Adrian Rebmann
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Adrian Rebmann has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
- Han van der Aa
University of Vienna
11 shared publications Han van der Aa
4 shared publications- Jana-Rebecca Rehse
University of Mannheim
4 shared publications - Timotheus Kampik
Umeå University
4 shared publications - Christian Warmuth
SAP Signavio
3 shared publications - Michael Grohs
University of Mannheim
3 shared publications
Latest Publications
- 27 September 2026
- Published by Springer Nature
- p. 211-227
Organizational Memory for Agentic Business Process Execution
- 27 September 2026
- Published by Springer Nature
- p. 211-227
Organizational Memory for Agentic Business Process Execution
- 27 September 2026
- Published by Springer Nature
- p. 179-194
Natural Language Querying for Large Declarative Rule Repositories
- 27 September 2026
- Published by Springer Nature
- p. 179-194
Natural Language Querying for Large Declarative Rule Repositories
- 27 September 2026
- Published by Springer Nature
- p. 755-772
Agent Behavior Mining: Generative AI Agent Governance in Business Processes
- 27 September 2026
- Published by Springer Nature
- p. 755-772
Agent Behavior Mining: Generative AI Agent Governance in Business Processes
- 27 September 2026
- Published by Springer Nature
- p. 651-669
Guiding Process Harmonization: A Classification Scheme for Differences Between BPMN Models
- 27 September 2026
- Published by Springer Nature
- p. 651-669
Guiding Process Harmonization: A Classification Scheme for Differences Between BPMN Models
- 2 August 2026
- Published by Springer Nature
- p. 85-97
Towards an Automated Assessment of Deviation Desirability
- 2 August 2026
- Published by Springer Nature
- p. 85-97
Towards an Automated Assessment of Deviation Desirability
- 1 April 2026
- Published by Springer Nature
- p. 405-411
LLMs for BPM—From First-Wave Features to Process Understanding and Lasting Impact(?)
- 1 April 2026
- Published by Springer Nature
- p. 405-411
LLMs for BPM—From First-Wave Features to Process Understanding and Lasting Impact(?)
- 27 August 2025
- Published by Springer Nature
- p. 183-203
Detecting Undesired Process Behavior by Means of Retrieval Augmented Generation
- 27 August 2025
- Published by Springer Nature
- p. 183-203
Detecting Undesired Process Behavior by Means of Retrieval Augmented Generation
- Open Access
LLMs that Understand Processes: Instruction-tuning for Semantics-Aware Process Mining
Process mining is increasingly using textual information associated with events to tackle tasks such as anomaly detection and process discovery. Such semantics-aware process mining focuses on what behavior should be possible in a process (i.e., expectations), thus providing an important complement to traditional, frequency-based techniques that focus on recorded behavior (i.e., reality). Large Language Models (LLMs) provide a powerful means for tackling semantics-aware tasks. However, the best performance is so far achieved through task-specific fine-tuning, which is computationally intensive and results in models that can only handle one specific task. To overcome this lack of generalization, we use this paper to investigate the potential of instruction-tuning for semantics-aware process mining. The idea of instruction-tuning here is to expose an LLM to prompt-answer pairs for different tasks, e.g., anomaly detection and next-activity prediction, making it more familiar with process mining, thus allowing it to also perform better at unseen tasks, such as process discovery. Our findings demonstrate a varied impact of instruction-tuning: while performance considerably improved on process discovery and prediction tasks, it varies across models on anomaly detection tasks, highlighting that the selection of tasks for instruction-tuning is critical to achieving desired outcomes.
Open AccessLLMs that Understand Processes: Instruction-tuning for Semantics-Aware Process Mining
Process mining is increasingly using textual information associated with events to tackle tasks such as anomaly detection and process discovery. Such semantics-aware process mining focuses on what behavior should be possible in a process (i.e., expectations), thus providing an important complement to traditional, frequency-based techniques that focus on recorded behavior (i.e., reality). Large Language Models (LLMs) provide a powerful means for tackling semantics-aware tasks. However, the best performance is so far achieved through task-specific fine-tuning, which is computationally intensive and results in models that can only handle one specific task. To overcome this lack of generalization, we use this paper to investigate the potential of instruction-tuning for semantics-aware process mining. The idea of instruction-tuning here is to expose an LLM to prompt-answer pairs for different tasks, e.g., anomaly detection and next-activity prediction, making it more familiar with process mining, thus allowing it to also perform better at unseen tasks, such as process discovery. Our findings demonstrate a varied impact of instruction-tuning: while performance considerably improved on process discovery and prediction tasks, it varies across models on anomaly detection tasks, highlighting that the selection of tasks for instruction-tuning is critical to achieving desired outcomes.
- Open Access
- 2 June 2025
- Published by Springer Nature in Process Science
- Vol. 2 (1) , 1-29
On the potential of large language models to solve semantics-aware process mining tasks
- 2 June 2025
- Published by Springer Nature in Process Science
- Vol. 2 (1) , 1-29
Open AccessOn the potential of large language models to solve semantics-aware process mining tasks
- Open Access
Detecting Undesired Process Behavior by Means of Retrieval Augmented Generation
Conformance checking techniques detect undesired process behavior by comparing process executions that are recorded in event logs to desired behavior that is captured in a dedicated process model. If such models are not available, conformance checking techniques are not applicable, but organizations might still be interested in detecting undesired behavior in their processes. To enable this, existing approaches use Large Language Models (LLMs), assuming that they can learn to distinguish desired from undesired behavior through fine-tuning. However, fine-tuning is highly resource-intensive and the fine-tuned LLMs often do not generalize well. To address these limitations, we propose an approach that requires neither a dedicated process model nor resource-intensive fine-tuning to detect undesired process behavior. Instead, we use Retrieval Augmented Generation (RAG) to provide an LLM with direct access to a knowledge base that contains both desired and undesired process behavior from other processes, assuming that the LLM can transfer this knowledge to the process at hand. Our evaluation shows that our approach outperforms fine-tuned LLMs in detecting undesired behavior, demonstrating that RAG is a viable alternative to resource-intensive fine-tuning, particularly when enriched with relevant context from the event log, such as frequent traces and activities.
Open AccessDetecting Undesired Process Behavior by Means of Retrieval Augmented Generation
Conformance checking techniques detect undesired process behavior by comparing process executions that are recorded in event logs to desired behavior that is captured in a dedicated process model. If such models are not available, conformance checking techniques are not applicable, but organizations might still be interested in detecting undesired behavior in their processes. To enable this, existing approaches use Large Language Models (LLMs), assuming that they can learn to distinguish desired from undesired behavior through fine-tuning. However, fine-tuning is highly resource-intensive and the fine-tuned LLMs often do not generalize well. To address these limitations, we propose an approach that requires neither a dedicated process model nor resource-intensive fine-tuning to detect undesired process behavior. Instead, we use Retrieval Augmented Generation (RAG) to provide an LLM with direct access to a knowledge base that contains both desired and undesired process behavior from other processes, assuming that the LLM can transfer this knowledge to the process at hand. Our evaluation shows that our approach outperforms fine-tuned LLMs in detecting undesired behavior, demonstrating that RAG is a viable alternative to resource-intensive fine-tuning, particularly when enriched with relevant context from the event log, such as frequent traces and activities.