MolPaQ assembles valid molecules from quantum latent patches via a pretrained beta-VAE, conditioner, and valence aggregator, reporting 100% validity, 99.75% novelty, and modest gains in QED and aromaticity over classical baselines.
While this framework excelled in likelihood-based training and supported validity, it was se- quential and non-modular
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generation
MolPaQ assembles valid molecules from quantum latent patches via a pretrained beta-VAE, conditioner, and valence aggregator, reporting 100% validity, 99.75% novelty, and modest gains in QED and aromaticity over classical baselines.