Scholar Feed

Speculative decoding

DynaSpec

DynaSpec: Context-aware Dynamic Speculative Sampling for Large-Vocabulary Language Models

Superseded baseline#34 of 151 most-superseded · first seen Oct 11, 2025

Superseded — cited as a baseline and beaten by newer methods

2 papers critique it · 1 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites DynaSpec as a baseline.

still rely on statically pre-clustered vocabularies, resulting in non-negligible training overhead.
— MicroSpec: Accelerating Speculative Decoding with Lightweight In-Context Vocabularies
“The latter can become a noticeable bottleneck on GPUs because it involves such operations as global ranking, partial sorting, irregular indexing and gathering a context-dependent subset of weights, which are less efficient than dense matrix multiplication.”
— SlimSpec: Low-Rank Draft LM-Head for Accelerated Speculative Decoding

Beaten on benchmarks

Head-to-head results where a newer method reports beating DynaSpec. Values are copied from the source paper's tables — verify against the cited paper.

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.