RESEARCH · RESEARCH · #2021
Harness^2: two-level recursive framework for evolving LLM agent harnesses
Harness^2 is a plug-and-play recursive agent-harnessing framework that uses two linked recursion levels: task-level recursion (a contrastive Propose‑Probe‑Compose step) to edit a current task's harness without ground-truth verifiers, and harnessing-level recursion to extract experience and update a continual improvement harness for future tasks. Evaluated across three professional benchmarks and four model‑harness configurations, single-step edits improved GLM-5 and Gemini 3.5 Flash over their base harnesses, with reported gains up to 11.5 points on WorkBuddy-Bench, 4.5 points on JobBench, and 11.8 percentage points on LAB.
KEY POINTS
- Harness^2 is a plug-and-play recursive agent-harnessing framework that uses two linked recursion levels: task-level recursion (a contrastive Propose‑Probe‑Compose step) to edit a current task's harness without ground-truth verifiers, and harnessing-level recursion to extract experience and update a continual improvement harness for future tasks.
- Evaluated across three professional benchmarks and four model‑harness configurations, single-step edits improved GLM-5 and Gemini 3.5 Flash over their base harnesses, with reported gains up to 11.5 points on WorkBuddy-Bench, 4.5 points on JobBench, and 11.8 percentage points on LAB.
- The paper offers a practical, generalizable method to automatically improve execution harnesses for LLM agents in open-world, unseen tasks, reducing reliance on task-specific verifiers and enabling continual adaptation.
WHY IT MATTERS
The paper offers a practical, generalizable method to automatically improve execution harnesses for LLM agents in open-world, unseen tasks, reducing reliance on task-specific verifiers and enabling continual adaptation.