Microsoft researchers find optimized "free pause tokens" boost LLM reasoning at near-zero cost

A Microsoft Research team led by John Langford ran State Prediction Separation through an automated optimization process and discovered "free pause tokens" that improve Transformer reasoning quality at almost no cost.

2026-09-08 ~ 2026-09-08 · 2 related posts

• Microsoft's John Langford finds optimized 'free pause tokens' boost transformer inference near-free — JohnCLangford · 2026-09-08
• Optimized SPS yields 'free pause tokens' that boost transformer inference at 10-20% overhead, down from ~700% — JohnCLangford · 2026-09-08