Improving Language Understanding by Generative Pre-Training Language Models are Unsupervised Multitask Learners Language Models are Few-Shot Learners 对应这三篇paper https://zhuanlan.zhihu.com/p/609367098 知乎上有对应的讲解 GPT1的核心思路是,通过language model做预训练,然后再针对下游的任务做fine tu…