AGREE is a dataset and task for evaluation of language models based on grammar agreement in Czech. The dataset consists of sentences with marked suffixes of past tense verbs. The task is to choose the right verb suffix which depends on gender, number and animacy of subject. It is challenging for language models because 1) Czech is morphologically rich, 2) it has relatively free word order, 3) high out-of-vocabulary (OOV) ratio, 4) predicate and subject can be far from each other, 5) subjects can be unexpressed and 6) various semantic rules may apply. The task provides a straightforward and easily reproducible way of evaluating language models on a morphologically rich language.
THE LINDAT/CLARIN PROJECT (LM2015071 and CZ.02.1.01/0.0/0.0/16_013/0001781; formerly LM2010013) IS FULLY SUPPORTED BY THE MINISTRY OF EDUCATION, SPORTS AND YOUTH OF THE CZECH REPUBLIC UNDER THE PROGRAMME LM OF "LARGE INFRASTRUCTURES".
Copyright (c) 2018 UFAL MFF UK. All rights reserved.