In the sequential recommendation model TiSASRec, the results of the baseline model SASRec are inconsisent with the actual?

Question

I am a novice in the recommender system. Recently, I was reading a paper related to sequential recommendation. In the process of running the official sample code of TiSASRec, I used the dataset given in github repo by removing the ratings and timestamps part and using it for SASRec. Finally, the results are quite different from the comparative results given in the paper.

Dataset of TiSASRec: SASRec:

The result of TiSASRec is not much different from the description in the paper.

The result I reproduced is:

• NDCG@10: 0.5711
• HR@10: 0.8053.

But the result of SASRec is

• NDCG@10: 0.7970
• HR@10: 0.9757.

But in the paper, the results are:

I will appreciate any kind of help. Thank you all.