Duplicate Matching returns a lot of false positives

[zygoballus]

At https://docs.symbiota.org/Editor_Guide/Editing_Searching_Records/duplicate_matching/ it says that "The Duplicate Matching tool allows the user to search the entire database (i.e., all collections in the Symbiota portal) for records that share the same collector name, collector number, and date." However, when I use the tool it seems to find a lot of false positives. For example, for a record collected by "B. & C. Durden" on 1965-07-03 (with no collector number), it returns matches to specimens collected by "Lois B. O'Brien" and "B. Daniel" (on the same day with no collector numbers) as well as all of the other specimens collected by B. & C. Durden on the same day. I guess I can understand blank collector numbers matching, but "B. & C. Durden" shouldn't match to "Lois B. O'Brien" and "B. Daniel". I imagine this is because it's splitting "B. & C. Durden" on the & and then finding "B." in the other records, but that seems like an overly loose strategy.