The post asks if the CitationQueryEngine is just a fancy prompt. The comments indicate that this is correct, and that the engine only slightly modifies the chunk text. Community members also express concerns about relying on large language models (LLMs) for citation, and suggest that fuzzy matching may be a more reliable approach. There is no explicitly marked answer in the provided information.
Yea after reading the implementation it wasn't what I thought and don't trust it to cite itself based on a prompt really, I'll look int othe fuzzy mathcing approach!
