SynCABEL: Synthetic Contextualized Augmentation for Biomedical Entity Linking
This addresses a bottleneck in biomedical entity linking for researchers and practitioners by reducing reliance on costly expert labeling, though it is incremental as it builds on existing methods with synthetic data.
The paper tackles the scarcity of expert-annotated training data in biomedical entity linking by introducing SynCABEL, a framework that uses large language models to generate synthetic training examples, achieving new state-of-the-art results on multilingual benchmarks and reducing the need for annotated data by up to 60%.
We present SynCABEL (Synthetic Contextualized Augmentation for Biomedical Entity Linking), a framework that addresses a central bottleneck in supervised biomedical entity linking (BEL): the scarcity of expert-annotated training data. SynCABEL leverages large language models to generate context-rich synthetic training examples for all candidate concepts in a target knowledge base, providing broad supervision without manual annotation. We demonstrate that SynCABEL, when combined with decoder-only models and guided inference establish new state-of-the-art results across three widely used multilingual benchmarks: MedMentions for English, QUAERO for French, and SPACCC for Spanish. Evaluating data efficiency, we show that SynCABEL reaches the performance of full human supervision using up to 60% less annotated data, substantially reducing reliance on labor-intensive and costly expert labeling. Finally, acknowledging that standard evaluation based on exact code matching often underestimates clinically valid predictions due to ontology redundancy, we introduce an LLM-as-a-judge protocol. This analysis reveals that SynCABEL significantly improves the rate of clinically valid predictions. Our synthetic datasets, models, and code are released to support reproducibility and future research.