Closed noone-code closed 1 month ago
Hi, we are currently occupied with other projects and have not tested this new method.
I am very curious about the performance of this latest and strongest method. It should demonstrate and push the performance limit of methods in this field. I look forward to you testing and comparing it to showcase the advantages of Lod-Loc.
According to the results published by GIM, the GIM trained model has good generalization performance. The currently published $GIM_{DKM}$ has the best performance (better than RoMA). I am curious if you have tested GIM's model?