Avatar：利用大语言模型实现科学工作流的自主端到端编排
Avatar: Toward Autonomous End-to-End Orchestration of Scientific Workflows using LLMs
浏览论文内容
中文总结 AI 辅助
Avatar提出基于actor的架构，通过可插拔决策策略实现科学工作流自主编排，规则模式复现原生执行，LLM模式减少55%计算浪费和40%GPU繁忙时间。
中文摘要 AI 辅助
科学工作流管理系统（WMSs）能够自动化执行，但其编排依赖于固定的、手工调整的规则。大语言模型（LLM）智能体有望实现更自主的编排，但目前尚不清楚应在何处引入智能体推理、如何限制其风险，以及何时真正有益。我们提出了Avatar，一种基于actor的架构，包含编排器、执行器和溯源监控器。每个actor的决策策略通过一个适配器验证的动作目录实现可插拔（基于规则或基于LLM），因此传统控制和智能体控制可在不同WMSs上运行于同一核心。我们使用Academy框架实现了该架构，并在三个工作负载上评估了Avatar。我们观察到，Avatar的规则模式复现了原生执行，且单一未更改的核心运行了全部三个工作负载。此外，基于LLM的Avatar报告计算浪费减少了55%，GPU繁忙时间减少了40%。总体而言，我们设想Avatar是迈向能够推理自身编排而非遵循预设规则的工作流系统的一步。
英文摘要
Scientific workflow management (WMSs) systems automate execution, yet orchestrate using fixed, hand-tuned rules. LLM agents promise more autonomous orchestration, but it remains unclear where to introduce agentic reasoning, how to bound its risk, and when it actually helps. We present Avatar, an actor-based architecture comprising an orchestrator, an executor, and a provenance monitor. Each actor's decision policy is pluggable (rule-based or LLM-backed) via a single adapter-validated action catalog, so conventional and agentic control run on the same core across different WMSs. We present an implementation using the Academy framework and evaluate Avatar across three workloads. We observe that Avatar's rule mode reproduces native execution, with a single unchanged core running all three. Moreover, LLM-backed Avatar reports a reduction of compute wastage by $55\%$ and cuts GPU-busy time by $40\%$. Overall, we envision Avatar as a step toward workflow systems that reason about their own orchestration rather than follow pre-fixed rules.
发表机构
- University of Chicago（芝加哥大学）
- Argonne National Laboratory（阿贡国家实验室）
机构由 AI 辅助整理，请以论文原文为准。
