2 ms·
There is already misinformation and incorrect facts in llm training data. It still gets things right by nature of how it’s designed to generate output.
by PUSH_AX 2y ago
There is already misinformation and incorrect facts in llm training data. It still gets things right by nature of how it’s designed to generate output.