Zero-Shot Open-Schema Entity Structure Discovery

2025-06-04

AI Summary

Existing LLM-based entity structure extraction methods heavily rely on predefined schemas or annotated data, resulting in incomplete structural coverage. This paper proposes a zero-shot open-domain entity structure discovery framework that requires neither schema priors nor human annotations. Our core innovation is an “Augment–Refine–Unify” collaborative mechanism: leveraging LLM-driven iterative prompting, structured reasoning, and multi-step self-verification to model the mutual reinforcement between entities and their attributes, enabling end-to-end structure discovery. Experiments across three diverse domains demonstrate that our method significantly outperforms supervised baselines in structural completeness while exhibiting strong generalization—requiring no fine-tuning, in-context examples, or domain-specific adaptation.

Technology Category

• Machine Learning: Structured Learning
• Natural Language Processing: Information Extraction
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Entity structure extraction, which aims to extract entities and their associated attribute-value structures from text, is an essential task for text understanding and knowledge graph construction. Existing methods based on large language models (LLMs) typically rely heavily on predefined entity attribute schemas or annotated datasets, often leading to incomplete extraction results. To address these challenges, we introduce Zero-Shot Open-schema Entity Structure Discovery (ZOES), a novel approach to entity structure extraction that does not require any schema or annotated samples. ZOES operates via a principled mechanism of enrichment, refinement, and unification, based on the insight that an entity and its associated structure are mutually reinforcing. Experiments demonstrate that ZOES consistently enhances LLMs' ability to extract more complete entity structures across three different domains, showcasing both the effectiveness and generalizability of the method. These findings suggest that such an enrichment, refinement, and unification mechanism may serve as a principled approach to improving the quality of LLM-based entity structure discovery in various scenarios.

Problem

• Extracts entities and attribute-value structures without predefined schemas
• Addresses incomplete extraction in existing LLM-based methods
• Enhances entity structure discovery across multiple domains

Innovation

• Zero-shot open-schema entity discovery
• Mutual reinforcement enrichment mechanism
• No predefined schemas or annotations