... with 1600 reju-venation steps to the 16-particle learner with 100 re-juvenation steps, for both models the former outper-forms the latter in both log-probability and token f-score. This suggests ... -2 39.22 50.14 -2 62.34PF100064.11 -2 34.87 57.88 -2 54.17PF1,10063.17 -2 45.32 66.88 -2 57.65PF16,10068.05 -2 35.71 70.05 -2 51.66PF1,160077.06 -2 28.79 74.47 -2 49.78Table 1: Results for both ... section. For the batch samplers, we use simu-lated annealing to facilitate the finding of high prob-ability solutions, and for the particle filters, we com-pare the performance of a ‘degenerate’ 1-particlelearner...