Hi Zindians,
The end is near! The test set will be released at midnight tonight and you have 24 hours to make your final submissions.
The test set will contain questions along with accompanying documents.
The test set follows the below format. You will need to submit the test file with your answers in it.
You can submit your top five 'chunks' retrieved from the source documents. Ordered by similarity score from left (highest) to right (lowest).
We wish you all the best!
The challenge introduction mentions .docx documents also. Is it now .pdf only?
The document formats can include both .docx and .pdf
Will the Document Number be populated or we'll have to populate that?
The document numbers will be provided. These numbers will indicate which documents you are expected to retrieve data from for a given query.
Where is the test set? Where can I find it please?
The test set will be made available on 16 May 2024 at 23:59 PM GMT.
For this challenge, the output size should be limited to 4,000 tokens (max). What does this mean ? The token length of all the output_1+output_2.....output_5 = 4000? or individually for each output we can have max 4000 tokens? Please reply
4000 tokens total (for all outputs combined).
Any anouncement of the winners for this challenge?
In the meantime, you may find these information sources useful
1. https://www.linkedin.com/posts/danielbrumund_its-been-an-absolute-please-to-co-host-todays-activity-7202358072055398402-L1OP?utm_source=share&utm_medium=member_desktop
2. https://www.linkedin.com/posts/adwitiya-singh-b35b321b7_ai-itu-aiforgood-activity-7203066978629005315-lgWW?utm_source=share&utm_medium=member_desktop
Ok. thanks. Any idea on the official communication on this who are winners?