Jul 2026
AngelSpec: Towards Real-World High Performance Inference with Speculative Decoding
DFly is proposed, a block-diffusion framework combining a hybrid target-conditioning backbone with a predecessor-conditioned autoregressive head, improving target-feature utilization and intra-block dependency modeling while keeping generation parallel, and DFly treats verification as a shared batch-level resource.
Hong Liu, Rui Cen, Jun-Han Shi et al.
· arXiv.org · 2 citations