连续三个空格时会将其中两个识别为PER

title = r'. ' #一个任意字符加三个空格
lac = LAC(mode='lac')
lac_result = lac.run(title)
d = {k: v for k, *v in zip(*lac_result)} #将结果转化为字典
print('lac结果:{}'.format(d))

输出： lac结果:{'.': ['w'], ' ': ['PER'], ' ': ['w']}

Start by running the reported Python snippet with LAC(mode='lac') and confirm how the three consecutive spaces are segmented and tagged. Trace the relevant LAC processing entry point to determine why two spaces receive the PER tag. Done means the same input preserves the three spaces without incorrectly labeling any as PER.

Written by the indexing model from the issue text.