Hi! Thanks for the effort!

It would be lovely to parse which datasets/benchmarks were used in the comparisons and select papers by dataset!

In many fields the datasets vary greatly depending on the subfield and its very difficult to find what other benchmarks could be used.