DCOCO: new speculative decoding method restores parallel draft context, up to 6.6x faster
arankomatsuzaki · x · 2026-10-26
A new speculative decoding method, DCOCO (Contextual Correction of Parallel Drafts), addresses the problem that parallel drafters lose context between predicted tokens. DCOCO restores that context without giving up parallelism, achieving up to 6.6x speedup over vanilla decoding.
More from Research
- MemTensor's MemT-1.0-Nano trends on Hugging Face, a memory-tuned Qwen3.6-35B-A3B finetune — MemTensor · 2026-10-26
- Mathematicians dissect OpenAI's fluid singularity paper: unprecedented smooth-force construction — Singularitarian · 2026-10-26
- EMNLP 2026 Findings: "Ask or Assume?" turns coding agents into proactive help-seekers — najoungkim · 2026-10-26
- Subliminal learning transfers hidden traits across models without a shared base model — burny_tech · 2026-10-26
- Implementing SynthID-Text From Scratch: Watermarking Doesn't Hurt Output Quality — ell-hol1 · 2026-10-26
- Two teams independently discovered the same interpretability method—a sign the idea is robust — sedielem · 2026-10-26