ML Science: Prompt Engine: Prompt processing and refinement with NL feedback
Post 22nd May , we would further like to work on enhancement of quality based on the data the model is used on and further based on the NL feedback we would be collecting.
1. Token expansion , tokenization , pre processing and post processing of prompts
2. Refinement with natural languages and with human feedback collection :https://arxiv.org/pdf/2303.16749.pdf\
1st Iteration
We'll add a simple, fix prefix to any python request to test if it improves the suggestion quality. The transformed prompt will be something like:
Here is a correct implementation of the function
Solution:
'''python
def hello_world():
print("Hello World!")
'''
Solution:
{User's actual request}
Edited by Hongtao Yang