用图神经网络与大模型结合，预测智能工作流性能表现。
GLOW: Graph-Language Co-Reasoning for Agentic Workflow Performance Prediction
- 将图结构与大模型推理融合，捕捉工作流的拓扑与语义特征。
- 在FLORA-Bench上准确率超越现有方法，排名效果更优。
- 适合关注自动化工作流评估与优化的研究者与工程师。
智能工作流（AWs）已成为解决复杂任务的有前景范式，但其自动生成的可扩展性受限于执行评估的高成本和延迟。现有性能预测方法作为替代方案，却无法同时捕捉工作流中复杂的拓扑依赖关系和深层语义逻辑。为此，我们提出GLOW框架，通过结合图神经网络（GNN）的图结构建模能力与大语言模型（LLM）的推理能力，实现统一的性能预测。具体地，我们引入一种面向图任务指令微调的图导向大模型，以提取具备拓扑感知的语义特征，并将其与GNN编码的结构表征融合。进一步采用对比对齐策略，优化潜在空间以区分高质量工作流。在FLORA-Bench上的大量实验表明，GLOW在预测准确率和排序效用方面均优于当前最优基线方法。
原文摘要 · Abstract (English)
Agentic Workflows (AWs) have emerged as a promising paradigm for solving complex tasks. However, the scalability of automating their generation is severely constrained by the high cost and latency of execution-based evaluation. Existing AW performance prediction methods act as surrogates but fail to simultaneously capture the intricate topological dependencies and the deep semantic logic embedded in AWs. To address this limitation, we propose GLOW, a unified framework for AW performance prediction that combines the graph-structure modeling capabilities of GNNs with the reasoning power of LLMs. Specifically, we introduce a graph-oriented LLM, instruction-tuned on graph tasks, to extract topologically aware semantic features, which are fused with GNN-encoded structural representations. A contrastive alignment strategy further refines the latent space to distinguish high-quality AWs. Extensive experiments on FLORA-Bench show that GLOW outperforms state-of-the-art baselines in prediction accuracy and ranking utility.
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。