Show HN: Explore large language models on any computer with 512MB of RAM

Re:Show HN: Explore large language models on any computer wi... by scarface_74 on Saturday June 17, 2023 @06:08PM (#36375193)

Is there theoretically a method to have an LLM to not hallucinate and just say “I don’t know” or just answer questions based on a certain domain of knowledge which it is well trained on?

Could a business ever trust an LLM based chatbot as much as the old school chatbot where certain questions reliably give certain answers and fails when it doesn’t know?

[ Parent | on HN ]

Re:Show HN: Explore large language models on any computer wi... by Animats on Sunday June 18, 2023 @12:12AM (#36377241)

> Is there theoretically a method to have an LLM to not hallucinate and just say “I don’t know” or just answer questions based on a certain domain of knowledge which it is well trained on?

That is a very good question. What Google returns for "LLM hallucination" mostly describes post-processing hacks to detect the problem, or pre-processing hacks to guide the LLM into using data directly relevant to the question asked. Not doing something to the core LLM system to get some measure of confidence out.

Anyone working on this?

[ Parent | on HN ]