用于从自然语言生成 BPMN 的多层 AI 流水线
摘要
一篇论文提出了一种多层人工智能流水线,通过结合微调后的 RoBERTa 实体提取器、基于大语言模型的控制流关系提取和本体填充,将自然语言流程描述转换为 BPMN 图,并报告其关系提取准确率达到 90.7%。
arXiv:2611.18157v1 Announce Type: new
Abstract: The automated transformation of unstructured natural language process descriptions into Business Process Model and Notation (BPMN) diagrams remains a challenging task, due to linguistic ambiguity and the difficulty to detect non-adjacent control-flow relationships. This paper proposes a solution for generating BPMN models from natural language descriptions through a multi-layered pipeline that integrates transformer-based architectures, Large Language Models and ontology population. The proposed method starts with natural language descriptions and separates textual processing, control-flow relation extraction, and structural validation into dedicated components. Entity extraction is performed using a fine-tuned RoBERTa-based transformer model while a control-flow engine captures non-adjacent relationships through LLM-based inference. To improve generalization under limited available data, the approach uses a dynamic few-shot strategy with con- trolled template input construction, enabling improved generalization under limited data availability. Additionally, an ontology population mechanism is introduced to represent the generated process models as semantic knowledge structures. For evaluation purposes we have used from the literature and industry reference processes. Additionally, we have validated the proposed method on a real-world heart disease patients transitional care process, capturing practical constraints and multi-actor collaboration, and serving as an expert-defined benchmark scenario. Experimental results demonstrate strong performance in the relation extraction task, achieving an accuracy of 90.7%, as well as maintaining structural alignment between the generated and reference BPMN models across multiple evaluation scenarios scenarios.
查看缓存全文
缓存时间: 2026/11/16 08:27
# A Multi-Layered AI Pipeline for BPMN Generation from Natural Language Source: [https://arxiv.org/abs/2611.18157](https://arxiv.org/abs/2611.18157) [View PDF](https://arxiv.org/pdf/2611.18157) > Abstract:The automated transformation of unstructured natural language process descriptions into Business Process Model and Notation \(BPMN\) diagrams remains a challenging task, due to linguistic ambiguity and the difficulty to detect non\-adjacent control\-flow relationships\. This paper proposes a solution for generating BPMN models from natural language descriptions through a multi\-layered pipeline that integrates transformer\-based architectures, Large Language Models and ontology population\. The proposed method starts with natural language descriptions and separates textual processing, control\-flow relation extraction, and structural validation into dedicated components\. Entity extraction is performed using a fine\-tuned RoBERTa\-based transformer model while a control\-flow engine captures non\-adjacent relationships through LLM\-based inference\. To improve generalization under limited available data, the approach uses a dynamic few\-shot strategy with con\- trolled template input construction, enabling improved generalization under limited data availability\. Additionally, an ontology population mechanism is introduced to represent the generated process models as semantic knowledge structures\. For evaluation purposes we have used from the literature and industry reference processes\. Additionally, we have validated the proposed method on a real\-world heart disease patients transitional care process, capturing practical constraints and multi\-actor collaboration, and serving as an expert\-defined benchmark scenario\. Experimental results demonstrate strong performance in the relation extraction task, achieving an accuracy of 90\.7%, as well as maintaining structural alignment between the generated and reference BPMN models across multiple evaluation scenarios scenarios\. ## Submission history From: Alexandru Miron Rancea \[[view email](https://arxiv.org/show-email/8c279308/2611.18157)\] **\[v1\]**Fri, 13 Nov 2026 13:31:08 UTC \(1,776 KB\)
相似文章
自然语言到SQL翻译的螺母与螺栓:模型流水线优化方法及其交互的系统性分析
本文系统地分析了多个用于自然语言到SQL翻译的流水线扩展组件,包括中间表示、合成数据、预处理和重排器,使用SmBoP和RASAT架构,发现它们的交互比简单组合所有组件更重要。
将文档AI投入生产:面向OCR与LLM管道的微服务架构
本文提出了一种面向生产级文档AI管道的微服务架构,该架构结合了分类、OCR和LLM提取,分享了设计决策和批量分析洞察,揭示了OCR(而非LLM解析)主导了延迟。
LLM在工艺流程图工程中的应用:从最优PFD到经过验证的P&ID
本文介绍了P&ID Pilot,一个端到端的AI流水线,结合遗传算法和LLM来生成最优的工艺流程图(PFD),并将其转换为经过验证的管道和仪表图(P&ID),从而减少人工工程工作量。
MADP: 面向可持续文档处理的多智能体流水线,带有人机回环
MADP是一种用于企业文档处理的多智能体架构,结合了深度学习和大型语言模型,并采用人机回环验证,实现了97%的自动化率以及资源消耗的大幅降低。
用于高中成绩单自动处理的多智能体AI系统:大规模协作式文档分析
本文介绍了一个多智能体AI系统,包含由协调智能体协调的专业智能体(模式识别、语义分析、视觉智能),用于自动处理多样化的高中成绩单。在来自13个州的40份成绩单测试集上,实现了96.7%的准确率和每份45秒的处理速度。