arXiv cs.CL· Gaotang Li, Bhavana Dalvi Mishra, Yifei Ming, Yanfei Chen, Zhongying CuiZhu, Lesly Miculicich, Long T. Le, Rui Meng, Chun-Liang Li, Hanghang Tong, Chen-Yu Lee, Tomas Pfister·· 12 小时前AI 评分45
OO-Harness:用面向对象设计演化智能体执行框架
Object-Oriented Harness: Evolving Agent Harnesses through Object-Oriented Design
AI 导读
研究者提出 Object-Oriented Harness(OO-Harness),一个用面向对象设计演化可执行 harness 类层级的框架:训练时由演化智能体基于执行记录,通过继承生成一系列专用 harness 子类,测试时由 harness 路由器为每个任务选择具体 harness。
正文 · 原文
Authors:Gaotang Li, Bhavana Dalvi Mishra, Yifei Ming, Yanfei Chen, Zhongying CuiZhu, Lesly Miculicich, Long T. Le, Rui Meng, Chun-Liang Li, Hanghang Tong, Chen-Yu Lee, Tomas Pfister
Abstract:Agent harnesses are executable code that specifies the prompts, tools, and control flow around a language model. A single global harness can struggle with heterogeneous tasks that require different execution strategies. In this work, we introduce Object-Oriented Harness (OO-Harness), a harness optimization framework that evolves a hierarchy of executable harness classes using object-oriented design. During training, an evolution agent uses execution records to create a family of specialized harnesses, each implemented as a subclass of the base or an existing specialized harness via inheritance. At test time, a harness router uses routing lessons accumulated during training to select a specific harness for each task. Across three long-horizon benchmarks spanning business workflows, workspace tasks, and tool use, OO-Harness achieves the best performance among the compared methods, with an absolute improvement of up to 15% over the base. Extensive ablation studies examine the contributions of its core components.
| Comments: | 20 pages, 10 figures |
| Subjects: | Computation and Language (cs.CL); Software Engineering (cs.SE) |
| Cite as: | arXiv:2610.28380 [cs.CL] |
| (or arXiv:2610.28380v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28380 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Gaotang Li [view email]
[v1]
Tue, 20 Oct 2026 15:32:52 UTC (4,910 KB)
来源:arXiv cs.CL · arxiv.org