Automated biomedical hypothesis generation with time-aware hypergraph contrastive learning: Automated biomedical hypothesis generation with

Amir Hassan Shariatmadari; Sikun Guo; Nathan C. Sheffield; Aidong Zhang; Kishlay Jha

doi:10.1007/s10115-026-02785-3

Back

Automated biomedical hypothesis generation with time-aware hypergraph contrastive learning: Automated biomedical hypothesis generation with

Journal article

Open access

Peer reviewed

Automated biomedical hypothesis generation with time-aware hypergraph contrastive learning: Automated biomedical hypothesis generation with

Amir Hassan Shariatmadari, Sikun Guo, Nathan C. Sheffield, Aidong Zhang and Kishlay Jha

Knowledge and information systems, Vol.68(1), 157

05/17/2026

DOI: 10.1007/s10115-026-02785-3

PMCID: PMC13180762

PMID: 42157957

Files and links (1)

url

https://doi.org/10.1007/s10115-026-02785-3View

Published (Version of record) Open Access

Abstract

Research in scientific domains now generates more than a million articles annually, overwhelming researchers and hindering discovery. This surge has sparked interest in biomedical hypothesis generation (HG), which aims to uncover implicit patterns among biomedical concepts. Most existing methods focus on pairwise link prediction, overlooking the complex, multi-concept relationships underlying many breakthroughs. We introduce HyHG , a temporal Hy pergraph contrastive learning framework for biomedical H ypothesis G eneration, which redefines hypotheses as hyperedges—sets of co-mentioned concepts in an article. By representing articles as hyperedges and organizing them into a temporal hypergraph, HyHG captures the evolution of scientific ideas over time. A transformer-based architecture learns from historical hyperedge sequences to predict future hyperedges—sets of concepts likely to co-occur in the future literature. To distinguish genuine hypotheses from misleading ones, HyHG employs a time-anchored contrastive loss and hard negative sampling based on minimal edits to real hyperedges. We demonstrate that HyHG achieves state-of-the-art performance on three biomedical datasets. Our code and data are available at: https://github.com/amir-hassan25/Temporal-Hypergraph-Contrastive-Learning.

Computer Science

Database Management

Data Mining and Knowledge Discovery

Information Storage and Retrieval

Information Systems and Communication Service

Information Systems Applications (incl.Internet)

IT in Business

Details

Title: Subtitle: Automated biomedical hypothesis generation with time-aware hypergraph contrastive learning: Automated biomedical hypothesis generation with
Creators: Amir Hassan Shariatmadari - University of Virginia
Sikun Guo - University of Virginia
Nathan C. Sheffield - University of Virginia
Aidong Zhang - University of Virginia
Kishlay Jha - University of Iowa
Resource Type: Journal article
Publication Details: Knowledge and information systems, Vol.68(1), 157
DOI: 10.1007/s10115-026-02785-3
PMID: 42157957
PMCID: PMC13180762
NLM abbreviation: Knowl Inf Syst
ISSN: 0219-1377
eISSN: 0219-3116
Publisher: Springer London
Grant note: IIS-2106913 / National Science Foundation (https://doi.org/10.13039/100000001) R01LM014012-01A1 / National Institutes of Health (https://doi.org/10.13039/100000002)
Language: English
Date published: 05/17/2026
Academic Unit: Electrical and Computer Engineering
Record Identifier: 9985164579502771

Metrics

1 Record Views