JSD of the SIXSA posteriors corrected by weighted importance sampling with respect to the BXA posteriors. Four training sample sizes are considered for the NDE: 500, 2500, 5,000 and 25,000 respectively. The horizontal dashed lines indicate the limit under which the posterior distributions can be considered as identical to the BXA ones (Dax et al. 2023). The JSD with the posteriors computed by jaxspec is also shown (a burn in of 1,000 and a chain length of 10,000).