Decoupled Contrastive Decoding via Expert-Aligned Drafting
Decoupled Contrastive Decoding is introduced, which drafts with an expert-aligned lightweight proposer and applies the amateur only in unchanged CD verification and achieves average greedy speedups of 1.65 to 1.95x over vanilla CD and reduces MMLU proposal-path latency by about 5 to 12x relative to amateur-coupled prop...
Zhi-Xuan Liu, Zhichen Dong, Yuanfu Wang et al.
· 0 citations