Counts@IITK at SemEval-2021 Task 8: SciBERT Based Entity And Semantic Relation Extraction For Scientific Data

TL;DR

This paper describes a SciBERT-based system for extracting quantities, entities, and relations from scientific text, achieving top ranks in several subtasks of the SemEval 2021 MeasEval challenge.

Contribution

The authors developed a novel span extraction and relation classification system using SciBERT, tailored for scientific data, and achieved competitive results in multiple subtasks.

Findings

Placed fifth overall in SemEval-2021 Task 8

First in Quantity and Unit subtasks

Second in MeasuredEntity, Modifier, and Qualifies subtasks

Abstract

This paper presents the system for SemEval 2021 Task 8 (MeasEval). MeasEval is a novel span extraction, classification, and relation extraction task focused on finding quantities, attributes of these quantities, and additional information, including the related measured entities, properties, and measurement contexts. Our submitted system, which placed fifth (team rank) on the leaderboard, consisted of SciBERT with [CLS] token embedding and CRF layer on top. We were also placed first in Quantity (tied) and Unit subtasks, second in MeasuredEntity, Modifier and Qualifies subtasks, and third in Qualifier subtask.