Pith. sign in

REVIEW 5 cited by

Few-shot Learning for Named Entity Recognition in Medical Text

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1811.05468 v1 pith:2UYLM6F6 submitted 2018-11-13 cs.CL cs.LGstat.ML

Maximilian Hofer , Andrey Kormilitzin , Paul Goldberg , Alejo Nevado-Holgado This is my paper · ORCID
classification cs.CLcs.LGstat.ML
keywords annotatedexamplesmodelsstate-of-the-artachievableentitygainsmedical
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Deep neural network models have recently achieved state-of-the-art performance gains in a variety of natural language processing (NLP) tasks (Young, Hazarika, Poria, & Cambria, 2017). However, these gains rely on the availability of large amounts of annotated examples, without which state-of-the-art performance is rarely achievable. This is especially inconvenient for the many NLP fields where annotated examples are scarce, such as medical text. To improve NLP models in this situation, we evaluate five improvements on named entity recognition (NER) tasks when only ten annotated examples are available: (1) layer-wise initialization with pre-trained weights, (2) hyperparameter tuning, (3) combining pre-training data, (4) custom word embeddings, and (5) optimizing out-of-vocabulary (OOV) words. Experimental results show that the F1 score of 69.3% achievable by state-of-the-art models can be improved to 78.87%.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Beyond Text: Characterizing Domain Expert Needs in Document Research

    cs.CL 2025-04 conditional novelty 6.0 of 10

    A qualitative interview study finds that expert document research is iterative, personal, and socially contextual, and argues NLP tools should model documents as social objects, not just text.

  2. Low-Resource Name Tagging Learned with Weakly Labeled Data

    cs.CL 2019-08 conditional novelty 6.0 of 10

    A weakly supervised NER model combines classification pretraining on noisy data with Partial-CRFs and non-entity sampling on clean data, improving F1 across five low-resource languages and a fine-grained food domain.

  3. Weakly Supervised Medical Entity Extraction and Linking for Chief Complaints

    cs.CL 2025-09 conditional novelty 5.0 of 10

    A split-and-match weak supervision pipeline trains BERT and BiLSTM models to extract and link medical entities from chief complaints without human annotation, achieving 67.5 F1 on a clinician-labeled test set.

  4. Task Decomposition for Efficient Annotation

    cs.CL 2026-06 unverdicted novelty 4.0 of 10

    Decomposing annotation tasks using centers from centering theory reduces aggregate inferential load via a degrees-of-freedom model and enables better sub-task allocation.

  5. Extracting OPQRST in Electronic Health Records using Large Language Models with Reasoning

    cs.CL 2025-09 conditional novelty 4.0 of 10

    Reasoning-style prompts improve few-shot LLM extraction of OPQRST items from EHR notes, but the result rests on an 85-note single-annotator evaluation with an LLM judge.

Pith tools