Fine-tuned Llama-3 and Mistral reach macro F1 0.69 to 0.88 and fitness 0.80 to 0.84 on five new semantics-aware process mining benchmarks, while few-shot in-context learning stays near random.
Information Systems 102, 101824 (2021)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DB 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Potential of Large Language Models to Solve Semantics-Aware Process Mining Tasks
Fine-tuned Llama-3 and Mistral reach macro F1 0.69 to 0.88 and fitness 0.80 to 0.84 on five new semantics-aware process mining benchmarks, while few-shot in-context learning stays near random.