Participants are highly encouraged to supplement the dataset by creating their own additional datasets and we expect this will help towards NLU model improvements. 

* To be compliant with the competition rules on external data use, please ensure that any datasets you create, source or use are publicly and freely available to all participants in the competition. You may share your datasets on [this forum thread](https://www.kaggle.com/c/chaii-hindi-and-tamil-question-answering/discussion/264581), in order to ensure it is made available for all participants to use.
* Please submit your data using [this form](https://forms.gle/P4ZdoDzyJ2iatzYo6) for Google to be able to release the dataset at the end of the contest
* Data and baseline model provided by Google as well as data contributed by participants will be released as an open-source dataset at the end of the contest. 
* Please be mindful to not violate the rights of a third party and ensure the data does not contain lewd, obscene, pornographic, racist, sexist, or otherwise inappropriate data to the contest. 
* We highly encourage you to create datasets from Wikipedia text. 
* Participants may find tools such as [cdQA-Annotator](https://github.com/cdqa-suite/cdQA-annotator) to annotate your data for ease and data consistency.