D-REAL: Acceptance-Length Optimization with Unbiased Gradients for Speculative Decoding

arXiv:2611.10432 · cs.AI · Submitted 2026-11-07 · Read on arXiv

cs.AI

Submitted: 2026-11-07

Updated: 2026-11-07

Code: https://github.com/tatsu-lab/stanf

Terminology

Sources

Related papers