Abstract:Abstract Background The abundance of biomedical text data coupled with advances in natural language processing (NLP) is resulting in novel biomedical NLP (BioNLP) applications. These NLP applications, or tasks, are reliant on the availability of domain-specific language models (LMs) that are trained on a massive amount of data. Most of the existing domain-specific LMs adopted bidirectional encoder representations from transformers (BERT) architecture which has limitations, and their generalizability is unproven as there is an absence of baseline results among common BioNLP tasks. Results We present 8 variants of BioALBERT, a domain-specific adaptation of a lite bidirectional encoder representations from transformers (ALBERT), trained on biomedical (PubMed and PubMed Central) and clinical (MIMIC-III) corpora and fine-tuned for 6 different tasks across 20 benchmark datasets. Experiments show that a large variant of BioALBERT trained on PubMed outperforms the state-of-the-art on named-entity recognition (+ 11.09% BLURB score improvement), relation extraction (+ 0.80% BLURB score), sentence similarity (+ 1.05% BLURB score), document classification (+ 0.62% F1-score), and question answering (+ 2.83% BLURB score). It represents a new state-of-the-art in 5 out of 6 benchmark BioNLP tasks. Conclusions The large variant of BioALBERT trained on PubMed achieved a higher BLURB score than previous state-of-the-art models on 5 of the 6 benchmark BioNLP tasks. Depending on the task, 5 different variants of BioALBERT outperformed previous state-of-the-art models on 17 of the 20 benchmark datasets, showing that our model is robust and generalizable in the common BioNLP tasks. We have made BioALBERT freely available which will help the BioNLP community avoid computational cost of training and establish a new set of baselines for future efforts across a broad range of BioNLP tasks.

[RE] ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

AUDIO ALBERT: A LITE BERT FOR SELF-SUPERVISED LEARNING OF AUDIO REPRESENTATION

Evaluation of BERT and ALBERT Sentence Embedding Performance on Downstream NLP Tasks

ARoBERT: An ASR Robust Pre-Trained Language Model for Spoken Language Understanding

KoreALBERT: Pretraining a Lite BERT Model for Korean Language Understanding

Innovative Bert-based Reranking Language Models for Speech Recognition

Are All Languages Created Equal in Multilingual BERT?

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

On Robustness and Bias Analysis of BERT-Based Relation Extraction

ABNIRML: Analyzing the Behavior of Neural IR Models

Grid Search Hyperparameter Benchmarking of BERT, ALBERT, and LongFormer on DuoRC

Ensemble ALBERT on SQuAD 2.0

Universal Text Representation from BERT: An Empirical Study

CoBERT: Self-Supervised Speech Representation Learning Through Code Representation Learning

Benchmarking for biomedical natural language processing tasks with a domain specific ALBERT

RelBERT: Embedding Relations with Language Models

ALBERT-Based Personalized Educational Recommender System: Enhancing Students’ Learning Outcomes in Online Learning

BERTer: The Efficient One

Benchmarking Retrieval-Augmented Large Language Models in Biomedical NLP: Application, Robustness, and Self-Awareness

mALBERT: Is a Compact Multilingual BERT Model Still Worth It?

Audio DistilBERT: A Distilled Audio BERT for Speech Representation Learning.