search

MikeWangWZHL / Solo-Performance-Prompting

Repo for paper "Unleashing Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration"
https://arxiv.org/abs/2307.05300
313 stars 28 forks source link

The baseline of CoT is not not in few-shot setting? #4

Open Tebmer opened 1 year ago

Tebmer commented 1 year ago

Hi there :)

I thoroughly read your work. Your work is awesome and intersting.

However, there is demonstrations in SPP (few-shot setting) but no demonstrations in the CoT baselines (zero-shot setting). So, the improvement may be attributed to the demonstrations? Could you provide the results in the CoT prompting with the same demonstrations (few-shot setting)?