Primary computational bottleneck in scoring: cosine computation Can we avoid all this computation? Yes, but may sometimes get it wrong a doc not in the top K may creep into the list of K output docs Is this such a bad thing?
Primary computational bottleneck in scoring: cosine computationCan we avoid all this computation?Yes, but may sometimes get it wronga doc not in the top K may creep into the list of K output docsIs this such a bad thing?