Evaluation datasets/code for downstream tasks (Section 7)

[wang-rq]

Hi AION team,

Thanks for releasing AION and the demos. This is a very exciting project. Could you share the datasets/cross-match code used for the downstream tasks reported in Section 7?

Any guidance or release of the evaluation tasks would greatly improve reproducibility and help the community build additional tasks on top of AION. Thanks again for your work.