Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LightKGG: Simple and Efficient Knowledge Graph Generation from Textual Data

Teng Lin

TL;DR

LightKGG tackles the KG construction bottleneck by enabling accurate knowledge graph extraction from text using small language models, avoiding costly heavy semantic parsing. It introduces two innovations: context-integrated graph extraction, which unifies entities, relations, and contextual cues into a cohesive graph, and topology-enhanced relation discovery, which uses graph structure to infer relationships with lightweight reasoning. Empirical results on SciERC and MINE demonstrate that LightKGG achieves close to LLM-baseline performance in Entity-F1 and Relation-F1 while substantially reducing compute, with ablations confirming the critical roles of context integration and topology-guided inference. The approach democratizes KG generation for resource-constrained environments and lays groundwork for extensions to multilingual and multimodal data.

Abstract

The scarcity of high-quality knowledge graphs (KGs) remains a critical bottleneck for downstream AI applications, as existing extraction methods rely heavily on error-prone pattern-matching techniques or resource-intensive large language models (LLMs). While recent tools leverage LLMs to generate KGs, their computational demands limit accessibility for low-resource environments. Our paper introduces LightKGG, a novel framework that enables efficient KG extraction from textual data using small-scale language models (SLMs) through two key technical innovations: (1) Context-integrated Graph extraction integrates contextual information with nodes and edges into a unified graph structure, reducing the reliance on complex semantic processing while maintaining more key information; (2) Topology-enhanced relationship inference leverages the inherent topology of the extracted graph to efficiently infer relationships, enabling relationship discovery without relying on complex language understanding capabilities of LLMs. By enabling accurate KG construction with minimal hardware requirements, this work bridges the gap between automated knowledge extraction and practical deployment scenarios while introducing scientifically rigorous methods for optimizing SLM efficiency in structured NLP tasks.

LightKGG: Simple and Efficient Knowledge Graph Generation from Textual Data

TL;DR

LightKGG tackles the KG construction bottleneck by enabling accurate knowledge graph extraction from text using small language models, avoiding costly heavy semantic parsing. It introduces two innovations: context-integrated graph extraction, which unifies entities, relations, and contextual cues into a cohesive graph, and topology-enhanced relation discovery, which uses graph structure to infer relationships with lightweight reasoning. Empirical results on SciERC and MINE demonstrate that LightKGG achieves close to LLM-baseline performance in Entity-F1 and Relation-F1 while substantially reducing compute, with ablations confirming the critical roles of context integration and topology-guided inference. The approach democratizes KG generation for resource-constrained environments and lays groundwork for extensions to multilingual and multimodal data.

Abstract

The scarcity of high-quality knowledge graphs (KGs) remains a critical bottleneck for downstream AI applications, as existing extraction methods rely heavily on error-prone pattern-matching techniques or resource-intensive large language models (LLMs). While recent tools leverage LLMs to generate KGs, their computational demands limit accessibility for low-resource environments. Our paper introduces LightKGG, a novel framework that enables efficient KG extraction from textual data using small-scale language models (SLMs) through two key technical innovations: (1) Context-integrated Graph extraction integrates contextual information with nodes and edges into a unified graph structure, reducing the reliance on complex semantic processing while maintaining more key information; (2) Topology-enhanced relationship inference leverages the inherent topology of the extracted graph to efficiently infer relationships, enabling relationship discovery without relying on complex language understanding capabilities of LLMs. By enabling accurate KG construction with minimal hardware requirements, this work bridges the gap between automated knowledge extraction and practical deployment scenarios while introducing scientifically rigorous methods for optimizing SLM efficiency in structured NLP tasks.
Paper Structure (18 sections, 1 figure, 2 tables)

This paper contains 18 sections, 1 figure, 2 tables.

Table of Contents

  1. Introduction
  2. Related works
  3. Rule-Based and Pattern-Matching Systems
  4. LLM-Driven KG Generation
  5. Hybrid and Lightweight Methods
  6. LightKGG
  7. LightKGG Framework
  8. Context-integrated Graph Extraction
  9. Graph Aggregation
  10. Topology-enhanced Relation Discovery
  11. Experiment
  12. Settings
  13. Ablation Study
  14. Main Results
  15. Ablation study results
  16. ...and 3 more sections

Figures (1)

  • Figure 1: Overview of LightKGG method.