Predicting Failures of LLMs to Link Biomedical Ontology Terms to Identifiers Evidence Across Models and Ontologies

Daniel B. Hier, Steven Keith Platt, Tayo Obafemi-Ajayi
8/27/2025
cs.CL

No confident code implementation found for this paper. 1 possible match can be shown below.

Abstract

Large language models often perform well on biomedical NLP tasks but may fail to link ontology terms to their correct identifiers. We investigate why these failures occur by analyzing predictions across two major ontologies, Human Phenotype Ontology and Gene Ontology, and two high-performing models, GPT-4o and LLaMa 3.1 405B. We evaluate nine candidate features related to term familiarity, identifier usage, morphology, and ontology structure. Univariate and multivariate analyses show that exposure to ontology identifiers is the strongest predictor of linking success.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. 1 weaker match is hidden by default — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Mapping Biomedical Ontology Terms to IDs: Effect of Domain Prevalence on Prediction Accuracy

88% match2024

From Memorization to Generalization: Fine-Tuning Large Language Models for Biomedical Term-to-Identifier Normalization

84% match2025

Do LLMs Dream of Ontologies?

83% match1 repo12024

Biomedical Language Models are Robust to Sub-optimal Tokenization

82% match1 repo42023

Latent Knowledge as a Predictor of Fact Acquisition in Fine-Tuned Large Language Models

80% match2026

Can Large Language Models abstract Medical Coded Language?

80% match2024

Cite this paper

@article{hier2025predicting,
  title  = {Predicting Failures of LLMs to Link Biomedical Ontology Terms to Identifiers Evidence Across Models and Ontologies},
  author = {Daniel B. Hier and Steven Keith Platt and Tayo Obafemi-Ajayi},
  year   = {2025},
  eprint = {2509.04458},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.04458v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.04458.svg)](https://codeofpaper.com/paper/2509.04458)

Discussion