GPT Understands, Too

Liu Xiao, Zheng Yanan, Du Zhengxiao, Ding Ming, Qian Yujie, Yang Zhilin, Tang Jie. Arxiv 2021

[Paper]    
Applications Few Shot GPT Model Architecture Prompting Training Techniques

Prompting a pretrained language model with natural language patterns has been proved effective for natural language understanding (NLU). However, our preliminary study reveals that manual discrete prompts often lead to unstable performance – e.g., changing a single word in the prompt might result in substantial performance drop. We propose a novel method P-Tuning that employs trainable continuous prompt embeddings in concatenation with discrete prompts. Empirically, P-Tuning not only stabilizes training by minimizing the gap between various discrete prompts, but also improves performance by a sizeable margin on a wide range of NLU tasks including LAMA and SuperGLUE. P-Tuning is generally effective for both frozen and tuned language models, under both the fully-supervised and few-shot settings.

Similar Work