🛰️ Daily AI Frontier
‹ back to 2026-08-05

Oilbird: Training-Free Speculative Decoding with Keys the Verifier Already Computes

Research Efficiency & Systems

Ranking

Overall 76
Content 100
Popularity 19

Observed public metrics from 1 member.

Merged summary

TL;DR - Oilbird is a training-free speculative decoding method that uses verifier-computed hidden states to retrieve semantically matching drafts missed by exact suffix lookup. It improves accepted draft length and reaches 4.4× autoregressive decoding speed on API-Bank.

  • Re-keys an existing context pool with hidden states already computed at each committed token.
  • Targets tool-calling workloads where repeated requests differ mainly in newly generated values.
  • Improves accepted length by 24–29% across three published drafters at matched pool size and budget.
  • Achieves 4.4× speedup on API-Bank, versus 3.9× for the strongest training-free baseline and 2.0× for EAGLE-3.

Sources (1)

Oilbird: Training-Free Speculative Decoding with Keys the Verifier Already Computes

arXiv cs.AI Tao Jin, Phuong Minh Nguyen, Zhenzhu Yan, Teeradaj Racharak, Naoya Inoue 2026-08-04 arXiv:2608.03839
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-03 14:32:55.274092 UTC

TL;DR - Oilbird is a training-free speculative decoding method that uses verifier-computed hidden states to retrieve semantically matching drafts missed by exact suffix lookup. It improves accepted draft length and reaches 4.4× autoregressive decoding speed on API-Bank.

  • Re-keys an existing context pool with hidden states already computed at each committed token.
  • Targets tool-calling workloads where repeated requests differ mainly in newly generated values.
  • Improves accepted length by 24–29% across three published drafters at matched pool size and budget.
  • Achieves 4.4× speedup on API-Bank, versus 3.9× for the strongest training-free baseline and 2.0× for EAGLE-3.
item →