Foundation model representations from images and transcriptomics carry complementary signals for cancer classification; multimodal fusion improves results mainly when no modality dominates, and conformal prediction recovers true labels in most failed point predictions on out-of-distribution data.
arXiv preprint arXiv:2603.27460 (2026)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Probing, Fusion, and Trustworthiness: A Systematic Evaluation of Foundation Model Representations for Multimodal Cancer Analysis
Foundation model representations from images and transcriptomics carry complementary signals for cancer classification; multimodal fusion improves results mainly when no modality dominates, and conformal prediction recovers true labels in most failed point predictions on out-of-distribution data.