ComMA@FIRE 2020: Exploring Multilingual JointTraining across different Classification Tasks

Research output: Chapter in Book or Conference Publication/ProceedingConference Publicationpeer-review

Abstract

n this paper, we give a description of the systems submitted to the three tracks of FIRE 2020 - HateSpeech and Offensive Content Identification in Indo-European Languages (HASOC), Sentiment Analy-sis of Dravidian Languages in Code-Mixed Text and Event Detection from News in Indian Languages(EDNIL). While the first two tasks were binary and multi-class text classification problems, EDNIL wasa sequence classification problem. For all the three tracks, we jointly fine-tuned mBERT, DistilBERT,RoBERTa and XLM-R using the dataset from all the languages for the given task
Original languageEnglish (Ireland)
Title of host publicationWorking Notes of FIRE 2020-Forum for Information Retrieval Evaluation, Hyderabad, India
Place of PublicationHyderabad, India
Publication statusPublished - 1 Dec 2020

Authors (Note for portal: view the doc link for the full list of authors)

  • Authors
  • Kumar, Ritesh; Lahiri, Bornini; Ojha, Atul Kr. and Bansal, Akanksha

Fingerprint

Dive into the research topics of 'ComMA@FIRE 2020: Exploring Multilingual JointTraining across different Classification Tasks'. Together they form a unique fingerprint.

Cite this