The selected rotation angle and deconvolution time window during S-wave receiver function (SRF) calculations and the final SRF quality control may introduce artificial interference. Here we overcome these problems by proposing a new strategy, the GC_SRF strategy, which employs grid search and correlation analysis to obtain reliable SRFs. Extensive tests using synthetic and real data suggest that the GC_SRF strategy is a suitable approach for constraining lithospheric thickness, and reproducible and robust depth estimations of the lithosphere–asthenosphere boundary (LAB) are obtained. Specifically, this GC_SRF strategy can restore the weak Sp phases from full wavefield synthetic seismograms. Similar, but clearer, discontinuity patterns that contain no artificial interference compared with those obtained in previous studies of southeastern Tibet are produced here. The post-stack migrated SRFs reveal that the Chuandian region has a thicker crust and either a poorly defined or unclear LAB, whereas a thinner crust and clear LAB is present at ~160 km depth beneath the Sichuan Basin. The absence of a continuous LAB in the Chuandian region may suggest lithospheric regrowth due to mantle plume processes.