Language models getting larger and larger

Language models trained on more and more data

Prompting

zero-shot learning

few-shot learning

chain-of-thought prompting

chain-of-thought prompting is an emergent property of model scale

Fine-tuning

Instruction fine-tuning

单纯的LLM并不是一个很好的助手，仅仅是一个能够预测下一个词的model