You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When chatting with the LLM for the chief complaint, the LLM must be able to recognize when it has enough information to construct a concise and informative Chief Complaint.
Problem
As of now, there is no way for the LLM to tell when it is finished.
Solution
We can provide the LLM with a function to call that will summarize the conversation into a concise Chief Complaint (using another LLM call), then can update the system prompt to encourage the LLM to use this function to finish the conversation.
Alternatives considered
We could also hard code an interaction limit (i.e. summarize the conversation after at most 5 interactions). Consider both LLM reasoning and hard limit.
Additional context
No response
Code of Conduct
I agree to follow this project's Code of Conduct and Contributing Guidelines
The text was updated successfully, but these errors were encountered:
Use Case
When chatting with the LLM for the chief complaint, the LLM must be able to recognize when it has enough information to construct a concise and informative Chief Complaint.
Problem
As of now, there is no way for the LLM to tell when it is finished.
Solution
We can provide the LLM with a function to call that will summarize the conversation into a concise Chief Complaint (using another LLM call), then can update the system prompt to encourage the LLM to use this function to finish the conversation.
Alternatives considered
We could also hard code an interaction limit (i.e. summarize the conversation after at most 5 interactions). Consider both LLM reasoning and hard limit.
Additional context
No response
Code of Conduct
The text was updated successfully, but these errors were encountered: