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

Instruction-tuned generative Large language models (LLMs) like ChatGPT andBloomz possess excellent generalization abilities, but they face limitations inunderstanding radiology reports, particularly in the task of generating theIMPRESSIONS section from the FINDINGS section. They tend to generate eitherverbose or incomplete IMPRESSIONS, mainly due to insufficient exposure tomedical text data during training. We present a system which leverageslarge-scale medical text data for domain-adaptive pre-training ofinstruction-tuned LLMs to enhance its medical knowledge and performance onspecific medical tasks. We show that this system performs better in a zero-shotsetting than a number of pretrain-and-finetune adaptation methods on theIMPRESSIONS generation task, and ranks 1st among participating systems in Task1B: Radiology Report Summarization at the BioNLP 2023 workshop.

shs-nlp at RadSum23: Domain-Adaptive Pre-training of Instruction-tuned LLMs for Radiology Report Impression Generation