Our reasons are twofold: first, collecting a sufficiently large number of individual headlines for training purposes may be impractical; second, as future work we intend to reuse the same models in the generation of newspapers sentences in general, and not only headlines.
Chomsky argued (on technical grounds; see Chomsky 1957, ch.1) that grammars must be enriched with a second type of rule, known as 'transformations.' Unlike phrase structure rules, transformations operate on whole sentences (or more strictly, their phrasemarkers); they allow for the generation of new sentences (/phrasemarkers) out of old ones.
The first generation of IPP sentences allowed almost no discretion to judges and resulted in an enormous increase in the number of people serving indefinite sentences.
However, the work in [6] was still limited in the sense that it was not applied to the generation of whole sentences taken from a corpus of actual language use, and by using large, computationally expensive language models that took gender and number factors into account, and which we have presently improved on.3.
They focus on generation of discrete sentence-level utterances.
FLN, on the other hand, is distinctive to humans and what is special about it is its power of recursion, that is, its ability to categorize linguistic objects into hierarchically organized classes, and (on the behavioral side) for the generation of infinitely many sentences out of finitely many words.
Furthermore, a positive shift in the memory of sentences and generation of morphological rules was observed.
In other words, posing problems in Monsakun is the division of the task to pose an arithmetic word problem into two sub-tasks: generation and integration of three sentences satisfying the required constraints and the replacement of the generation (sub-) tasks by selecting tasks of sentences.
When the military suppressed the uprising, killing thousands of protesters, it rounded up an entire generation of student leaders and sentenced them to long jail terms.
