Find answers from the community

Updated last year

any idea if how bing makes the

At a glance
any idea if how bing makes the autogenerated next question proposals? is there any way to do it with llamaindex?
Attachment
IMG_0982_2.png
L
1 comment
Probably just an LLM call, something like

Plain Text
prompt = "Here is the latest message and response:\nMessage: {message}\nResponse: {response}\n\nWrite 3 possible follow-up questions, numbered on each line."

questions = llm.complete(prompt.format(message=message, response=str(response))

# process the text
questions = str(questions).split("\n"


Could also ask it to respond with JSON or use an OpenAIPydnaticProgram to get structured outputs
Add a reply
Sign up and join the conversation on Discord