The missing highlights due to PDF processors #2
Description
There could be around 20% of highlighted sentences that were lost in the final annotation data because of the python PDF highlight extraction library. This needs to be dealt with.