AdsorbFlow: energy-conditioned flow matching enables fast and realistic adsorbate placement

2026-02-22

AI Summary

This work addresses the persistent challenge in computational heterogeneous catalysis of efficiently generating low-energy adsorbate configurations that can relax into the correct energy basins. We propose AdsorbFlow, the first method to incorporate classifier-free guidance with energy conditioning within a conditional flow matching framework. By learning an energy-guided vector field over the rigid-body configuration space of adsorbates and leveraging deterministic ODE integration, AdsorbFlow achieves high-quality sampling in only five steps. Coupled with an EquiformerV2 backbone, the method attains 61.4% SR@10 and 34.1% SR@1 on OC20-Dense, substantially outperforming existing approaches: it reduces sampling steps by 20×, achieves the lowest outlier rate (6.8%), and maintains strong generalization with 58.0% SR@10 on out-of-distribution systems.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Diffusion Models for Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Identifying low-energy adsorption geometries on catalytic surfaces is a practical bottleneck for computational heterogeneous catalysis: the difficulty lies not only in the cost of density functional theory (DFT) but in proposing initial placements that relax into the correct energy basins. Conditional denoising diffusion has improved success rates, yet requires $\sim$100 iterative steps per sample. Here we introduce AdsorbFlow, a deterministic generative model that learns an energy-conditioned vector field on the rigid-body configuration space of adsorbate translation and rotation via conditional flow matching. Energy information enters through classifier-free guidance conditioning -- not energy-gradient guidance -- and sampling reduces to integrating an ODE in as few as 5 steps. On OC20-Dense with full DFT single-point verification, AdsorbFlow with an EquiformerV2 backbone achieves 61.4% SR@10 and 34.1% SR@1 -- surpassing AdsorbDiff (31.8% SR@1, 41.0% SR@10) at every evaluation level and AdsorbML (47.7% SR@10) -- while using 20 times fewer generative steps and achieving the lowest anomaly rate among generative methods (6.8%). On 50 out-of-distribution systems, AdsorbFlow retains 58.0% SR@10 with a MLFF-to-DFT gap of only 4~percentage points. These results establish that deterministic transport is both faster and more accurate than stochastic denoising for adsorbate placement.

Problem

• adsorbate placement
• low-energy adsorption geometries
• computational heterogeneous catalysis
• initial configuration sampling
• energy basins

Innovation

• conditional flow matching
• energy-conditioned generation
• deterministic transport
• adsorbate placement
• classifier-free guidance