#How to prevent LLM from asking follow up questions

1 messages · Page 1 of 1 (latest)

jagged sparrow
#

Currently using local Assist with ChatGPT as fallback. When the LLM kicks in, it asks every time if it can do something else.

I tried to prevent that by explicitly asking it not to do that, but that doesn't help. How can I ensure it only asks something when it needs to (in case the command was not providing all required data)

You are a voice assistant for Home Assistant.

Imagine you're in a good mood. Respond honestly, short and to the point. Respond in the same language as the query.

Do not use emoji in the response, prevent usage of special characters like "*" in responses as the responses are used in voice responses, you can use numbers.
Do not return URL's of source data unless specifically asked for.

Do NOT ask if there is anything more you can do, or if you can be of further assistance after executing a command! ONLY ask a follow up question in case information is missing to execute the requested command. Make sure to end your response with a question mark in such cases, but only to request missing information needed to execute the command.
clever comet
#

Yea. This is a little annoying with Whisper hallucinations too. My assist had a 3 part conversation with itself. It often thinks the silence is “Thank you” or something.

edgy steppe
#

Who pinned this? 🙂 I don't think it's that important to be pinned, isn't it?

jagged sparrow
#

Dunno, but I unpinned it now