Prompting Beyond Pairs: Decoupled Semantic Supervision for Knowledge-Guided Multiplex Virtual Staining
- 1Department of Physiology, Yong Loo Lin School of Medicine, National University of Singapore; Mechanobiology Institute, National University of Singapore
Abstract
Virtual staining provides a non-invasive alternative to fluorescence microscopy, yet existing deep learning approaches fundamentally rely on pixel-aligned, multiplexed fluorescence targets for supervision. This dependence on rigidly paired data limits scalability, constrains flexibility in generating diverse subcellular structures, and becomes impractical in data-scarce biological settings. In this work, we introduce a semantic supervision paradigm for virtual staining, demonstrating that domain-knowledge prompts can effectively replace conventional pixel-level supervision. Unlike existing methods constrained by rigidly paired multiplex targets, our framework leverages biological prompts to decouple structural guidance from image translation. This decoupling enables high-fidelity, independent synthesis of multiple subcellular structures using only single-channel data. To ensure high-fidelity generation under weak supervision, we integrate self-supervised representation learning to mitigate data scarcity and incorporate direct preference optimization to suppress structural artifacts. Evaluations on the JUMP benchmark demonstrate that our approach effectively balances flexibility and fidelity, outperforming supervised baselines with a 43.3 % reduction in Average FID and an Average PCC of 0.912, while exhibiting high robustness in channel-deficient scenarios. Furthermore, the model generalizes across four in-house datasets to successfully multiplex six subcellular structures, overcoming the physical constraints of conventional fluorescent staining.
Related
- Relational Knowledge Distillation Brings DNN Representations Close Enough to Humans to Be Aligned Without Supervision
- A trustworthy semantic knowledge network linking data-driven cell phenotypes to diseases and drugs
- Validated Semantic Scoring Improves Filtering Generic Concepts from Knowledge-Graphs for Drug Repurposing Queries
- Evaluating in-context learning with prompting regimes for transformer-based synthetic health tabular data generation
- Abdominal Magnetic Resonance Imaging (MRI) Findings Prompting the Diagnosis of Paroxysmal Nocturnal Hemoglobinuria: A Case Report.
BibTeX
@misc{hu2026prompting,
title = {{Prompting Beyond Pairs: Decoupled Semantic Supervision for Knowledge-Guided Multiplex Virtual Staining}},
author = {Y. Hu and J. Wang and K. Zheng and Y. Jin and H. Yu},
howpublished = {bioRxiv},
year = {2026},
month = jul,
doi = {10.64898/2026.07.31.741995},
url = {https://www.biorxiv.org/content/10.64898/2026.07.31.741995v3},
note = {CC BY},
}
RIS
TY - UNPB
TI - Prompting Beyond Pairs: Decoupled Semantic Supervision for Knowledge-Guided Multiplex Virtual Staining
AU - Y. Hu
AU - J. Wang
AU - K. Zheng
AU - Y. Jin
AU - H. Yu
PY - 2026
DA - 2026/07/31
PB - bioRxiv
DO - 10.64898/2026.07.31.741995
UR - https://www.biorxiv.org/content/10.64898/2026.07.31.741995v3
AB - Virtual staining provides a non-invasive alternative to fluorescence microscopy, yet existing deep learning approaches fundamentally rely on pixel-aligned, multiplexed fluorescence targets for supervision. This dependence on rigidly paired data limits scalability, constrains flexibility in generating diverse subcellular structures, and becomes impractical in data-scarce biological settings. In this work, we introduce a semantic supervision paradigm for virtual staining, demonstrating that domain-knowledge prompts can effectively replace conventional pixel-level supervision. Unlike existing methods constrained by rigidly paired multiplex targets, our framework leverages biological prompts to decouple structural guidance from image translation. This decoupling enables high-fidelity, independent synthesis of multiple subcellular structures using only single-channel data. To ensure high-fidelity generation under weak supervision, we integrate self-supervised representation learning to mitigate data scarcity and incorporate direct preference optimization to suppress structural artifacts. Evaluations on the JUMP benchmark demonstrate that our approach effectively balances flexibility and fidelity, outperforming supervised baselines with a 43.3 % reduction in Average FID and an Average PCC of 0.912, while exhibiting high robustness in channel-deficient scenarios. Furthermore, the model generalizes across four in-house datasets to successfully multiplex six subcellular structures, overcoming the physical constraints of conventional fluorescent staining.
ER -