RobustSGPO: Search-Space Control for Agent Harness Evolution

2026-09-08

AI Summary

该研究通过引入RobustSGPO方法，控制搜索空间以优化代理执行框架，解决了语义梯度提示优化中编辑范围和操作选择的问题，提高了任务完成率和测试质量。

Abstract

Semantic-gradient-based prompt optimization (SGPO) improves agent harnesses using execution feedback, but its local update rule leaves the choice of edit scope and operation unresolved. We introduce RobustSGPO, which specifies the requested edit, constructs and checks the patch, and continues search from either the incumbent or retained snapshots. We evaluate permission scheduling, cumulative controls, and task-family transfer in the AgentX brainstorming workflow using 120 tasks, 95 runs, and 7,350 candidate attempts. Periodic $1\to2\to3$ scheduling exceeds fixed maximum permission by 0.28 test-score points. RobustSGPO increases completion on 30 held-out tasks from 60.0% to 80.0% and improves test quality from 3.77 to 4.14 under a 20-million-token budget. Category retention reduces source-task degradation after a shift, whereas random retention reaches a higher destination endpoint. Search-space control benefits quality through executable edits and alternative starting points, with measurable retention overhead.

Problem

• Semantic-gradient-based prompt optimization
• edit scope
• operation selection

Innovation

• RobustSGPO
• Search-Space Control
• Semantic-Gradient-Based Prompt Optimization
• Agent Harness Evolution
• Execution Feedback