Category-free complement selection in causal adjunct phrases

Improving the complement/adjunct distinction in CCGbank

One of the challenges of adapting the Penn Treebank for a specific formalism is that the target annotation often requires information represented imperfectly or not at all in the original corpus. When this occurs, the information must either be guessed with heuristics, or annotated manually. Recently, a third option has become available, due to the release of resources that supplement the Penn ...

Knowledge Selection in Category Learning

Inferring Unobserved Category Features With Causal Knowledge

One central function of categories is to allow people to infer the presence of features that cannot be directly observed. Although the effect of observing past category members on such inferences has been considered, the effect of theoretical or causal knowledge about the category has not. We compared the effects of causal laws on feature prediction with the effects of the inter-feature correla...

Category-Driven Content Selection

In this paper, we introduce a content selection method where the communicative goal is to describe entities of different categories (e.g., astronauts, universities or monuments). We argue that this method provides an interesting basis both for generating descriptions of entities and for semi-automatically constructing a benchmark on which to train, test and compare data-to-text generation systems.

Clitics , pronouns , noun phrases and the Head - Complement parameter

There are two levels of rhythmic structure which are assumed to play a role in language acquisition: the basic rhythmic level and the first phrasal level. While the former is supposed to guide infants in language discrimination (Mehler and Nespor 2004), the latter signals the order of words within phrases. In particular, in Nespor et al. (1996), the hypothesis that relative prominence, that is ...