Fig. 6

Under the MS model setting as in Figure 5, we fix \(n=500\), \(m=100\), and we apply Naive Padding (left panel) and Centered Padding (right panel) in \((\mathbf{G} ,\mathbf{H} )\in (\mathcal {M}^{10}_{100},\mathcal {M}^{10}_{500})\). We plot the matching accuracy (averaged over 100 Monte Carlo replicates) obtained by M-FAQ (with 10 seeds) versus \(c_b\)