Reproduce “A Survey on Open Information Extraction from Rule-based Model to Large Language Model” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Pai Liu, Wenyang Gao, Wenjie Dong, Lin Ai, Ziwei Gong, Songfang Huang, Zongsheng Li, Ehsan Hoque and 2 more
A comprehensive survey analyzes the evolution of Open Information Extraction (OpenIE) from its rule-based foundations through neural networks to the Large Language Model (LLM) era, detailing shifts in methodologies, task settings, and evaluation metrics from 2007 to 2024. The work further examines how LLMs are transforming OpenIE, converging its definition with standard information extraction, and outlining critical future research directions.
Key insights
- OpenIE research has progressed through three distinct paradigm shifts, evolving from rule-based and statistical methods to neural networks, and most recently, to generative approaches powered by Large Language Models.
- Large Language Models are blurring the traditional boundaries between OpenIE and standard IE by enabling flexible schema design and instruction-following, potentially leading to a unified information extraction paradigm.
- Valuable lessons and approaches from traditional OpenIE research, such as structured extraction methods and an emphasis on faithfulness, can inform and enhance the development of more robust and reliable LLM-based information extraction systems.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.