Scopus

Generative Reasoning Meets Multimodal Evaluation: A Novel Approach to Activity and Package Diagram Comparison in UML Modeling

Năm XB 2026 Tạp chí / Hội thảo International Conference in Cybersecurity, IoT, Data Science, and Digital Forensics Vol 5, 114–127 Đơn vị CNTT DOI / Link https://doi.org/10.1007/978-3-032-19488-6_7 ↗

Tác giả

Van-Viet Nguyen ^✉ ; Huu-Khanh Nguyen ; Kim-Son Nguyen ; Minh-Hue Luong Thi ; Anh-Tu Bui ; Duc-Quang Vu ; The-Vinh Nguyen ^✉

Tóm tắt

Large Language Models (LLMs) have been utilized for many downstream tasks such as writing, reasoning and software engineering. However, only a fraction of studies focused on generating code, especially in the field of software requirements. Thus, this paper presents a unique framework that leverages generative reasoning and multimodal evaluation to compare two essential Unified Modeling Language (UML) diagram types: Activity Diagrams, representing behavioral aspects and Package Diagrams, representing structural organization. In the proposed approach, LLaMA 3.2-1B-Instruct was utilized to generate user-oriented prompts, while DeepSeek-R1-Distill-Qwen-32B performs reasoning and automatically produces PlantUML code for diagram construction. To ensure the reliability and accuracy of the generated diagrams, multimodal large models Qwen2.5-VL-3B-Instruct, LLaMA-3.2-11B-Vision-Instruct and Aya-Vision-8B are employed within the MMMU Benchmark for evaluation and scoring. Experimental results demonstrated that the combination of reasoning based generation and multimodal assessment provided significant improvements in diagram consistency, semantic correctness, and cross-model verification. The study contributes a novel methodology for automated UML modeling which highlighted the advantages and limitations of Activity versus Package diagrams when generated and validated through generative AI. This study not only contributes to the automation process in software design but also proposes a new framework for evaluating the ability of reasoning on diagrams in model-driven engineering.

Tài liệu tham khảo

[1] Booch, G., Rumbaugh, J., Jacobson, I.: The Unified Modeling Language User Guide, 2nd edn. Addison-Wesley Professional, Boston, MA, USA (2005)

[2] Zhang, Q., et al.: A survey on large language models for software engineering. arXiv preprint arXiv:2312.15223 (2023)

[3] Chen, M., et al.: Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374 (2021)

[4] Dehaerne, E., Dey, B., Halder, S., De Gendt, S., Meert, W.: Code generation using machine learning: a systematic review. IEEE Access 10, 82434–82455 (2022)

[5] Nguyen, T.-V., Phung, T.-N., Cuong, D.-D.: A bibliometric and thematic analysis of systematic reviews of artificial intelligence in education. In: International Conference on Advances in Information and Communication Technology, pp. 337–351. Springer (2024)

[6] Van Viet, N., et al.: Revolutionizing education: an extensive analysis of large language models integration 4(4), 10–21 (2024). https://irjstem.com/wp-content/uploads/2025/02/IRJSTEM-V4N4-2024-Paper02.pdf

[7] Khanh, N.H., Van, V.N., Vinh, N.T., Cong, N.H.: Phi-3 meets law: fine-tuning mini language models for legal document understanding. Res. Dev. Appl. Inf. Commun. Technol. 2024(3), 136–142 (2024). ISSN: 1859-3526

[8] de Faria, F.F.G., do Prado Leite, J.C.S.: Generating UML models from text: a systematic literature review. In: Proceedings of the 22nd Brazilian Symposium on Software Engineering, pp. 131–140 (2008)

[9] Guo, D., et al.: DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning. arXiv preprint arXiv:2501.12948 (2025)

[10] Wu, J., Gan, W., Chen, Z., Shicheng Wan, Z., Yu, P.S.: Multimodal large language models: a survey. In: 2023 IEEE International Conference on Big Data (BigData), pp. 2247–2256. IEEE (2023)

[11] Yue, X., et al.: MMMU: a massive multi-discipline multimodal understanding and reasoning benchmark for expert AGI. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 9556–9567 (2024)

[12] Wang, P., et al.: Qwen2-VL: Enhancing vision-language model’s perception of the world at any resolution. arXiv preprint arXiv:2409.12191 (2024)

[13] Touvron, H., et al.: LLaMA: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)

[14] Üstün, A., et al.: Aya model: an instruction finetuned open-access multilingual language model. In: Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 15894–15939, Bangkok, Thailand (2024). Association for Computational Linguistics

[15] Sommerville, I.: Software Engineering, 10th edn. Pearson, Boston, MA, USA (2016)

[16] Van Viet, N., The Vinh, N.: Large language models in software engineering: a systematic review and vision. J. Educ. Sustain. Innovation 2(2), 146–156 (2024)

[17] Meng, Y., Ban, A.: Automated UML class diagram generation from textual requirements using NLP techniques. JOIV: Int. J. Inform. Vis. 8(3-2), 1905–1915 (2024)

[18] Wei, J., Wang, X., Schuurmans, D., Le, Q.V., Zhou, D.: Chain-of-thought prompting elicits reasoning in large language models. In: Advances in Neural Information Processing Systems (NeurIPS), vol. 35, pp. 24824–24837 (2022)

[19] Wei, J., et al.: Chain-of-thought prompting elicits reasoning in large language models. In: Proceedings of the 36th International Conference on Neural Information Processing Systems, NIPS ’22, Red Hook, NY, USA (2022). Curran Associates Inc

[20] Rouabhia, D., Hadjadj, I.: Enhancing class diagram dynamics: A natural language approach with ChatGPT. arXiv preprint arXiv:2406.11002 (2024)

[21] Munde, A.R.: Evaluation of vision language model on UML diagrams

[22] Touvron, H., et al.: LLaMA: Open and efficient foundation language models (2023)

[23] DeepSeek-AI. DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning (2025)

[24] Wang, P., et al.: Qwen2-VL: Enhancing vision-language model’s perception of the world at any resolution. arXiv preprint arXiv:2409.12191 (2024)

[25] Bai, J., et al.: Qwen-VL: A versatile vision-language model for understanding, localization, text reading, and beyond. arXiv preprint arXiv:2308.12966 (2023)

[26] Üstün, A., et al.: Aya model: An instruction finetuned open-access multilingual language model. arXiv preprint arXiv:2402.07827 (2024)

← Quay lại danh sách bài báo