Aurora unifies speculative decoder training and serving via asynchronous RL on inference traces, delivering 1.5x day-0 speedup on frontier models and 1.25x adaptation gains on distribution shifts.
Fast inference from transformers via speculative decoding, 2023
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2representative citing papers
STS repurposes draft-model attention scores from speculative decoding to build token-and-head-wise sparsity masks, delivering 2.67x speedup at ~90% sparsity on NarrativeQA with negligible accuracy loss.
citing papers explorer
-
When RL Meets Adaptive Speculative Training: A Unified Training-Serving System
Aurora unifies speculative decoder training and serving via asynchronous RL on inference traces, delivering 1.5x day-0 speedup on frontier models and 1.25x adaptation gains on distribution shifts.
-
STS: Efficient Sparse Attention with Speculative Token Sparsity
STS repurposes draft-model attention scores from speculative decoding to build token-and-head-wise sparsity masks, delivering 2.67x speedup at ~90% sparsity on NarrativeQA with negligible accuracy loss.