BeLink: Biomedical Entity Linking Meets Generative Re-Ranking

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Shlyk, Darya, Montanelli, Stefano, Hunter, Lawrence
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914588199485440
author Shlyk, Darya
Montanelli, Stefano
Hunter, Lawrence
author_facet Shlyk, Darya
Montanelli, Stefano
Hunter, Lawrence
contents Despite recent progress, Biomedical Entity Linking (BEL) with large language models (LLMs) remains computationally inefficient and challenging to deploy in practical settings. In this work, we demonstrate that instruction-tuning of open-source generative models can offer an effective solution when applied at the re-ranking stage of the BEL pipeline. We propose a set-wise instruction-tuning formulation that enables fast and accurate candidate selection. Our method demonstrates strong performance on multiple BEL benchmarks, yielding significant improvements in linking accuracy (3%-24%) while reducing inference time compared to the state-of-the-art. We integrate our generative re-ranker into BeLink, a modular, end-to-end system designed for practical real-world BEL applications.
format Preprint
id arxiv_https___arxiv_org_abs_2605_22501
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle BeLink: Biomedical Entity Linking Meets Generative Re-Ranking
Shlyk, Darya
Montanelli, Stefano
Hunter, Lawrence
Computation and Language
Artificial Intelligence
Information Retrieval
Despite recent progress, Biomedical Entity Linking (BEL) with large language models (LLMs) remains computationally inefficient and challenging to deploy in practical settings. In this work, we demonstrate that instruction-tuning of open-source generative models can offer an effective solution when applied at the re-ranking stage of the BEL pipeline. We propose a set-wise instruction-tuning formulation that enables fast and accurate candidate selection. Our method demonstrates strong performance on multiple BEL benchmarks, yielding significant improvements in linking accuracy (3%-24%) while reducing inference time compared to the state-of-the-art. We integrate our generative re-ranker into BeLink, a modular, end-to-end system designed for practical real-world BEL applications.
title BeLink: Biomedical Entity Linking Meets Generative Re-Ranking
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2605.22501