Building virtual cells with generative models to simulate cellular behavior in silico is emerging as a promising paradigm for accelerating drug discovery. However, prior image-based generative approaches can produce implausible cell images that violate basic physical and biological constraints. To address this, we propose to post-train virtual cell models with reinforcement learning (RL), leveraging biologically meaningful evaluators as reward functions. We design seven rewards spanning three categories-biological function, structural validity, and morphological correctness-and optimize the state-of-the-art CellFlux model to yield CellFluxRL. CellFluxRL consistently improves over CellFlux across all rewards, with further performance boosts from test-time scaling. Overall, our results present a virtual cell modeling framework that enforces physically-based constraints through RL, advancing beyond "visually realistic" generations towards "biologically meaningful" ones.
@article{arxiv.2603.21743,
title = {CellFluxRL: Biologically-Constrained Virtual Cell Modeling via Reinforcement Learning},
author = {Dongxia Wu and Shiye Su and Yuhui Zhang and Elaine Sui and Emma Lundberg and Emily B. Fox and Serena Yeung-Levy},
journal= {arXiv preprint arXiv:2603.21743},
year = {2026}
}