import re
text = "對語〈渭北　江東〉〈　平北　安東〉摘句〈力障百川東〉"
tokens = re.findall(r"([^〈〉]*)((?:〈[^〉]+〉)+)", text)
for i, (word, desc_blocks) in enumerate(tokens):
    print(f"Token {i}: WORD='{word}' DESCS={desc_blocks}")