metadata

annotations_creators: []
language_creators: []
language:
  - cz
  - de
  - it
license: []
multilinguality:
  - translation
pretty_name: merlin
size_categories:
  - unknown
source_datasets: []
task_categories:
  - conditional-text-generation
task_ids:
  - machine-translation

MERLIN corpus

Project URL: https://merlin-platform.eu/C_mcorpus.php

Dataset URL: https://clarin.eurac.edu/repository/xmlui/handle/20.500.12124/6

The MERLIN corpus is a written learner corpus for Czech, German, and Italian that has been designed to illustrate the Common European Framework of Reference for Languages (CEFR) with authentic learner data. The corpus contains learner texts produced in standardized language certifications covering CEFR levels A1-C1. The MERLIN annotation scheme includes a wide range of language characteristics that provide researchers with concrete examples of learner performance and progress across multiple proficiency levels.