# CRNDiff: Count-Native Diffusion Framework via Chemical Reaction Networks

arXiv: https://arxiv.org/abs/2609.31149 · HTML: https://mlsanity.com/paper/2609.31149/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Flow matching, time series and science, one of 14 directions in Diffusion models.
- Builds on 16 of its 37 references that are in the map.
- First posted on arXiv 25 Sep 2026.

## Placed in

- Diffusion models / [Flow matching, time series and science](https://mlsanity.com/diffusion/flow-matching-time-series-and-science/)

## Builds on (most cited first, 16 of 16)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Structured Denoising Diffusion Models in Discrete State-Spaces](https://mlsanity.com/paper/2107.03006/) (2021)
- [Simple and Effective Masked Diffusion Language Models](https://mlsanity.com/paper/2406.07524/) (2024)
- [Simplified and Generalized Masked Diffusion for Discrete Data](https://mlsanity.com/paper/2406.04329/) (2024)
- [A Continuous Time Framework for Discrete Denoising Models](https://mlsanity.com/paper/2205.14987/) (2022)
- [Practical and Asymptotically Exact Conditional Sampling in Diffusion Models](https://mlsanity.com/paper/2306.17775/) (2023)
- [A General Framework for Inference-time Scaling and Steering of Diffusion Models](https://mlsanity.com/paper/2501.06848/) (2025)
- [Derivative-Free Guidance in Continuous and Discrete Diffusion Models with Soft Value-Based Decoding](https://mlsanity.com/paper/2408.08252/) (2024)
- [Generator Matching: Generative modeling with arbitrary Markov processes](https://mlsanity.com/paper/2410.20587/) (2024)
- [Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces](https://mlsanity.com/paper/2305.11089/) (2023)
- [Learning to Jump: Thinning and Thickening Latent Counts for Generative Modeling](https://mlsanity.com/paper/2305.18375/) (2023)
- [Multi-Modal and Multi-Attribute Generation of Single Cells with CFGen](https://mlsanity.com/paper/2407.11734/) (2024)
- [Count Bridges enable Modeling and Deconvolving Transcriptomic Data](https://mlsanity.com/paper/2603.04730/) (2026)
- [Flow Matching for Count Data](https://mlsanity.com/paper/2605.07746/) (2026)
- [CountsDiff: A Diffusion Model on the Natural Numbers for Generation and Imputation of Count-Based Data](https://mlsanity.com/paper/2604.03779/) (2026)

## Abstract (arXiv)

Scientific measurements such as single-cell RNA (scRNA) sequencing often take the form of nonnegative integer counts, whereas continuous-state diffusion models approximate this discrete structure using continuous coordinates. Building on stochastic chemical reaction networks (CRNs), a class of count-native Markov jump processes, we introduce CRNDiff, a structured framework that combines count-space diffusion with inference-time conditioning on rare subpopulations. An independent birth--death instantiation yields a closed-form transition kernel for forward noising. This kernel enables reverse sampling via forward-filtering backward-sampling (FFBS) and supports data-driven selection of the terminal noising time, eliminating the need for a validation sweep. This tractability also lets us introduce tilted Feynman--Kac (FK) steering, a method for sampling target subpopulations from a frozen generator without retraining. By tilting posterior marginals before FK particle correction, steering mitigates importance-weight concentration when the target population is rare. Using scRNA-seq data from the human heart cell atlas, we test the ability of CRNDiff to generate cell-type-specific distributions. Across the three evaluated target populations, CRNDiff achieves the highest conditional fidelity among the evaluated generative models, with larger mean purity margins for rarer target populations. Generated cells preserve marker-level differential-expression structure. Replacing real training cells for the target classes with generated cells yields downstream classification performance approaching that of the real-data reference.