4 datasets found

The dataset is used to evaluate the proposed semantics-native communication framework. It consists of a set of actions, concepts, and symbols, and is used to test the...

The Penn Treebank dataset contains one million words of 1989 Wall Street Journal material annotated in Treebank II style, with 42k sentences of varying lengths.

The dataset used in the paper is a training dataset for the SPAWN model, consisting of 10000 sentences with reduced relative clauses.

The dataset used in the paper is a comprehension-to-production priming paradigm, where participants or SPAWN models are presented with three primes with the same structure,...