# Cell-JEPA: Latent Representation Learning for Single-Cell Transcriptomics

arXiv: https://arxiv.org/abs/2602.02093 · HTML: https://mlsanity.com/paper/2602.02093/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Feb 2026.

## Built on by (most cited first, 3 of 3)

- [AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling](https://mlsanity.com/paper/2605.05586/) (2026)
- [JEPA-Anything: Learning Predictive Models across Different Worlds](https://mlsanity.com/paper/2609.20800/) (2026)
- [Orthogonal JEPA: Factorized Predictive States for Latent World Models](https://mlsanity.com/paper/2608.20065/) (2026)

## Abstract (arXiv)

Single-cell foundation models learn by reconstructing masked gene expression, implicitly treating technical noise as signal. With dropout rates exceeding 90%, reconstruction objectives encourage models to encode measurement artifacts rather than stable cellular programs. We introduce Cell-JEPA, a joint-embedding predictive architecture that shifts learning from reconstructing sparse counts to predicting in latent space. The key insight is that cell identity is redundantly encoded across genes. We show predicting cell-level embeddings from partial observations forces the model to learn dropout-robust features. On cell-type clustering, Cell-JEPA achieves 0.72 AvgBIO in zero-shot transfer versus 0.53 for scGPT, a 36% relative improvement. On perturbation prediction within a single cell line, Cell-JEPA improves absolute-state reconstruction but not effect-size estimation, suggesting that representation learning and perturbation modeling address complementary aspects of cellular prediction.