HarnessPro：用帕累托优化让 Agent Harness 自我进化，性能与成本兼顾
HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs
一篇讲怎么让 Agent 框架自动进化的论文，产出的 harness 比之前的 RHO 又快又便宜，代码已开源可以直接看。
HarnessPro 是一种无标签的 harness 递归自我改进方法，用 LLM 生成的测试代替真实答案来评估候选 harness。它通过专门 agent 诊断执行轨迹、按不同性能-成本优先级更新，最终输出 lite、medium、heavy 三档 harness 以适配不同部署预算。实验覆盖 Terminal-Bench 2、SWE-Bench Pro 和 ProgramBench，使用 GPT-5.5 搭配 Codex 与 GLM-5.2 搭配 Claude Code。在 GPT-5.5 + Codex 设置下，heavy 模式相比 RHO 平均 pass rate 提升 2.0 个百分点，单任务成本平均降低 26.1%。
HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs
Harness development plays a crucial role in building capable AI agents. Recent approaches improve harnesses through recursive self-improvement (RSI), but their performance gains often come with higher execution costs. We introduce HarnessPro, a label-free method for evolving harnesses to improve task performance while reducing execution cost. In this setting, candidate harnesses are evaluated using LLM-generated tests without access to ground-truth solutions or official tests. Specialized agents diagnose execution trajectories, plan updates with different performance-cost priorities, and implement them as candidate harnesses. Rather than returning a single harness, HarnessPro uses normalized weighted squared-distance selection to provide cost-focused (lite), balanced (medium), and performance-focused (heavy) harnesses for different deployment budgets. Selected harnesses enter the next iteration, where execution feedback guides further updates. Our experiments cover Terminal-Bench 2, SWE-Bench Pro, and ProgramBench, using GPT-5.5 with Codex and GLM-5.2 with Claude Code. Across these benchmarks using GPT-5.5 with Codex, the heavy mode of HarnessPro achieves an average pass-rate gain of 2.0 percentage points and an average per-task cost reduction of 26.1% relative to RHO. This advantage also holds when using GLM-5.2 with Claude Code. Code is available at https://github.com/pzs19/HarnessPro