cs60075_team2 at SemEval-2021 Task 1 : Lexical Complexity Prediction using Transformer-based Language Models pre-trained on various text corpora

This paper describes the performance of the team cs60075_team2 at SemEval\n2021 Task 1 - Lexical Complexity Prediction. The main contribution of this\npaper is to fine-tune transformer-based language models pre-trained on several\ntext corpora, some being general (E.g., Wikipedia, BooksCorpus), some being the\ncorpora from which the CompLex Dataset was extracted, and others being from\nother specific domains such as Finance, Law, etc. We perform ablation studies\non selecting the transformer models and how their individual complexity scores\nare aggregated to get the resulting complexity scores. Our method achieves a\nbest Pearson Correlation of $0.784$ in sub-task 1 (single word) and $0.836$ in\nsub-task 2 (multiple word expressions).\n