Ehsan Hemmati

Hi, @Guaguago Can you share how you calculate PPL?

@Guaguago Thanks. Perplexity is usually calculated for the test set, However maybe the authors used the generated text to compute the perplexity since they don't have ground truth target. In...

@dathath Thanks. Can you correct me if I am wrong? are perplexities usually computed on the test set? that is using the NLL of the trained model on the target...

For some reasons that I cannot figure out yet, when I generate single samples vs when I generate using all inputs from a text file, the first gives better results....

@dathath I think I figure out the reason. In this regard, should the perturb and unperturbed generation happen together? The reason I am asking is when I comment out the...