{"type":"ScholarlyArticle","title":"SynCABEL: Synthetic Contextualized Augmentation for Biomedical Entity Linking","authors":[{"name":"Adam Remaki","slug":"adam-remaki","given":"Adam","family":"Remaki","affiliation":""},{"name":"Christel Gérardin","slug":"christel-gerardin","given":"Christel","family":"Gérardin","affiliation":""},{"name":"Eulàlia Farré-Maduell","slug":"eulalia-farre-maduell","given":"Eulàlia","family":"Farré-Maduell","affiliation":""},{"name":"Martin Krallinger","slug":"martin-krallinger","given":"Martin","family":"Krallinger","affiliation":""},{"name":"Xavier Tannier","slug":"xavier-tannier","given":"Xavier","family":"Tannier","affiliation":""}],"abstract":"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.\r\n• HuggingFace Datasets & Models\r\n• GitHub Repository","doi":"10.24963/ijcai.2026/763","url":"https://ijcai.org/proceedings/2026/763","pdf":"https://ijcai.org/proceedings/2026/0763.pdf","markdown":"https://ijcai.org/proceedings/2026/763.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/763","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"AI and Health","track":"AI and Health","number":763,"pages":{"first":6859,"last":6867},"keywords":["Health data mining: Health data mining","Medical knowledge representation: Medical knowledge representation","LLM in medicine: LLM in medicine","EHR analysis: EHR analysis","Biomedical NLP: Biomedical NLP"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}