用知识图谱和两阶段模型生成合规且高效的业务流程，医疗编码任务提升超29%。
Opus: A Large Work Model for Complex Workflow Generation
- 基于意图生成流程，结合知识图谱与大工作模型构建有上下文的流程图。
- 在医疗编码任务中，大模型和小模型分别提升38%和29%的生成效果。
- 适合需要高合规性、可审计流程的外包与企业自动化场景。
本文提出Opus，一种面向复杂业务流程外包（BPO）场景的新型框架，旨在通过生成与优化工作流实现成本降低与质量提升，同时遵守行业规范与运营约束。工作流从“意图”生成，即客户输入、输出与流程上下文的一致对齐，以有向无环图（DAG）形式表示，节点为包含工具调用与人工评审等可执行指令的任务序列。采用两阶段方法：第一阶段利用大型工作模型（LWM）结合工作知识图谱（WKG）生成工作流；第二阶段将工作流转化为工作流图（WFG），通过路径优化确定最优方案。实验表明，现有大语言模型（LLM）在可靠获取详细流程数据及生成行业合规工作流方面存在困难。本文核心贡献包括：将工作知识图谱（WKG）融入大型工作模型（LWM），实现上下文感知、语义对齐、结构化且可审计的工作流生成；提出两阶段方法，融合意图驱动生成与图优化；并发布Opus Alpha 1 Large与Opus Alpha 1 Small，在医疗编码任务中分别优于现有SOTA LLM 38%与29%。
原文摘要 · Abstract (English)
This paper introduces Opus, a novel framework for generating and optimizing Workflows tailored to complex Business Process Outsourcing (BPO) use cases, focusing on cost reduction and quality enhancement while adhering to established industry processes and operational constraints. Our approach generates executable Workflows from Intention, defined as the alignment of Client Input, Client Output, and Process Context. These Workflows are represented as Directed Acyclic Graphs (DAGs), with nodes as Tasks consisting of sequences of executable Instructions, including tools and human expert reviews. We adopt a two-phase methodology: Workflow Generation and Workflow Optimization. In the Generation phase, Workflows are generated using a Large Work Model (LWM) informed by a Work Knowledge Graph (WKG) that encodes domain-specific procedural and operational knowledge. In the Optimization phase, Workflows are transformed into Workflow Graphs (WFGs), where optimal Workflows are determined through path optimization. Our experiments demonstrate that state-of-the-art Large Language Models (LLMs) face challenges in reliably retrieving detailed process data as well as generating industry-compliant workflows. The key contributions of this paper include integrating a Work Knowledge Graph (WKG) into a Large Work Model (LWM) to enable the generation of context-aware, semantically aligned, structured and auditable Workflows. It further introduces a two-phase approach that combines Workflow Generation from Intention with graph-based Workflow Optimization. Finally, we present Opus Alpha 1 Large and Opus Alpha 1 Small that outperform state-of-the-art LLMs by 38% and 29% respectively in Workflow Generation for a Medical Coding use case.
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。