JetSpec trains a causal draft head to produce branch-consistent trees aligned with target autoregressive scores, achieving up to 9.64x speedup on MATH-500 and outperforming prior SD baselines on Qwen3 models.
Direct alignment of draft model for specula- tive decoding with chat-fine-tuned llms.arXiv preprint arXiv:2403.00858
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
No detectable safety divergence between target-only and speculative decoding at temperature zero under TAIS criteria on 48,072 samples across safety benchmarks.
ConFu boosts speculative decoding acceptance rates 8-20% over EAGLE-3 by letting draft models use contemplate tokens and MoE to anticipate future generation direction.
citing papers explorer
-
JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting
JetSpec trains a causal draft head to produce branch-consistent trees aligned with target autoregressive scores, achieving up to 9.64x speedup on MATH-500 and outperforming prior SD baselines on Qwen3 models.
-
Speculative Decoding at Temperature Zero: A Scoped Safety-Invariance Screen with a 48,072-Sample Expansion
No detectable safety divergence between target-only and speculative decoding at temperature zero under TAIS criteria on 48,072 samples across safety benchmarks.
-
ConFu: Contemplate the Future for Better Speculative Sampling
ConFu boosts speculative decoding acceptance rates 8-20% over EAGLE-3 by letting draft models use contemplate tokens and MoE to anticipate future generation direction.