Human evaluation data

[vishaal27]

Hi, thanks for your great work and for releasing the code directly. I had a question about your human evaluation data: would you be able to make the per-sample level caption similarity task data public i.e. the human data collected from Fig 10a? It would be super helpful if I can get access to this data, so please let me know if this is feasible to accommodate?