aiqbench analysis
OntologyAligner leverages large language models and hierarchical reasoning to solve the specific challenge of biomedical text normalization, achieving state-of-the-art accuracy on Human Phenotype Ontology datasets. By integrating ontology-aligned retrieval with selective hierarchy-guided refinement, this framework demonstrates a robust method for mapping free-text expressions to standardized concepts across multiple biological taxonomies. The release of PhenoNormBench provides a critical benchmark for evaluating future models in structured biomedical data integration tasks.
OntologyAligner: Ontology-Aligned Retrieval and Hierarchy-Guided Large Language Model Reranking for Biomedical Ontology Normalization
arXiv:2609.10055v1 Announce Type: new Abstract: Biomedical ontology normalization maps free-text expressions to standardized concepts, enabling consistent integration and analysis of biomedical data. This task remains challenging because lexical variation and subtle distinctions among hierarchically related concepts can obscure concept boundaries. We present OntologyAligner, a three-stage framework that combines ontology-aligned retrieval, large language model candidate reranking, and selective hierarchy-guided refinement. We also construct PhenoNormBench, a unified benchmark comprising 13,390 samples from seven Human Phenotype Ontology datasets. OntologyAligner achieved state-of-the-art performance on HPO normalization, with 88.78% Macro Top-1 Accuracy and 86.75% Micro Top-1 Accuracy, exceeding the strongest baseline by 4.85 and 5.07 percentage points, respectively. Ablation analyses showed complementary contributions from all three stages, and sensitivity analyses demonstrated stability across candidate-set sizes and model backbones. Applications to MONDO, MEDIC, and NCBITaxon further established portability to other ontologies. OntologyAligner offers a generalizable framework for accurate mapping of biomedical text to structured ontology concepts. PhenoNormBench and the code are publicly available at https://github.com/zhelishisongjie/OntologyAligner.
Abstract:Biomedical ontology normalization maps free-text expressions to standardized concepts, enabling consistent integration and analysis of biomedical data. This task remains challenging because lexical variation and subtle distinctions among hierarchically related concepts can obscure concept boundaries. We present OntologyAligner, a three-stage framework that combines ontology-aligned retrieval, large language model candidate reranking, and selective hierarchy-guided refinement. We also construct PhenoNormBench, a unified benchmark comprising 13,390 samples from seven Human Phenotype Ontology datasets. OntologyAligner achieved state-of-the-art performance on HPO normalization, with 88.78% Macro Top-1 Accuracy and 86.75% Micro Top-1 Accuracy, exceeding the strongest baseline by 4.85 and 5.07 percentage points, respectively. Ablation analyses showed complementary contributions from all three stages, and sensitivity analyses demonstrated stability across candidate-set sizes and model backbones. Applications to MONDO, MEDIC, and NCBITaxon further established portability to other ontologies. OntologyAligner offers a generalizable framework for accurate mapping of biomedical text to structured ontology concepts. PhenoNormBench and the code are publicly available at this https URL.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2609.10055 [cs.AI]
(or arXiv:2609.10055v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.10055 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 254 words
Related coverage
Angles of attack for continual learning safety
LessWrong · Jun 16, 2026
Toward Culturally Aligned LLMs through Ontology-Guided Multi-Agent Reasoning
arXiv · Jun 6, 2026
An Ontology-Guided, Deduplication-Aware Extraction Layer for Knowledge Graph Construction from Heterogeneous Documents
arXiv · Aug 3, 2026
Operationalizing Genie Ontology in Your Data Stack
Databricks · Sep 1, 2026