English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

End-to-end relation extraction (E2ERE) is an important and realisticapplication of natural language processing (NLP) in biomedicine. In this paper,we aim to compare three prevailing paradigms for E2ERE using a complex datasetfocused on rare diseases involving discontinuous and nested entities. We usethe RareDis information extraction dataset to evaluate three competingapproaches (for E2ERE): NER $\rightarrow$ RE pipelines, joint sequence tosequence models, and generative pre-trained transformer (GPT) models. We usecomparable state-of-the-art models and best practices for each of theseapproaches and conduct error analyses to assess their failure modes. Ourfindings reveal that pipeline models are still the best, whilesequence-to-sequence models are not far behind; GPT models with eight times asmany parameters are worse than even sequence-to-sequence models and lose topipeline models by over 10 F1 points. Partial matches and discontinuousentities caused many NER errors contributing to lower overall E2E performances.We also verify these findings on a second E2ERE dataset for chemical-proteininteractions. Although generative LM-based methods are more suitable forzero-shot settings, when training data is available, our results show that itis better to work with more conventional models trained and tailored for E2ERE.More innovative methods are needed to marry the best of the both worlds fromsmaller encoder-decoder pipeline models and the larger GPT models to improveE2ERE. As of now, we see that well designed pipeline models offer substantialperformance gains at a lower cost and carbon footprint for E2ERE. Ourcontribution is also the first to conduct E2ERE for the RareDis dataset.

Comparison of pipeline, sequence-to-sequence, and GPT models for end-to-end relation extraction: experiments with the rare disease use-case