# ActiveSaddler adapts scenarios while optimizing agent harnesses

DAIR.AI describes ActiveSaddler, a paper from Microsoft and colleagues that changes which scenarios are used to optimize an agent harness. With the same optimizer, test Pass@1 improves by 4.4 points on GAIA2 and 7.5 points on Terminal-Bench 2.0 versus a fixed scenario order.

Language: en
Time zone: America/Los_Angeles
Current time (UTC, cached as of): 2026-11-22T13:00:00.000Z
Query date (America/Los_Angeles): 2026-11-22

HTML: https://didcodexreset.com/news/a8ef35607d3d44a0aa958a6e.html

Content language: en
Localization state: sameLanguage

Source: DAIR.AI · Published 10/3/2026, 22:51:10

ActiveSaddler groups recurring failures into failure patterns and treats each pattern as an arm in a non-stationary bandit. It tracks how much the harness is still learning from each pattern and splits the budget between revisiting known weaknesses and finding new ones.

DAIR.AI says current harness optimizers change how the harness is updated but keep the training scenarios fixed, so feedback continues to come from tasks that stop being informative as the harness improves. ActiveSaddler adapts those scenarios as well. With the same optimizer, test Pass@1 improves by 4.4 points on GAIA2 and 7.5 points on Terminal-Bench 2.0 compared with a fixed scenario order. The paper is [ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization](https://academy.dair.ai/papers/activesaddler-automated-curriculum-learning-for-agent-harness-optimization-2610.00906).

Tags: ActiveSaddler, Agent harness, Microsoft, GAIA2, Terminal-Bench

[View original post](https://x.com/dair_ai/status/2106620084542419257)
