ReLiK: Retrieve, Read and LinK: Fast and Accurate Entity Linking and Relation Extraction on an Academic Budget
Primary Area: representation learning for computer vision, audio, language, and other modalities
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: Information Extraction, Entity Linking, Relation Extraction, Natural Language Processing, NLP
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
TL;DR: We propose a new unified framework for closed Information Extraction capable of Entity Linking and Relation Extraction with high performance on an academic-budget.
Abstract: Entity Linking (EL) and Relation Extraction (RE) are fundamental tasks in Natural Language Processing, serving as critical components in various applications such as Information Retrieval, Question Answering, and Knowledge Graph Construction. However, existing approaches often suffer from either a lack of flexibility, low-performance issues, or computational inefficiency. In this paper, we propose ReLiK, a Retriever-Reader architecture, where, given an input text, the Retriever module undertakes the identification of candidate entities or relations that could potentially appear within the text. Subsequently, the Reader module is tasked to discern the pertinent retrieved entities or relations and establish their alignment with the corresponding textual spans. Notably, we put forward an innovative input representation that incorporates the candidate entities or relations alongside the text, making it possible to link entities or extract relations in a single forward pass in contrast with previous Retriever-Reader-based methods, which necessitate a forward pass for each candidate. Our formulation of EL and RE achieves state-of-the-art performance in both in-domain and out-of-domain benchmarks while using academic budget training and with up to 40x inference speed with respect to other competitors. Finally, we propose a model for closed Information Extraction (cIE), i.e. EL + RE, which sets a new state of the art by employing a shared Reader that simultaneously extracts entities and relations.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 9353
Loading