Hi Folks,
I am using
prompt = f"reconstruct the following paragraph using IEEE Standard guidelines and make sentence in active voice and present perfect tense: '{text}'"
response = openai.ChatCompletion.create(
engine="gpt-3.5-turbo",
messages = [
{'role': 'system', 'content': 'You are a helpful assistant.'},
{'role': 'user', 'content': prompt}
])
But I am getting the vague results like below:
-
“I have inspection page 5 of the IEEE documentation / specification. Please supply i with a special sentence or hole that you can like i to rephrase accord to the guideline give in the specification.”
-
"“I sorry, but
Table of content
do n’t give i with sufficient context to rephrase the sentence. Could you please supply i with more information or a complete sentence that necessary to be rephrase ?.”,
"
How can I smartly set the prompt so that it could avoid the false positive?
Whatever the words I have provided in the prompt at least those words shouldn’t be come in result.