用样条曲线替代位置编码，让Transformer更灵活地处理序列
Spline-based Transformers
- 用样条曲线在隐空间中表示序列，无需传统位置编码
- 在2D合成数据到真实图像/3D动画数据上表现更优
- 可直接操控隐空间控制点生成新序列，交互性强
我们提出Spline-based Transformers，一种新型Transformer模型，无需位置编码。受计算机动画中样条曲线工作流的启发，该模型将输入序列嵌入为隐空间中的平滑轨迹。克服了位置编码在序列长度外推上的缺陷，同时提供了一种用户直接操作隐空间控制点以生成新轨迹和序列的新方式。我们在多种数据集上验证了该方法的优越性，涵盖从二维合成数据到大规模真实世界数据（图像、3D形状、动画）。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。