Question about ogbench-scene task result

Thanks for your excellent work！

I would like to inquire about the implementation of ground truth data replay for different tasks in the OGBench scene scenario. Specifically, could you please advise on the proper methodology to execute this functionality and whether there are any existing documentation or guidelines available for reference?


