AUG: Automated UML Diagrams Generated by Large Language Models
by 佳 卢·Updated 1mo ago
80.0 MB1files
Available on 1 platform
Sign in to view source links and access this dataset
Description
figshare hosts a dataset supporting the AUG tool for automated UML diagram generation. The dataset likely contains evaluation data and benchmarks for UML class, use case, and sequence diagrams generated by the GLM4-9B model. The repository includes code, datasets, and evaluation data, with the dataset last updated on May 5, 2026.
Use Cases
Benchmarking automated UML generation tools based on the provided evaluation data
Training or fine-tuning language models for software engineering tasks based on the UML diagram corpus
Studying the accuracy of model-generated diagrams based on reported precision, recall, and F1 scores
Comparing model performance against tools like ChatGPT 4o based on the comparative evaluation results
Strengths
Evaluation results show precision, recall, and F1 scores of 78.68%, 67.37%, and 72.59%, respectively
A questionnaire study reported high user satisfaction and time efficiency for the tool
The dataset is openly available under a CC-BY-4.0 license
Limitations
Column-level documentation is absent; field semantics must be inferred after download
Row count is unknown, which may limit suitability assessment
The dataset is relatively small at 80.0 MB
Provenance
Source
figshare
Collection Method
Likely generated as part of the development and evaluation of the AUG tool.