conftrace_

token acceptance rate

3 papers

Explore in graph

Also known as

TAR

Co-occurring keywords · papers mentioning both

large language model (2) speculative decoding (2) knowledge distillation (2) draft model calibration (1) dynamic draft tree (1) lossless acceleration (1) online adaptation (1) speculative sampling (1) draft model (1) inference acceleration (1)

Papers

SelecTKD: Selective Token-Weighted Knowledge Distillation for LLMs CVPR 2026 EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees EMNLP 2024 Online Speculative Decoding ICML 2024