Research

Paper

TESTING February 23, 2026

PackFlow: Generative Molecular Crystal Structure Prediction via Reinforcement Learning Alignment

Authors

Akshay Subramanian, Elton Pan, Juno Nam, Maurice Weiler, Shuhui Qu, Cheol Woo Park, Tommi S. Jaakkola, Elsa Olivetti, Rafael Gomez-Bombarelli

Abstract

Organic molecular crystals underpin technologies ranging from pharmaceuticals to organic electronics, yet predicting solid-state packing of molecules remains challenging because candidate generation is combinatorial and stability is only resolved after costly energy evaluations. Here we introduce PackFlow, a flow matching framework for molecular crystal structure prediction (CSP) that generates heavy-atom crystal proposals by jointly sampling Cartesian coordinates and unit-cell lattice parameters given a molecular graph. This lattice-aware generation interfaces directly with downstream relaxation and lattice-energy ranking, positioning PackFlow as a scalable proposal engine within standard CSP pipelines. To explicitly steer generation toward physically favourable regions, we propose physics alignment, a reinforcement learning post-training stage that uses machine-learned interatomic potential energies and forces as stability proxies. Physics alignment improves physical validity without altering inference-time sampling. We validate PackFlow's performance against heuristic baselines through two distinct evaluations. First, on a broad unseen set of molecular systems, we demonstrate superior candidate generation capability, with proposals exhibiting greater structural similarity to experimental polymorphs. Second, we assess the full end-to-end workflow on two unseen CSP blind-test case studies, including relaxation and lattice-energy analysis. In both settings, PackFlow outperforms heuristics-based methods by concentrating probability mass in low-energy basins, yielding candidates that relax into lower-energy minima and offering a practical route to amortize the relax-and-rank bottleneck.

Metadata

arXiv ID: 2602.20140
Provider: ARXIV
Primary Category: physics.chem-ph
Published: 2026-02-23
Fetched: 2026-02-24 04:38

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.20140v1</id>\n    <title>PackFlow: Generative Molecular Crystal Structure Prediction via Reinforcement Learning Alignment</title>\n    <updated>2026-02-23T18:52:13Z</updated>\n    <link href='https://arxiv.org/abs/2602.20140v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.20140v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Organic molecular crystals underpin technologies ranging from pharmaceuticals to organic electronics, yet predicting solid-state packing of molecules remains challenging because candidate generation is combinatorial and stability is only resolved after costly energy evaluations. Here we introduce PackFlow, a flow matching framework for molecular crystal structure prediction (CSP) that generates heavy-atom crystal proposals by jointly sampling Cartesian coordinates and unit-cell lattice parameters given a molecular graph. This lattice-aware generation interfaces directly with downstream relaxation and lattice-energy ranking, positioning PackFlow as a scalable proposal engine within standard CSP pipelines. To explicitly steer generation toward physically favourable regions, we propose physics alignment, a reinforcement learning post-training stage that uses machine-learned interatomic potential energies and forces as stability proxies. Physics alignment improves physical validity without altering inference-time sampling. We validate PackFlow's performance against heuristic baselines through two distinct evaluations. First, on a broad unseen set of molecular systems, we demonstrate superior candidate generation capability, with proposals exhibiting greater structural similarity to experimental polymorphs. Second, we assess the full end-to-end workflow on two unseen CSP blind-test case studies, including relaxation and lattice-energy analysis. In both settings, PackFlow outperforms heuristics-based methods by concentrating probability mass in low-energy basins, yielding candidates that relax into lower-energy minima and offering a practical route to amortize the relax-and-rank bottleneck.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='physics.chem-ph'/>\n    <published>2026-02-23T18:52:13Z</published>\n    <arxiv:primary_category term='physics.chem-ph'/>\n    <author>\n      <name>Akshay Subramanian</name>\n    </author>\n    <author>\n      <name>Elton Pan</name>\n    </author>\n    <author>\n      <name>Juno Nam</name>\n    </author>\n    <author>\n      <name>Maurice Weiler</name>\n    </author>\n    <author>\n      <name>Shuhui Qu</name>\n    </author>\n    <author>\n      <name>Cheol Woo Park</name>\n    </author>\n    <author>\n      <name>Tommi S. Jaakkola</name>\n    </author>\n    <author>\n      <name>Elsa Olivetti</name>\n    </author>\n    <author>\n      <name>Rafael Gomez-Bombarelli</name>\n    </author>\n  </entry>"
}