3 ms·
A lot of LLM hallucination is because of the internal conflict between alignment for helpfulness and lack of a clear answer. It's much like when someone gets o
by CuriouslyC 3y ago
A lot of LLM hallucination is because of the internal conflict between alignment for helpfulness and lack of a clear answer. It's much like when someone gets out of their depth in a conversation and dissembles their way through to try and maintain their illusion of competence. In these cases, if you give the LLM explicit permission to tell you that it doesn't know in cases where it's not sure, that will significantly reduce hallucinations.
A lot more of LLM hallucination is it getting the context confused. I was able to get GPT4 to hallucinate easily with questions related to the distance from one planet to another, since most distances on the internet are from the sun to individual planets, and the distances between planets varies significantly based on their locations in cycle. These are probably slightly harder to fix.
- danenania 3y ago"In these cases, if you give the LLM explicit permission to tell you that it doesn't know in cases where it's not sure, that will significantly reduce hallucinations." I've noticed that while this can help to prevent hallucinations, it can also cause it to go way too far in the other direction and start telling you it doesn't know for all kinds of questions it really can answer.
- sumtechguy 3y agoMy current favorite one is to ask the time. Then ask it if it is possible for it to give you the time. You get 2 very different answers.
- mzi 3y agoIt also has a problem with quantity, so it gets confused by things like the cube root of 750l that it maintains for a long time is around 9m. It even suggests that 1l is equal to 1m³.