MétaCan
Menu
Back to cohort
Record W4390789641 · doi:10.1017/s1351324923000542

Lightweight transformers for clinical natural language processing

2024· article· en· W4390789641 on OpenAlexfundno aff
Omid Rohanian, Mohammadmahdi Nouriborji, Hannah Jauncey, Samaneh Kouchaki, Farhad Nooralahzadeh, Lei Clifton, Laura Merson, David A. Clifton

Bibliographic record

VenueNatural Language Engineering · 2024
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsnot available
FundersNIHR Imperial Biomedical Research CentreInstituto de Salud Carlos IIIMedical Research CouncilNational Institutes of HealthKementerian Kesihatan MalaysiaAll-India Institute of Medical SciencesHorizon 2020 Framework ProgrammePrince Charles Hospital FoundationForeign, Commonwealth and Development OfficeNational Institute for Health Research Health Protection Research UnitUniversity of OxfordNorges ForskningsrådImperial College LondonConselho Nacional de Desenvolvimento Científico e TecnológicoUniversity of Cape TownPublic Health EnglandWellcome TrustUniversity College DublinSunnybrook Research InstituteInstitut National de la Santé et de la Recherche MédicaleCanadian Institutes of Health ResearchNational Institute for Health and Care ResearchMinistero della SaluteEuropean Federation of Pharmaceutical Industries and AssociationsEuropean CommissionBill and Melinda Gates Foundation
KeywordsComputer scienceTransformerNatural language processingArtificial intelligenceElectrical engineering

Abstract

fetched live from OpenAlex

Abstract Specialised pre-trained language models are becoming more frequent in Natural language Processing (NLP) since they can potentially outperform models trained on generic texts. BioBERT (Sanh et al., Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. arXiv preprint arXiv: 1910.01108, 2019) and BioClinicalBERT (Alsentzer et al., Publicly available clinical bert embeddings. In Proceedings of the 2nd Clinical Natural Language Processing Workshop, pp. 72–78, 2019) are two examples of such models that have shown promise in medical NLP tasks. Many of these models are overparametrised and resource-intensive, but thanks to techniques like knowledge distillation, it is possible to create smaller versions that perform almost as well as their larger counterparts. In this work, we specifically focus on development of compact language models for processing clinical texts (i.e. progress notes, discharge summaries, etc). We developed a number of efficient lightweight clinical transformers using knowledge distillation and continual learning, with the number of parameters ranging from $15$ million to $65$ million. These models performed comparably to larger models such as BioBERT and ClinicalBioBERT and significantly outperformed other compact models trained on general or biomedical data. Our extensive evaluation was done across several standard datasets and covered a wide range of clinical text-mining tasks, including natural language inference, relation extraction, named entity recognition and sequence classification. To our knowledge, this is the first comprehensive study specifically focused on creating efficient and compact transformers for clinical NLP tasks. The models and code used in this study can be found on our Huggingface profile at https://huggingface.co/nlpie and Github page at https://github.com/nlpie-research/Lightweight-Clinical-Transformers , respectively, promoting reproducibility of our results.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.015
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.014
Threshold uncertainty score0.045

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.015
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0020.001
Science and technology studies0.0000.001
Scholarly communication0.0020.006
Open science0.0020.004
Research integrity0.0010.003
Insufficient payload (model declined to judge)0.0140.009

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.012
GPT teacher head0.304
Teacher spread0.293 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations30
Published2024
Admission routes1
Has abstractyes

Explore more

Same venueNatural Language EngineeringSame topicTopic ModelingFrench-language works237,207