cs.LGOct 7, 2025

ESS-Flow: Training-free guidance as Bayesian inference in source space

Authors: Adhithyan Kalaivanan, Zheng Zhao, Jens Sjölund, Fredrik Lindsten

Organizations: Department of Computer Science Linköping University, Sweden · Department of Information Technology Uppsala University, Sweden

Abstract

Guiding pretrained flow-based generative models for conditional generation or to produce samples with desired target properties enables solving diverse tasks without retraining on paired data. We present ESS-Flow, a gradient-free method that leverages the Gaussian prior of the source distribution in flow-based models to perform Bayesian inference directly in the source space using elliptical slice sampling. ESS-Flow only requires forward passes through the generative model and observation process, making it applicable even when gradients are unavailable, such as with non-differentiable simulation-based observations or quantization in the generation process. We demonstrate its effectiveness on scientific applications including generating materials with target properties, protein backbone structures with improved designability, and synthesizable, drug-like molecules.

Figures & tables

Appendix figures & tables10 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Flow-Direct: Feedback-Efficient and Reusable Guidance for Flow Models via Non-Parametric Guidance Field

    May 8, 2026Kim Yong Tan, Yueming Lyu, Ivor Tsang +1Diffusion Model GuidanceBlack-Box Optimization

  2. Conservative Flows: A New Paradigm of Generative Models

    May 7, 2026Eshed Gal, Md Shahriar Rahim Siddiqui, Moshe Eliasof +1Langevin DynamicsGenerative Flow Networks