Spline-based Transformers
We introduce Spline-based Transformers, a novel class of Transformer models that eliminate the need for positional encoding. Inspired by workflows using splines in computer animation, our Spline-based Transformers embed an input sequence of elements as a smooth
trajectory in latent space. Overcoming drawbacks of positional encoding
such as sequence length extrapolation, Spline-based Transformers also
provide a novel way for users to interact with transformer latent spaces
by directly manipulating the latent control points to create new latent
trajectories and sequences. We demonstrate the superior performance of
our approach in comparison to conventional positional encoding on a
variety of datasets, ranging from synthetic 2D to large-scale real-world
datasets of images, 3D shapes, and animations.
Publication link: https://la.disneyresearch.com/publica...
In questa pagina del sito puoi guardare il video online Spline-based Transformers della durata di ore minuti seconda in buona qualità , che l'utente ha caricato DisneyResearchHub 23 settembre 2024, condividi il link con amici e conoscenti, su youtube questo video è già stato visto 5,506 volte e gli è piaciuto like spettatori. Buona visione!
