Hello! Your method of scoring videos across multiple dimensions is fascinating. I experimented with videos of the same content but different bitrates and discovered that higher bitrate videos scored lower instead(average of ten times). What could be causing this? Could there be an issue with how I reproduced the experiment? I would greatly appreciate your insights on this matter.