Spline-based Transformers

Abstract

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.

References in corpus (7)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Learning Transferable Visual Models From Natural Language Supervision
• Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
• Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
• Action2Motion: Conditioned Generation of 3D Human Motions
• Emotional Speech-Driven Animation with Content-Emotion Disentanglement
• Position Interpolation Improves ALiBi Extrapolation