Summary of Gui Testing Arena: a Unified Benchmark For Advancing Autonomous Gui Testing Agent, by Kangjia Zhao et al.
GUI Testing Arena: A Unified Benchmark for Advancing Autonomous GUI Testing Agent
by Kangjia Zhao, Jiahui Song, Leigang Sha, Haozhan Shen, Zhi Chen, Tiancheng Zhao, Xiubo Liang, Jianwei Yin
First submitted to arxiv on: 24 Dec 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a comprehensive environment for evaluating the entire process of automated GUI testing, GTArena. It divides the testing process into three subtasks: test intention generation, test task execution, and GUI defect detection. A benchmark dataset is constructed based on these subtasks to conduct a comprehensive evaluation. The performance of different models is evaluated using real mobile applications, artificially injected defects, and synthetic data. Additionally, the paper proposes a method to explore the correlation between multimodal language model performance in specific scenarios and their general capabilities. Experimental results show that even advanced models struggle across all sub-tasks of automated GUI testing, highlighting a significant gap between current capabilities and practical applicability. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Automated GUI testing is an important area of research. The paper proposes a new way to evaluate this process called GTArena. It breaks down the evaluation into three parts: understanding what tests need to be run, running the tests, and finding any problems with the user interface. A special dataset is created to test different models and see how well they do. The results show that even the best models don’t do well across all parts of GUI testing, which means there’s still a lot of work to be done. |
Keywords
» Artificial intelligence » Language model » Synthetic data
