HarnessBridge: 用于LLM智能体框架的可学习双向控制器
HarnessBridge: Learnable Bidirectional Controller for LLM Agent Harness
- University of California, Los Angeles（加州大学洛杉矶分校）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出HarnessBridge，一种轻量级可学习框架控制器，通过双向投影参数化智能体-环境接口，减少令牌使用和轨迹长度，并泛化到更大模型。
AI中文摘要：
大型语言模型越来越多地被部署为用于长周期任务的智能体，但其性能不仅受模型能力和环境设计的影响，还受调节智能体-环境交互的框架的影响。现有的框架大多是手动设计的，随着轨迹变长和交互变得更加复杂，它们难以扩展。在这项工作中，我们探究框架是否可以通过一个可学习的即插即用模块生成，该模块可以以端到端的方式进行训练。我们引入了HarnessBridge，一种轻量级可学习框架控制器，它将智能体-环境接口参数化为双向投影。HarnessBridge学习两个双向投影：观测投影，将原始轨迹提炼为紧凑的、与决策相关的状态；以及动作投影，将提议的动作转换为可执行的转换或基于轨迹的拒绝。我们在框架监督数据集上通过统一指令调优训练HarnessBridge。在Terminal-Bench~2.0和SWE-bench Verified上，HarnessBridge匹配或超越了强大的专用框架，同时大幅减少了令牌使用和轨迹长度，并从较小的生成器泛化到较大的商业模型。
英文摘要：
Large language models are increasingly deployed as agents for long-horizon tasks, yet their performance is shaped not only by model capability and environment design, but also by the harness that mediates agent--environment interaction. Existing harnesses are largely manually engineered, making them difficult to scale as trajectories grow longer and interactions become more complex. In this work, we ask whether harness can be generated by a learnable plug-in module that can be trained in an end-to-end fashion. We introduce HarnessBridge, a lightweight learnable harness controller that parameterizes the agent--environment interface as a bidirectional projection. HarnessBridge learns two bidirectional projections: observation projection, which distills raw trajectories into compact, decision-relevant states, and action projection, which converts proposed actions into executable transitions or trajectory-grounded rejections. We train HarnessBridge on a harness supervision dataset via unified instruction tuning. On Terminal-Bench~2.0 and SWE-bench Verified, HarnessBridge matches or surpasses strong specialized harnesses while substantially reducing token usage and trajectory length, and generalizes from smaller generators to larger commercial models.
