“…No intermediate representations are available so researchers can straight-forwardly use them to develop or evaluate popular tasks in NLG pipelines (Reiter and Dale, 2000), such as Discourse Ordering, Lexicalization, Aggregation, Referring Expression Generation, among others. Moreover, these new corpora, like many other resources in Computational Linguistics more in general, are only available in English, limiting the development of NLGapplications to other languages, which is currently an emerging theme in NLG research community -see, for instance, the increased availability of SimpleNLG tools to languages other than English (Mazzei et al, 2016;Bollmann, 2011;Vaudry and Lapalme, 2013;Ramos-Soto et al, 2017) and the recent Multilingual Surface Realization task (Mille et al, 2018).…”