wafngar·7 miesięcy temu·discussVery common in ML research these days - claim novelty / cite prior work in an obfuscated way and so on.
wafngar·w zeszłym roku·discussBut they have built a fully “independent” pipeline. Deepseek and others probably trained in gpt4, o1 or whatever data.
wafngar·w zeszłym roku·discussShould be a relevant reference: https://arxiv.org/abs/2111.13587Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformers John Guibas, Morteza Mardani, Zongyi Li, Andrew Tao, Anima Anandkumar, Bryan Catanzaro