DeLS-Spec improves block-parallel speculative decoding by fusing DFlash logits with an independently trained lightweight local head and a unigram prior correction.
Speculative diffusion decoding: Accelerating language generation through diffusion
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting
DeLS-Spec improves block-parallel speculative decoding by fusing DFlash logits with an independently trained lightweight local head and a unigram prior correction.