"whitelisting" lemmatised as "whiteliste"

[chrisjbryant]

As it says in the title, the lemmatisation of "whitelisting" is wrong.

How to reproduce the behaviour

>>> import spacy
>>> nlp = spacy.load("en_core_web_sm")
>>> sent = nlp("I am whitelisting this word.")
>>> for tok in sent:
...     print(tok.text, tok.lemma_, tok.pos_)
... 
I I PRON
am be AUX
whitelisting whiteliste VERB
this this DET
word word NOUN
. . PUNCT

Your Environment

• spaCy version: 3.8.7
• Platform: Linux-6.8.0-1029-gcp-x86_64-with-glibc2.35
• Python version: 3.12.11
• Pipelines: en_core_web_sm (3.8.0)

[cyriaka90]

Similar lemmatisation issue (spaCy version: 3.8.8): "I am riding on my blue bike.": riding -> lemma: "rid" (expected: "ride")

But probably belongs to this master thread: https://github.com/explosion/spaCy/issues/3052