LLMs that Understand Processes: Instruction-tuning for Semantics-Aware Process Mining

TL;DR

This paper explores instruction-tuning of Large Language Models to enhance their ability in semantics-aware process mining tasks, aiming for better generalization across multiple process mining applications.

Contribution

It introduces an instruction-tuning approach for LLMs in process mining, enabling models to perform well on multiple tasks without task-specific fine-tuning.

Findings

Performance improved on process discovery and prediction tasks.

Impact on anomaly detection varies across models.

Task selection for instruction-tuning is crucial.

Abstract

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.