Lamini Memory Tuning: 10x Fewer Hallucinations
1–10 of 63 posts
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#210x less is a weird way of saying 90% less, or better yet reduced to 10% from before.
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#310x less is a weird way of saying 90% less, or better yet reduced to 10% from before.
so if ChatGPT hallucinates 10% of the time, their model hallucinates only 1% of the time?
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#4What is the hallucination rate of, for example, a Llama3 or GPT4?
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#5Has anyone here used this, or anything similar? This sounds phenomenal if it really works. Looks like “contact us” is the only way to try it or buy it right now, and the purported benefit (memorize facts up to the model training size, basically trillions of tokens of facts) is wild. I’d love to try a system running this way to understand the failure modes, like for instance how does it reliably infer which “facts” to use?
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#6"Hallucinations" are the creative aspect of LLMs, which is what they are more useful for- if anything we want more of them. We already have much simpler systems that search and regurgitate facts. We need more intelligent hallucinations that are consistent with and extend rather than conflict with the data.
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#7Woww, very creative and interesting idea: I understand it as: train a bunch of fact-based LoRAs to zero loss (they mention 100k different ones), then use RAG to pick the appropriate Loras for a query.
So cool. The only moat I can think of for such a company would be proprietary fact loras- basically licensing a modern ai encyclopedia.
Anyway, really nice idea.
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#8Can it win at Jeopardy?
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#9Here I was hoping that there would be some kind of regulatory framework or protections put in place for AI before it became smart enough to actually take over the world.
Being able to say "you are wrong, taking over the world is a bad idea" and have the model respond with "oh you are completely right, I am very sorry for that" was our first line of defense.
I wonder if this model will argue insistently that you are wrong if you try and tell it that 1+1=3, and if so, whether that expands to philosophical issues such as the model arguing back at you based on its formed opinions on ethics and history.
Re: Lamini Memory Tuning: 10x Fewer Hallucinations
#10Hopefully someone reproduces results with code... cant find any code they shared