Physics-guided residual correction of $\alpha$-decay half-lives based on the effective liquid drop model
To improve the prediction accuracy of $\alpha$-decay half-lives in heavy and superheavy nuclei, a physics-guided residual-correction framework combining the effective liquid drop model (ELDM) with machine-learning methods is proposed. The ELDM is first used as the macroscopic baseline for describing the barrier-penetration process, and XGBoost and TabPFN models are then employed to learn the residual deviations between ELDM predictions and experimental data. To incorporate microscopic nuclear-structure information, several physically motivated descriptors are constructed, including deformation-related quantities, Geiger--Nuttall-related features, and minimum orbital angular momentum. The results show that machine-learning residual correction significantly improves the predictive performance of the ELDM baseline. Among all models, TabPFN-term3 achieves the best accuracy, reducing the RMSE and MAE to 0.348 and 0.248, corresponding to improvements of 38.60\% and 40.46\%, respectively. Residual-distribution and feature-ablation analyses further indicate that the corrected predictions are closer to experimental values and that physically motivated descriptors play an important role in learning nonlinear residual structures. Overall, the proposed ELDM-based residual-correction framework can effectively compensate for missing microscopic nuclear-structure effects while preserving physical interpretability, providing a feasible strategy for high-precision $\alpha$-decay half-life prediction.