is there a way not to match short queries like

golf

->

gold

without bluntly increasing

min_len_1typo

? i would like to keep the typo tolerance for short words (i.e.

teh

->

the

), but prevent results for

gold

from appearing as results for

golf

, since there are lots of results for both queries. is there a way to either reduce the textmatch score if its a typo-match, or downrank typos when there are sufficient results for the exact match?

Maybe using drop_tokens_mode=left_to_right or typo_tokens_threshold=<value>: https://typesense.org/docs/29.0/api/search.html#typo-tolerance-parameters