A Contrastive-Learning-Based Pre-Training Framework for Optical Property Prediction of Low-Data Rhodamines with Interpretable Multitask Graph Neural Networks

, respectively, outperforming machine learning, single-task, and no-pre-training GNN baselines. External dataset tests and comparisons with theoretical calculations reveal the superiority of our proposed model. Then, attention-based interpretability identifies chemically meaningful regions, including the conjugated backbone and amino substituents, which is consistent with the known photophysical mechanisms. Finally, we designed three new rhodamine derivatives exhibiting high Stokes shifts, with minimum 9 nm deviation between predicted and experimental values. These findings demonstrate that this framework enables accurate fluorescence property prediction and mechanism-informed molecular design, offering a promising theoretical guide for designing next-generation probes.