Preprint · 2026 · Open access
A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data
arXiv (Cornell University) · Published 27 June 2026 · DOI 10.48550/arxiv.2606.28798
In brief
“A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data” is a 2026 preprint by Chengyuan Liu and 3 co-authors, published in arXiv (Cornell University). OpenAlex counts 0 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 0
- References
- 24
- Authors
- 4
Abstract
The main objective of this paper is to propose a general framework for prediction based on different sources of multimodal data in the healthcare domain. We evaluated whether frozen medical large language model (LLM) representations can serve as a shared embedding space for multimodal primary diagnosis category prediction. We propose a backbone-flexible probing framework that serializes structured electronic health record (EHR) variables and leakage-pruned discharge-note sections into a shared frozen-LLM representation space and recovers primary diagnosis categories with linear probes, without fine-tuning the LLM. On a MIMIC-IV cohort of 13,645 admissions (seven diagnosis categories), we probed inputs of different modalities (Structured-only, Unstructured-only, and Combined) across transformer layers and six frozen backbones, compared against XGBoost and three text baselines (CAML, LAAT, PLM-ICD) under a matched split and context budget, and tested robustness across datasets on MIMIC-III. The Combined terminal-layer probe reached 88.86% top-1 accuracy on MIMIC-IV, significantly outperforming CAML and PLM-ICD and statistically comparable to LAAT; Combined accuracy ranged from 87.69% to 89.25% across six backbones. On MIMIC-III, in-domain Combined probing reached 74.18% accuracy and significantly exceeded all three baselines, and in a transfer-learning test a 2.8M-parameter adapter on frozen MIMIC-IV representations reached 92.51% accuracy, exceeding adapted LAAT (85.54%). Frozen LLM embeddings unify structured and narrative EHR information: the Combined configuration was strongest in every setting, and its representations transferred across datasets through a compact adapter. Multimodal probing of frozen LLM representations provides a practical, backbone-flexible approach for studying EHR modalities and adapting clinical representations across datasets.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Liu, C., Zhang, X., Li, Y., & Chen, G. (2026). A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.28798
MLA 9
Liu, Chengyuan, et al. “A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2606.28798.
Chicago
Liu, Chengyuan, Xinyue Zhang, Yao Li, and Guanting Chen. 2026. “A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.28798.
Harvard
Liu, C. et al. (2026) ‘A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data’, arXiv (Cornell University). doi:10.48550/arxiv.2606.28798.
IEEE
C. Liu, X. Zhang, Y. Li, and G. Chen, “A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2606.28798.
BibTeX
@misc{liu2026general,
title = {{A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data}},
author = {Liu, Chengyuan and Zhang, Xinyue and Li, Yao and Chen, Guanting},
howpublished = {arXiv (Cornell University)},
year = {2026},
doi = {10.48550/arxiv.2606.28798},
url = {https://doi.org/10.48550/arxiv.2606.28798}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- frozen large language models
- electronic health records
- discharge notes
- linear array transducer
- transfer learning
- adapter fine-tuning
- multimodal fusion
- MIMIC-IV
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Liu, C., Zhang, X., Li, Y., & Chen, G. (2026). A General Multimodal Probing Framework for Data Fusion and Prediction with Frozen Large Language Models: Evidence from Multimodal EHR Data. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.28798
How many times has this paper been cited?
0 times, according to OpenAlex on 19 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 arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Chengyuan Liu, Xinyue Zhang, Yao Li, Guanting Chen.
What is the DOI of this paper?
10.48550/arxiv.2606.28798, permanently at https://doi.org/10.48550/arxiv.2606.28798. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 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.