Pre-trained Large Language Models Use Fourier Features to Compute Addition — mentioned by astrobe_

Fri, 22 Aug 2025 17:56:34 GMT

"Isn't it just temporary, and AI will get better?" There are 2 instances of the word "understand" in the first paragraph, 3 if you count the beginning of the second. In my book, "understanding" is a synonym for "intelligence" - the roots of the word are "read between the lines", where something else that just knowledge is, the ability to use and manipulate knowledge [1]. But the thing is, despite this tech being classified as "artificial intelligence", it does not understand a thing - or so little. So, if we extrapolate Betteridge's law of headlines, no it is not temporary for this type of…

Paper and discussion history. A mention is not an endorsement.

Pre-trained Large Language Models Use Fourier Features to Compute Addition — HN submission

Thu, 06 Feb 2025 10:31:06 GMT

Pre-Trained Large Language Models Use Fourier Features for Addition (2024)

149 points; 40 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Pre-trained Large Language Models Use Fourier Features to Compute Addition · HackerPapers discussions

Pre-trained Large Language Models Use Fourier Features to Compute Addition — mentioned by astrobe_

Pre-trained Large Language Models Use Fourier Features to Compute Addition — HN submission