Search Articles

View query in Help articles search

Search Results (1 to 1 of 1 Results)

Download search results: CSV END BibTex RIS


Enhancing Clinical Relevance of Pretrained Language Models Through Integration of External Knowledge: Case Study on Cardiovascular Diagnosis From Electronic Health Records

Enhancing Clinical Relevance of Pretrained Language Models Through Integration of External Knowledge: Case Study on Cardiovascular Diagnosis From Electronic Health Records

This framework consists of 3 major components, which are the base PLM, pretrained knowledge-specific adapters, and the knowledge controller (CTRL) that adaptively activates the adapters, as illustrated in Figure 2. Generally, when pretraining a knowledge adapter, the parameters of the base PLM are frozen, and only the adapter is optimized. In this way, we inject specific knowledge into an adapter.

Qiuhao Lu, Andrew Wen, Thien Nguyen, Hongfang Liu

JMIR AI 2024;3:e56932