the deviation of results

[Honerlaaco]

Hi! When I reproduced the experimental results of the Facebook dataset in Table 2, some results showed a deviation of 1-4, such as sage-∞, gap-edp and mlp-dp, and some results even showed a large error, such as for SAGE-NDP, my result was 76.9; Also, for gap-ndp, my result is 79.2. I want to know if this is possible or if there is something wrong with my practice