Papers › MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding...

MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretraining

27 Dec 2020EMNLP (ClinicalNLP) 2020 11arXiv:2012.13978archive 2025-07-28

Zhi Wen, Xing Han Lu, Siva Reddy

One of the biggest challenges that prohibit the use of many current NLP methods in clinical settings is the availability of public datasets. In this work, we present MeDAL, a large medical text dataset curated for abbreviation disambiguation, designed for natural language understanding pre-training in the medical domain. We pre-trained several models of common architectures on this dataset and empirically showed that such pre-training leads to improved performance and convergence speed when fine-tuning on downstream medical tasks.

PaperPDFConference PDFCode

Code

mcGill-NLP/medal officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Mortality PredictionNatural Language Understanding

Datasets

Introduced by this paper, per the archive.

MeDAL

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Mortality Prediction MIMIC-III ELECTRA (pretrained) Accuracy 0.8443 #9 of 13 Archive leaderboard report
Mortality Prediction MIMIC-III ELECTRA (from scratch) Accuracy 0.8325 #10 of 13 Archive leaderboard report
Mortality Prediction MIMIC-III LSTM+SA (pretrained) Accuracy 0.8298 #11 of 13 Archive leaderboard report
Mortality Prediction MIMIC-III LSTM (pretrained) Accuracy 0.828 #12 of 13 Archive leaderboard report
Mortality Prediction MIMIC-III LSTM+SA (from scratch) Accuracy 0.7996 #13 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutDense ConnectionsDropoutELECTRALSTMLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionSigmoid ActivationSoftmaxTanh ActivationWeight DecayWordPiece

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections