让大模型长期执行任务不崩溃，靠文件管理状态
InfiAgent: An Infinite-Horizon Framework for General-Purpose Autonomous Agents
- 把长期状态存成文件，每次只读最近动作和快照
- 80篇论文综述任务中覆盖率达基线3倍以上
- 无需微调，20B开源模型媲美大厂闭源系统
大语言模型代理能推理和用工具，但在长周期任务中常因上下文无限增长和误差累积而失效。现有方法如上下文压缩或检索增强会牺牲信息保真度与推理稳定性。我们提出InfiAgent，一种通用框架，通过将持久状态外置于文件中心的状态抽象，使代理的推理上下文始终保持有限，无论任务时长如何。每一步，代理从工作区状态快照加固定窗口的近期动作重建上下文。在DeepResearch和80篇论文文献综述任务上的实验表明，无需任务特定微调，使用20B开源模型的InfiAgent在性能上可与更大规模的专有系统竞争，并显著优于以上下文为中心的基线，在长周期任务中的覆盖率高出三倍以上。结果支持显式状态外化作为稳定长周期代理的实际基础。Github仓库：https://github.com/ChenglinPoly/infiAgent
原文摘要 · Abstract (English)
LLM agents can reason and use tools, but they often break down on long-horizon tasks due to unbounded context growth and accumulated errors. Common remedies such as context compression or retrieval-augmented prompting introduce trade-offs between information fidelity and reasoning stability. We present InfiAgent, a general-purpose framework that keeps the agent's reasoning context strictly bounded regardless of task duration by externalizing persistent state into a file-centric state abstraction. At each step, the agent reconstructs context from a workspace state snapshot plus a fixed window of recent actions. Experiments on DeepResearch and an 80-paper literature review task show that, without task-specific fine-tuning, InfiAgent with a 20B open-source model is competitive with larger proprietary systems and maintains substantially higher long-horizon coverage than context-centric baselines. These results support explicit state externalization as a practical foundation for stable long-horizon agents. Github Repo:https://github.com/ChenglinPoly/infiAgent
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。