![[2506.02529] Automated Web Application Testing: End-to-End Test Case Generation with Large Language Models and Screen Transition Graphs](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2506.02529] Automated Web Application Testing: End-to-End Test Case Generation with Large Language Models and Screen Transition Graphs
Summary
This paper presents an automated system for generating end-to-end test cases for web applications using large language models and screen transition graphs, enhancing testing efficiency.
Why It Matters
As web applications become increasingly complex, ensuring their reliability is crucial. This research addresses the challenges of automating web application testing, particularly in navigation and form interactions, potentially leading to more robust software development practices.
Key Takeaways
- Introduces a novel integration of graph structures and LLMs for site navigation testing.
- Employs state graphs to automate form-filling test cases effectively.
- Provides a comprehensive dataset for evaluating form-interaction testing.
- Demonstrates improved test coverage and robustness through experimental results.
- Addresses the limitations of current automated testing methods in dynamic web environments.
Computer Science > Software Engineering arXiv:2506.02529 (cs) [Submitted on 3 Jun 2025 (v1), last revised 19 Feb 2026 (this version, v2)] Title:Automated Web Application Testing: End-to-End Test Case Generation with Large Language Models and Screen Transition Graphs Authors:Nguyen-Khang Le, Quan Minh Bui, Minh Ngoc Nguyen, Hiep Nguyen, Trung Vo, Son T. Luu, Shoshin Nomura, Minh Le Nguyen View a PDF of the paper titled Automated Web Application Testing: End-to-End Test Case Generation with Large Language Models and Screen Transition Graphs, by Nguyen-Khang Le and 7 other authors View PDF HTML (experimental) Abstract:Web applications are critical to modern software ecosystems, yet ensuring their reliability remains challenging due to the complexity and dynamic nature of web interfaces. Recent advances in large language models (LLMs) have shown promise in automating complex tasks, but limitations persist in handling dynamic navigation flows and complex form interactions. This paper presents an automated system for generating test cases for two key aspects of web application testing: site navigation and form filling. For site navigation, the system employs screen transition graphs and LLMs to model navigation flows and generate test scenarios. For form filling, it uses state graphs to handle conditional forms and automates Selenium script generation. Key contributions include: (1) a novel integration of graph structures and LLMs for site navigation testing, (2) a state graph-b...
