Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation Paper • 2610.14162 • Published about 1 month ago • 27
The Hedgehog & the Porcupine: Expressive Linear Attentions with Softmax Mimicry Paper • 2402.04347 • Published Feb 6, 2024 • 15
FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores Paper • 2311.05908 • Published Nov 10, 2023 • 14