mlsanityData to 14 Oct 2026

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

arXiv 2509.04458 ↗First posted 27 Aug 2025cs.CLcited by 3

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 11 of its 37 references that are in the map. First posted on arXiv 27 Aug 2025.

Builds on

11 of 37 references.

References · 11

  1. 2022Emergent Abilities of Large Language Modelscited by 2,512
  2. 2022Large Language Models Struggle to Learn Long-Tail Knowledgecited by 458
  3. 2021Neural Natural Language Processing for Unstructured Data in Electronic Health Records: a Reviewcited by 35
  4. 2023Identifying and Extracting Rare Disease Phenotypes with Large Language Modelscited by 18
  5. 2023An evaluation of GPT models for phenotype concept recognitioncited by 10
  6. 2024Generalizable and Scalable Multistage Biomedical Concept Normalization Leveraging Large Language Modelscited by 7
  7. 2024ChocoLlama: Lessons Learned From Teaching Llamas Dutchcited by 5
  8. 2024High Throughput Phenotyping of Physician Notes with Large Language and Hybrid NLP Modelscited by 4
  9. 2024EHRmonize: A Framework for Medical Concept Abstraction from Electronic Health Records using Large Language Modelscited by 4
  10. 2024A Large Language Model Outperforms Other Computational Approaches to the High-Throughput Phenotyping of Physician Notescited by 3
  11. 2024A Simplified Retriever to Improve Accuracy of Phenotype Normalizations by Large Language Modelscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025From Memorization to Generalization: Fine-Tuning Large Language Models for Biomedical Term-to-Identifier Normalizationcited by 1
  2. 2026Automated Standardization of Legacy Biomedical Metadata Using an Ontology-Constrained LLM Agentcited by 1
  3. 2026Latent Knowledge as a Predictor of Fact Acquisition in Fine-Tuned Large Language Modelscited by 0