As pointed out in other threads, Chinese splitting/parsing has recently turned terrible. It was reasonably good but now lots of
particles, prepositions, etc. are consistently being attached to words around them, creating a huge number of phrases and expressions that should be considered isolated. That drastically changes the number of words/lingqs, creating lots of new and unknown words.
Please fix it or at least reverse whatever you have done. The previous way was much better. Otherwise we will need to waste a lot of time to edit manually every BS expression created by unnecessary agglutination of words.