3 ms·
Yes, several. • https://arxiv.org/pdf/2106.14742.pdf https://arxiv.org/pdf/2106.14742.pdf • https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/met.2098 ht
by ben_w 3y ago
Yes, several.
• https://arxiv.org/pdf/2106.14742.pdf https://arxiv.org/pdf/2106.14742.pdf
• https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/met.2098 https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/met.2...
• https://ieeexplore.ieee.org/document/9671442 https://ieeexplore.ieee.org/document/9671442
Also non-transformer models, because of the scaling complexity on input: https://arxiv.org/pdf/2212.12794.pdf https://arxiv.org/pdf/2212.12794.pdf