发表机构
The University of Hong Kong; Hong Kong Quantum AI Lab Limited（香港大学; 香港量子人工智能实验室有限公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过融合经典树模型与图预测器的简单预测级集成，在多个数据集上实现优于复杂多模态方法的分子性质回归精度，证明化学引导的表示选择与简单融合即可达到高准确性和可迁移性。
AI 中文摘要
分子连续性质的精确预测支持化学发现与设计。许多近期方法在日益复杂的神经架构中组合多种分子表示，旨在实现更高精度。然而，经典机器学习模型在基于化学意义的分子表示训练时仍具有竞争力。因此，我们基于Morgan指纹、RDKit分子描述符以及构象衍生的几何特征构建了一个化学引导的经典分支，这些特征被选择以提供分子结构的互补视角。我们将基于这些表示训练的四个树学习器与CoMPT图预测器相结合，并使用Ridge回归集成其折外预测。在ESOL、FreeSolv和Lipophilicity数据集上，融合模型实现了比Chemprop更低的平均均方根误差（RMSE），且优于包括CoMPT、MoleSG、KFLM2和MCMPP在内的竞争性图方法和多模态方法。在单独整理的AqSolDB数据集上重新训练后，相同方法无需重新设计或调整参数，相对于经典集成降低了平均绝对误差（MAE），并实现了比Chemprop更低的平均MAE。这些结果表明，准确且可迁移的分子性质回归不一定需要复杂的联合多模态架构：当化学洞察指导分子表示的选择时，经典学习仍具有竞争力，而图模型可以通过简单的预测级融合补充互补信息。
英文摘要
Accurate prediction of continuous molecular properties supports chemical discovery and design. Many recent approaches combine multiple molecular representations within increasingly complex neural architectures, aiming to achieve higher accuracy. However, classical machine-learning models remain competitive when trained on chemically meaningful molecular representations. We therefore constructed a chemistry-guided classical branch from Morgan fingerprints, RDKit molecular descriptors, and conformer-derived geometric features selected to provide complementary views of molecular structure. We combined four tree learners trained on these representations with a CoMPT graph predictor and integrated their out-of-fold predictions using Ridge regression. Across ESOL, FreeSolv, and Lipophilicity datasets, the fused model achieved lower mean root-mean-square error (RMSE) than Chemprop and competitive graph and multimodal methods, including CoMPT, MoleSG, KFLM2, and MCMPP. After retraining on the separately curated AqSolDB dataset, the same approach, without redesign or retuning, reduced mean absolute error (MAE) relative to the classical ensemble and achieved lower mean MAE than Chemprop. These results show that accurate and transferable molecular property regression does not necessarily require a complex joint multimodal architecture: classical learning remains competitive when chemical insight guides the choice of molecular representations, and a graph model can add complementary information through simple prediction-level fusion.
