Microsoft's John Langford finds optimized 'free pause tokens' boost transformer inference near-free
JohnCLangford · x · 2026-09-08
John Langford shares a fun paper: his team ran State Prediction Separation through an optimization process, finding it works at higher op and yields 'free pause tokens' that let a transformer make higher-quality inference at near-zero extra cost.
A lightweight approach to letting models think longer without a meaningful inference cost increase.
More from Research
- New Blog Breaks Black-Box Distillation Into Five Practical Steps — chrisrohlf · 2026-10-26
- DetectorScope turns black-box AI text detectors into auditable ones with sparse autoencoders — _akhaliq · 2026-10-26
- OPST paper shows LLMs self-improve with zero human data, guided only by topic descriptions — _akhaliq · 2026-10-26
- SJTU's VaM lets robots learn online from mistakes, 95%+ success over 4-hour autonomous runs — chris_j_paxton · 2026-10-26
- Hugging Face uses small open models to reverse engineer a game binary byte-for-byte — yb2698 · 2026-10-26
- Author hopes contracting formalism becomes AI's "MDP moment" for agent economics — xuanalogue · 2026-10-26