Microsoft's FOCUS compresses agent context at test time: 48% less context, +8.9 points success
dair_ai · x · 2026-10-02
A new Microsoft paper introduces FOCUS, a training-free method that compresses agent context at test time by identifying which past interactions the agent's next decisions actually depend on and dropping the rest. It works as a standalone layer in front of closed-API models with no fine-tuning. On tool-calling, QA, web and multi-turn dialogue benchmarks it cuts peak context by up to 48% and improves task success by up to 8.9 points versus running on full history.
More from coding & agent
- AI in finance shows a power law: top 1% users generate 35x more files with agents — peterjliu · 2026-11-19
- Linux LTS patch flood: 3318 patches for 6.6 alone, 'no way humans are reviewing this' — pkqzy888 · 2026-11-19
- Caliper Harness: open-source coding harness disciplines small local models with plans, checks, reviews — Saraozte01 · 2026-11-19
- AI systems lack executive function, not scale: four manager-layer strategies to fix it — Designer-Material633 · 2026-11-19
- Google employee replaces doomscrolling with 3 fully automated personal AI newsletters — DynamicWebPaige · 2026-11-19
- Designer spins up an agent to build state-airline brand gallery mid-flight delay — beh_zod · 2026-11-19