# Preprint reports benchmark gains from agents revising their execution workflows

_Published Thursday, October 15, 2026 at 1:05 AM EDT · AI, Science · Latest · Tier 2 — Notable_

Researchers report that Harness^2, a framework that lets AI agents revise the workflows governing their execution, improved performance across three professional benchmarks and four model-workflow configurations. The preprint reports gains of up to 11.5 points on WorkBuddy-Bench, 4.5 points on JobBench, and 11.8 percentage points on LAB.

The framework probes proposed workflow changes without ground-truth verifiers, then draws on those edits to improve adaptation to future tasks. A single revision step consistently improved GLM-5 and Gemini 3.5 Flash over their base workflows, with further gains as the revision budget increased. The reported results concern benchmark tasks rather than demonstrated production deployments.

## Sources

- [cs.AI updates on arXiv.org](https://arxiv.org/abs/2610.14224)

---
Canonical: https://techandbusiness.org/newswire/egxYmu5SEDvI7hOMmbsR-J
Published: 2026-10-15T05:05:54.313Z
Story chronology: 2026-10-15T04:00:00.000Z
Retrieved: 2026-10-15T07:38:15.823Z
Publisher: Tech & Business (techandbusiness.org)