HarnessMM Cuts Agent Harness Token Use by 88%

A Tsinghua–Microsoft Research paper introduces HarnessMM, which reuses established agent decisions with guardrails and recursive self-improvement, cutting harness token use from 96K to 11.6K (about 88%) while improving success rates.

2026-10-26 ~ 2026-10-26 · 2 related posts

• HarnessMM paper cuts agent token use 87.9% on DeepSeek harness by reusing decisions — blaizedsouza · 2026-10-26
• HarnessMM cuts agent tokens per task by up to 88% while raising success rates — omarsar0 · 2026-10-26