About the type of performance result.

[carottX]

I've been reading your essay and trying to reproduce it with the code you provided. I have a simple question: whether hit@1, hit@3 and mrr in your result table is hit@n(filtered) or unfiltered. Thank you for great work!

[Inrstning]

By the way, could you share the best configuration files (YAML) for the four datasets? I tried configuring the YAML file for ICEWS14 myself, but the metrics were slightly lower than those reported in the literature.

[carottX]

Hi, I've been trying to do the same thing. The metrics I produce is lower than reported. That's why I am asking this.

[thePOET8]

Same. Set as this #3 But the metrics I produce is lower than reported

[UUP68]

Yes, my reproduction results for hit@1 are much lower than those reported in the paper, and my classmates have the same results. We have reason to suspect that the authors may have altered the actual experimental results.