BioHiCL: Hierarchical Multi-Label Contrastive Learning for Biomedical Retrieval with MeSH Labels

2026-04-16

AI Summary

This work addresses the limitations of existing biomedical retrieval models, which rely on coarse-grained binary relevance signals and struggle to capture fine-grained semantic overlap and hierarchical relationships among texts. To overcome this, the study introduces hierarchical multi-label contrastive learning into biomedical retrieval for the first time, leveraging the tree-structured MeSH ontology to construct structured supervision signals. The authors propose BioHiCL (Base/Large), a lightweight generative retrieval model that effectively encodes both semantic and hierarchical dependencies among labels. Evaluated across biomedical retrieval, sentence similarity, and question answering tasks, BioHiCL significantly outperforms current state-of-the-art methods while maintaining efficient inference performance.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Image and Video Retrieval

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Effective biomedical information retrieval requires modeling domain semantics and hierarchical relationships among biomedical texts. Existing biomedical generative retrievers build on coarse binary relevance signals, limiting their ability to capture semantic overlap. We propose BioHiCL (Biomedical Retrieval with Hierarchical Multi-Label Contrastive Learning), which leverages hierarchical MeSH annotations to provide structured supervision for multi-label contrastive learning. Our models, BioHiCL-Base (0.1B) and BioHiCL-Large (0.3B), achieve promising performance on biomedical retrieval, sentence similarity, and question answering tasks, while remaining computationally efficient for deployment.

Problem

• biomedical retrieval
• hierarchical relationships
• semantic overlap
• MeSH labels
• multi-label learning

Innovation

• Hierarchical Multi-Label Contrastive Learning
• MeSH Labels
• Biomedical Retrieval
• Structured Supervision
• Semantic Overlap