PanelGPT: Prompt Language Models with a Penal Discussion

Dear authors of the Prompt-Engineering-Guide team,

Thanks for the helpful repo and introduction to prompt engineering! It provides a lot of insights!

Here is how I tried to improve over previous zero-shot CoT prompts inspired by the panel discussions:

In conferences and workshops, there are always penal discussions among experts, and people exchange their opinions on a given topic, improving the understanding of new concepts, changing perspectives of thinking, and reaching a more comprehensive understanding of prevailing debates or discussions.

Getting inspiration from the panel discussions, we introduce new zero-shot prompting magic words that improve language models' reasoning ability: **Panel Discussion!**

We've experimented with large-scale datasets, with ablation studies, and present the results in this repo: https://github.com/holarissun/PanelGPT

I was wondering if the authors could kindly update our method to the prompt list.

Best regards,
Hao