3 ms·
And how do you explain that they can do advanced mathematics?
by pixel_popping 6mo ago
And how do you explain that they can do advanced mathematics?
- jqpabc123 6mo agoHow do you explain when they can't? If they *really* understand math, why is the error rate so high? According to a 2025 Stanford HAI report, large language models fail basic multi-step arithmetic up to 40% of the time without external tools. https://medium.com/@dojolabs.main/why-does-ai-get-math-wrong-understanding-ai-calculation-errors-1586128f1da5 https://medium.com/@dojolabs.main/why-does-ai-get-math-wrong...
- pixel_popping 6mo ago2025... have you checked latest models? and even, this talk is irrelevant when we know in few months/years this exact topic will be solved.
- jqpabc123 6mo agowe know in few months/years this exact topic will be solved You may know this somehow --- but I don't. Without a fundamental re-design, the basic problem will remain. I don't believe it is possible to apply statistics to predict answers without significant errors.
- pixel_popping 6mo agoyes but most humans (also without tools, to do a fair comparison) also make significant errors, WAY more than Opus 4.7 and GPT-5.4 xhigh
- jqpabc123 6mo agoyes but most humans (also without tools, to do a fair comparison) also make significant error Humans adopted the use of computers because they provided accurate answers at low cost. At least until recently. Now, LLMs provide questionable answers at high cost.
- pixel_popping 6mo agobut LLMs + Tools (computers) right now beat large majority of humans (and we know this gap will keep widening), so how does that make them "not intelligent"?