Linear Transformers with Learnable Kernel Functions are Better In-Context Models Paper • 2402.10644 • Published Feb 16 • 79
