4 ms·
I wonder in LLMs either need to be rethought/retrained to say "I don't know" instead of guessing (I have no clue what would go into this) or... if we somehow a
by MuffinFlavored 3y ago
I wonder in LLMs either need to be rethought/retrained to say "I don't know" instead of guessing (I have no clue what would go into this) or...
if we somehow are able to take a "test-driven-approach" with them.
Upfront, we tell it what criteria must pass. Connect it to a "test/task" runner of some sort per prompt. Then, it keeps trying behind the scenes to come up with an answer that passes the criteria (to prevent humans from having to waste iterations on 'no, you got it wrong, try again')
However, I will admit... most of the time when the LLM can't solve it, retrying and retrying over and over slightly different ways really doesn't typically end up in success. If the task is "too complex" for the LLM, I have yet to find a way to break it down context wise and overcome the limitation.
- zitterbewegung 3y agoIn some situations you can make the prompt tell you. "I don't know". For example "You are a friendly D&D dungeon master and know all of the rules of D&D and run the game. If you interact with a player you should observe all of the rules. If you don't know if they are following the rules then say "I don't know"
- viraptor 3y ago> I wonder in LLMs either need to be rethought/retrained to say "I don't know" instead of guessing (I have no clue what would go into this) You can already do this yourself to some extent by providing examples in the prompt. Ask your question, explicitly add that you're allowing the "don't know" answer and add that for this similar question the answer is X, for that the answer is don't know, for that the answer is Y. It's a bit of work, but it works in practice. Check out the concept of priming. Yeah, it would be nice if this was pre-trained - maybe in the future.
- trifurcate 3y agoNon-instruct-tuned LLMs are far more likely to say "I don't know".