM Knowledge Analysis via Extensive Materialization
topic Computation and Language
Artificial Intelligence
Databases
url https://arxiv.org/abs/2411.04920