21–30 of 63 posts
Doesn't this make the "AI" even less creative and more like full-text-search instead? What makes some data a "fact"? If everything is written in the training data, in the end, won't everything be treated like a fact? So the LLM will have 100% accuracy and 0% creativity.
Doesn't this make the "AI" even less creative and more like full-text-search instead? What makes some data a "fact"? If everything is written in the training data, in the end, won't everything be treated like a fact? So the LLM will have 100% accuracy and 0% creativity.
Doesn't this make the "AI" even less creative and more like full-text-search instead? What makes some data a "fact"? If everything is written in the training data, in the end, won't everything be treated like a fact? So the LLM will have 100% accuracy and 0% creativity.
Sounds like compression to me
Am I the only one that cringes at "10x fewer?" How do I multiply positive numbers and get something smaller? Is "1/10th" or "90% less" not better arithmetic? Maybe I should have done more gooder at math but it hurts my ears (eyes).
Am I the only one that cringes at "10x fewer?" How do I multiply positive numbers and get something smaller? Is "1/10th" or "90% less" not better arithmetic? Maybe I should have done more gooder at math but it hurts my ears (eyes).
fractions are gonna blow your mind
Am I the only one that cringes at "10x fewer?" How do I multiply positive numbers and get something smaller? Is "1/10th" or "90% less" not better arithmetic? Maybe I should have done more gooder at math but it hurts my ears (eyes).
I am also annoyed by most modern tech marketing using percentages incorrectly and inconsistently. But 150% is a bigger number than 1.5x so I suppose their hands are tied.
Am I the only one that cringes at "10x fewer?" How do I multiply positive numbers and get something smaller? Is "1/10th" or "90% less" not better arithmetic? Maybe I should have done more gooder at math but it hurts my ears (eyes).
Am I the only one that cringes at "10x fewer?" How do I multiply positive numbers and get something smaller? Is "1/10th" or "90% less" not better arithmetic? Maybe I should have done more gooder at math but it hurts my ears (eyes).
eg. if hallucinations occur roughly 1 in 20 prompts then 10x fewer is 1 in 200 prompts, rather than 0.1 in 20 prompts.
Doesn't this make the "AI" even less creative and more like full-text-search instead? What makes some data a "fact"? If everything is written in the training data, in the end, won't everything be treated like a fact? So the LLM will have 100% accuracy and 0% creativity.
"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.
Going from 50% "accurate" to 90% "accurate" may actually be more insidious because it changes the utility from being a coin flip to trying to determine which 10% is inaccurate, or downplaying the existence of inaccuracies because at 90% it is "mostly correct".