MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915929428852736 |
|---|---|
| author | Naqvi, Syed Rameez Peng, Lu |
| author_facet | Naqvi, Syed Rameez Peng, Lu |
| contents | Molecular generative models must jointly ensure validity, diversity, and property control, yet existing approaches typically trade off among these objectives. We present MOLPAQ, a modular quantum-classical generator that assembles molecules from quantum-generated latent patches. A \b{eta}-VAE pretrained on QM9 learns a chemically aligned latent manifold; a reduced conditioner maps molecular descriptors into this space; and a parameter-efficient quantum patch generator produces entangled node embeddings that a valence-aware aggregator reconstructs into valid molecular graphs. Adversarial fine-tuning with a latent critic and chemistry-shaped reward yields 100\% RDKit validity, 99.75\% novelty, and 0.905 diversity. Beyond aggregate metrics, the pretrained quantum generator, steered by the conditioner, improves mean QED by approx. 2.3\% and increases aromatic motif incidence by approx. 10-12\% relative to a parameter-matched classical generator, highlighting its role as a compact topology-shaping operator. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08575 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generation Naqvi, Syed Rameez Peng, Lu Machine Learning Artificial Intelligence Molecular generative models must jointly ensure validity, diversity, and property control, yet existing approaches typically trade off among these objectives. We present MOLPAQ, a modular quantum-classical generator that assembles molecules from quantum-generated latent patches. A \b{eta}-VAE pretrained on QM9 learns a chemically aligned latent manifold; a reduced conditioner maps molecular descriptors into this space; and a parameter-efficient quantum patch generator produces entangled node embeddings that a valence-aware aggregator reconstructs into valid molecular graphs. Adversarial fine-tuning with a latent critic and chemistry-shaped reward yields 100\% RDKit validity, 99.75\% novelty, and 0.905 diversity. Beyond aggregate metrics, the pretrained quantum generator, steered by the conditioner, improves mean QED by approx. 2.3\% and increases aromatic motif incidence by approx. 10-12\% relative to a parameter-matched classical generator, highlighting its role as a compact topology-shaping operator. |
| title | MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generation |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2604.08575 |