A new benchmarking framework shows virtual cell models overestimate performance on standard tests, drop sharply on unseen contexts and perturbations, and produce inconsistent rankings across metrics.
Large-scale foundation model on single-cell tran- scriptomics.Nature Methods, 21(8):1481–1491
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
q-bio.CB 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Benchmarking virtual cell models for in-the-wild perturbation response
A new benchmarking framework shows virtual cell models overestimate performance on standard tests, drop sharply on unseen contexts and perturbations, and produce inconsistent rankings across metrics.