Reproduce “Large Process Models: A Vision for Business Process Management in the Age of Generative AI” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Timotheus Kampik, Christian Warmuth, Adrian Rebmann, Ron Agam, Lukas N.P. Egger, Andreas Gerber, Johannes Hoffart, Jonas Kolk and 7 more
Researchers from SAP and several universities propose Large Process Models (LPMs), a neuro-symbolic framework that integrates generative AI, particularly large language models, with traditional Business Process Management (BPM) methods. This framework aims to enhance process design, analysis, and execution by fusing diverse process data and knowledge, making BPM more accessible, observable, and adaptable while addressing the limitations of both purely symbolic and purely statistical AI approaches.
Key insights
- Fusing the statistical correlation power of generative AI with the analytical precision and reliability of symbolic knowledge-based systems is crucial for trustworthy and effective BPM.
- Context-specificity, achieved through fine-tuning LLMs with domain, industry, and organizational knowledge, is essential for tailoring process insights and recommendations.
- Iterative development with continuous human feedback is vital to ensure the usefulness, relevance, and compliance of LPM inferences, especially given the stochastic nature of generative models.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.