The field of process industry automation and analysis is witnessing significant developments, driven by the integration of large language models, graph embeddings, and domain knowledge. Researchers are exploring innovative approaches to enhance the accuracy and efficiency of process modeling, control logic engineering, and process discovery. The use of graph-aware neighborhood contrastive learning methodologies and large language models is improving the performance of language models in domain-specific applications. Additionally, the development of automated workflows for generating control logic from natural language requirements is reducing human labor and increasing automation. The integration of domain knowledge into process discovery pipelines is also leading to more reliable and accurate process models. Noteworthy papers include: Spec2Control, which introduces a highly automated LLM workflow for generating graphical control logic, and Integrating Domain Knowledge into Process Discovery Using Large Language Models, which proposes an interactive framework for incorporating domain knowledge into process discovery using LLMs.