A systematic review of relation extraction task since the emergence of Transformers

2025-11-05

AI Summary

This survey addresses relation extraction (RE) research in the Transformer era (2019–2024). To overcome limitations of manual literature reviews, we propose the first automated framework for systematic literature acquisition and annotation tailored to RE. Our methodology integrates 34 survey papers, 64 benchmark datasets, and 104 models, enabling multidimensional analysis across methodological evolution, benchmark resources, and semantic web techniques. Unlike prior work, our approach constructs a structured, holistic knowledge map covering models, data, and evaluation—thereby clarifying developmental trajectories of dominant paradigms, including prompt learning, instruction tuning, and knowledge-enhanced modeling. We identify four critical open challenges: robustness to annotation noise, few-shot generalization, cross-domain transferability, and model interpretability. The resulting synthesis provides researchers with a reusable analytical framework and an authoritative reference system for advancing RE research.

Technology Category

• Natural Language Processing: Information Extraction
• Reasoning under Uncertainty: Relational Probabilistic Models
• Knowledge Representation and Reasoning: Common-Sense Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

This article presents a systematic review of relation extraction (RE) research since the advent of Transformer-based models. Using an automated framework to collect and annotate publications, we analyze 34 surveys, 64 datasets, and 104 models published between 2019 and 2024. The review highlights methodological advances, benchmark resources, and the integration of semantic web technologies. By consolidating results across multiple dimensions, the study identifies current trends, limitations, and open challenges, offering researchers and practitioners a comprehensive reference for understanding the evolution and future directions of RE.

Problem

• Analyzing relation extraction research since Transformer models emerged
• Reviewing methodological advances and benchmark datasets from 2019-2024
• Identifying current trends, limitations and open challenges in RE

Innovation

• Automated framework for collecting publications
• Analyzing surveys datasets and models systematically
• Integrating semantic web technologies in extraction