Journal article · 2026 · Open access

Multimodal cell foundation models require synergistic integration beyond alignment

Till Richter, Eric Zimmermann, James Hall, Fabian Joachim Theis, Srivatsan Raghavan, Peter S. Winter, Ava P. Amini, Lorin A. Crawford

Nature Machine Intelligence · Published 9 November 2026 · DOI 10.1038/s42256-026-01328-3

Open PDFView at the publisherCite this paper

In brief

“Multimodal cell foundation models require synergistic integration beyond alignment” is a 2026 journal article by Till Richter and 7 co-authors, published in Nature Machine Intelligence. OpenAlex counts 2 citations as of 18 November 2026, which puts it in the top 5% of works published in 2026 by citations. It is open access under a CC BY licence: the full text is free to read at Nature Machine Intelligence.

Citations
2
Most cited in
2026 (2)
Field-weighted impact
2.53
References
59
Authors
8

Abstract

Integrating measurements across modalities and scales into unified quantitative frameworks of cellular function is a central goal in computational biology. Although powerful unimodal foundation models exist, the scarcity of large-scale paired data limits joint multimodal training. This constraint favours compositional foundation models, which fuse frozen unimodal experts through a learned interface. However, it remains unclear when such fusion provides task-relevant information beyond the strongest unimodal representation rather than merely aggregating redundant signal. Here we introduce Synergistic Information Score, a metric grounded in partial information decomposition that quantifies information gain arising only from cross-modal interactions. Extending theory from self-supervised learning, we show that standard alignment-based objectives with frozen encoders collapse to detect linear redundancies, which constrains their ability to capture nonlinear synergistic states. This distinction is especially important for tasks that connect tissue morphology with gene expression. Benchmarking ten fusion methods on spatial transcriptomics datasets, we demonstrate that tasks dominated by linear redundancies are well served by unimodal baselines, whereas complex niche definitions benefit from synergy-aware integration. Finally, we perform a scaling analysis and show that fine-tuning a dominant unimodal expert is the most sample-efficient strategy for standard tasks. Together, these results suggest that progress in multimodal modelling of cellular state requires shifting from alignment-based objectives to synergy-driven integration that exploits complementary cross-modal signal.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Richter, T., Zimmermann, E., Hall, J., Theis, F. J., Raghavan, S., Winter, P. S., Amini, A. P., & Crawford, L. A. (2026). Multimodal cell foundation models require synergistic integration beyond alignment. Nature Machine Intelligence. https://doi.org/10.1038/s42256-026-01328-3

MLA 9

Richter, Till, et al. “Multimodal cell foundation models require synergistic integration beyond alignment.” Nature Machine Intelligence, 2026. https://doi.org/10.1038/s42256-026-01328-3.

Chicago

Richter, Till, Eric Zimmermann, James Hall, Fabian Joachim Theis, Srivatsan Raghavan, Peter S. Winter, Ava P. Amini, and Lorin A. Crawford. 2026. “Multimodal cell foundation models require synergistic integration beyond alignment.” Nature Machine Intelligence. https://doi.org/10.1038/s42256-026-01328-3.

Harvard

Richter, T. et al. (2026) ‘Multimodal cell foundation models require synergistic integration beyond alignment’, Nature Machine Intelligence. doi:10.1038/s42256-026-01328-3.

IEEE

T. Richter et al., “Multimodal cell foundation models require synergistic integration beyond alignment,” Nature Machine Intelligence, 2026, doi: 10.1038/s42256-026-01328-3.

BibTeX

@article{richter2026multimodal,
  title     = {{Multimodal cell foundation models require synergistic integration beyond alignment}},
  author    = {Richter, Till and Zimmermann, Eric and Hall, James and Theis, Fabian Joachim and Raghavan, Srivatsan and Winter, Peter S. and Amini, Ava P. and Crawford, Lorin A.},
  journal   = {Nature Machine Intelligence},
  year      = {2026},
  doi       = {10.1038/s42256-026-01328-3},
  url       = {https://doi.org/10.1038/s42256-026-01328-3}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics

  • multimodal foundation models
  • partial information decomposition
  • cross-modal fusion
  • spatial transcriptomics
  • sample efficiency
  • compositional models

Work with this paper

Most-cited papers in Biochemistry, Genetics and Molecular Biology

All highly cited Biochemistry, Genetics and Molecular Biology papers

Questions about this paper

How do I cite this paper in APA style?

Richter, T., Zimmermann, E., Hall, J., Theis, F. J., Raghavan, S., Winter, P. S., Amini, A. P., & Crawford, L. A. (2026). Multimodal cell foundation models require synergistic integration beyond alignment. Nature Machine Intelligence. https://doi.org/10.1038/s42256-026-01328-3

How many times has this paper been cited?

2 times, according to OpenAlex on 18 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Nature Machine Intelligence: sign in and open the PDF from this page.

Who wrote this paper?

Till Richter, Eric Zimmermann, James Hall, Fabian Joachim Theis, Srivatsan Raghavan, Peter S. Winter, Ava P. Amini, Lorin A. Crawford.

What is the DOI of this paper?

10.1038/s42256-026-01328-3, permanently at https://doi.org/10.1038/s42256-026-01328-3. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 18 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.