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
Right now, the LLM prompt for text-followup instructs that the LLM should populate the response_full tag with something like "I do not have any further details to provide". After conversation with @JRegimbal , this is not great behavior since then a future application would need to parse this (and it may vary slightly if the LLM is not paying proper attention), to know that there is nothing meaningful in this tag. The schema now defines the response_full tag as optional, and this work item is to change the LLM prompt to simply omit this tag entirely if the LLM itself does not think it can produce anything meaningful.
The text was updated successfully, but these errors were encountered:
Right now, the LLM prompt for text-followup instructs that the LLM should populate the response_full tag with something like "I do not have any further details to provide". After conversation with @JRegimbal , this is not great behavior since then a future application would need to parse this (and it may vary slightly if the LLM is not paying proper attention), to know that there is nothing meaningful in this tag. The schema now defines the response_full tag as optional, and this work item is to change the LLM prompt to simply omit this tag entirely if the LLM itself does not think it can produce anything meaningful.
The text was updated successfully, but these errors were encountered: