I wouldn’t be surprised if this is the case and I come from ML research which tends to have a very high standard of transparency and we still end up with nonsen

I wouldn’t be surprised if this is the case and I come from ML research which tends to have a very high standard of transparency and we still end up with nonsense results