iMETRE: Incorporating Markers of Entity Types for Relation Extraction

TL;DR

This paper introduces iMETRE, a relation extraction method that uses entity type markers and fine-tuning on a financial dataset, achieving improved F1 scores and discussing potential limitations.

Contribution

The paper presents a novel approach incorporating typed entity markers for relation extraction in financial data, enhancing performance over existing methods.

Findings

Achieved an F1 score of 69.65% on REFinD dataset.

Demonstrated the effectiveness of typed entity markers in relation extraction.

Discussed limitations and future directions for the approach.

Abstract

Sentence-level relation extraction (RE) aims to identify the relationship between 2 entities given a contextual sentence. While there have been many attempts to solve this problem, the current solutions have a lot of room to improve. In this paper, we approach the task of relationship extraction in the financial dataset REFinD. Our approach incorporates typed entity markers representations and various models finetuned on the dataset, which has allowed us to achieve an F1 score of 69.65% on the validation set. Through this paper, we discuss various approaches and possible limitations.