FR-Spec: Accelerating Large-Vocabulary Language Models via Frequency-Ranked Speculative Sampling Paper • 2502.14856 • Published Feb 20 • 8
From Hours to Minutes: Lossless Acceleration of Ultra Long Sequence Generation up to 100K Tokens Paper • 2502.18890 • Published Feb 26 • 30
DuoDecoding: Hardware-aware Heterogeneous Speculative Decoding with Dynamic Multi-Sequence Drafting Paper • 2503.00784 • Published Mar 2 • 13
