OntologyAligner: Ontology-Aligned Retrieval and Hierarchy-Guided Large Language Model Reranking for Biomedical Ontology Normalization
Summary
OntologyAligner presents a three-stage framework combining ontology-aligned retrieval and LLM reranking for biomedical ontology normalization, achieving state-of-the-art performance on HPO tasks and introducing the PhenoNormBench benchmark.
View Cached Full Text
Cached at: 09/11/26, 08:43 AM
# OntologyAligner: Ontology-Aligned Retrieval and Hierarchy-Guided Large Language Model Reranking for Biomedical Ontology Normalization Source: [https://arxiv.org/abs/2609.10055](https://arxiv.org/abs/2609.10055) Authors:[Jie Song](https://arxiv.org/search/cs?searchtype=author&query=Song,+J),[Zhichuan Xu](https://arxiv.org/search/cs?searchtype=author&query=Xu,+Z),[Ziyu Lu](https://arxiv.org/search/cs?searchtype=author&query=Lu,+Z),[Meng Xiao](https://arxiv.org/search/cs?searchtype=author&query=Xiao,+M),[Cheng Bi](https://arxiv.org/search/cs?searchtype=author&query=Bi,+C),[Yuxin Zhang](https://arxiv.org/search/cs?searchtype=author&query=Zhang,+Y),[Xin Zheng](https://arxiv.org/search/cs?searchtype=author&query=Zheng,+X),[Xiaoran Li](https://arxiv.org/search/cs?searchtype=author&query=Li,+X),[Qiongfang Cao](https://arxiv.org/search/cs?searchtype=author&query=Cao,+Q),[Hao Yang](https://arxiv.org/search/cs?searchtype=author&query=Yang,+H),[Bairong Shen](https://arxiv.org/search/cs?searchtype=author&query=Shen,+B) [View PDF](https://arxiv.org/pdf/2609.10055) > 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](https://github.com/zhelishisongjie/OntologyAligner)\. ## Submission history From: Jie Song \[[view email](https://arxiv.org/show-email/0a48e8ca/2609.10055)\] **\[v1\]**Wed, 9 Sep 2026 11:30:25 UTC \(1,127 KB\)
Similar Articles
Learning to Fuse LLMs with Ontology Rankers for Rare-Disease Diagnosis
This paper presents a fusion model that integrates large language models (LLMs) with ontology rankers to improve rare-disease diagnosis, achieving higher recall while preserving structured evidence for clinical decision support.
Topology-Enhanced Alignment for Large Language Models: Trajectory Topology Loss and Topological Preference Optimization
This paper introduces a topology-enhanced alignment framework for LLMs, utilizing trajectory topology loss and topological preference optimization based on persistent homology to regularize semantic trajectories in hidden space.
BeLink: Biomedical Entity Linking Meets Generative Re-Ranking
BeLink introduces a set-wise instruction-tuning formulation for generative re-ranking in biomedical entity linking, achieving 3-24% accuracy improvements and faster inference compared to state-of-the-art systems.
Fully Automated Identification of Lexical Alignment and Preference-Stage Shifts in Large Language Models
This paper introduces two automated metrics, Lexical Alignment Score and Triangulated Preference Shift, to identify lexical overuse in LLMs and attribute it to preference learning stages. The method is tested on six model families using PubMed abstracts, replicating prior findings without manual intervention.
Hypergraph as Language
This paper proposes Hyper-Align, a framework that serializes hypergraph structures into tokens via HIDT-O and HIP, enabling LLMs to process high-order relationships, and introduces HyperAlign-Bench for evaluation.