When will the full dataset be released?

#12
by DongmingZhang - opened

Hi Turing team,

Thank you for releasing Open-RL - this dataset looks very valuable for reinforcement learning training on STEM reasoning problems!

I noticed that the current HuggingFace repository only contains 40 sample entries. The README mentions "Future Directions" with NuRL-style nudging augmentation, but there's no specific timeline mentioned.

Could you please share:

  1. When do you plan to release the full dataset?
  2. Is there any way to request access to the complete data before public release?
  3. Will the "nudge augmentation layer" be released together with the full dataset?

This would greatly help with planning our RL training experiments. Looking forward to your update!

Hi @DongmingZhang ,

Could you please share details here ?
We will get in touch with you with all the details

saurabh-gnirut changed discussion status to closed

Sign up or log in to comment