researchers with a paper here

1 paper

cs.CL2026

DominoTree: Conditional Tree-Structured Drafting with Domino for Speculative Decoding

Saw S. Lin, Jyh-Shing Roger Jang

Speculative decoding accelerates LLM inference by drafting tokens and verifying them in parallel. Block-diffusion drafters such as DFlash model only per-position marginals, and tre…