ymcui/cmrc2018
A Chinese span-extraction reading comprehension dataset for training and evaluating question-answering models.
Not currently ranked — collecting fresh signals.
star history
CMRC 2018 is a benchmark dataset released at EMNLP-IJCNLP 2019 for Chinese machine reading comprehension, specifically designed for span-extraction question answering tasks. The repository provides training, dev, and test data along with submission guidelines through CodaLab and a Hugging Face datasets integration. It includes a public leaderboard tracking state-of-the-art systems on this benchmark.
Frequently asked
- What is ymcui/cmrc2018?
- A Chinese span-extraction reading comprehension dataset for training and evaluating question-answering models.
- Is cmrc2018 open source?
- Yes — ymcui/cmrc2018 is open source, released under the CC-BY-SA-4.0 license.
- What language is cmrc2018 written in?
- ymcui/cmrc2018 is primarily written in Python.
- How popular is cmrc2018?
- ymcui/cmrc2018 has 456 stars on GitHub.
- Where can I find cmrc2018?
- ymcui/cmrc2018 is on GitHub at https://github.com/ymcui/cmrc2018.