Implementation · DiFR (Divergence From Reference)

Speculative decoding and multi-model sampling not evaluated

On this page

← All known flaws

SignificantOpen questionOpen

The algorithms and experiments cover sampling from a single LLM. Speculative decoding was not evaluated. The authors sketch an extension to one speculative-decoding algorithm, without experiments. They note that other variants would need modified verification and extra metadata.

Sources: [1]

Search

Full search page