... the case of similar train and test corpora, wherethe mean of the distribution (8.50) is close to the median(8.0), but not in the case of divergent corpora, where themean (6.04) and median (1.0) ... range of techniques, from interpolation at either the countlevel or the model level (Bacchiani and Roark, 2003;Bacchiani et al., 2006) to using explicit models of syntax or semantics. Hsu and ... tune, and test on threedisjoint corpora. We consider two different train-ing sets: one of 110M words of NYT from 2000and 2001 (NYT00+01), and one of 110M words of AFP from 2002, 2005, and 2006...