The field of web application testing is moving towards more robust and automated methods, with a focus on improving test coverage and reliability. Recent developments have seen the integration of large language models (LLMs) and graph structures to model navigation flows and generate test scenarios. Gamification strategies are also being employed to boost user engagement and guide users towards the adoption of best practices. Additionally, novel methods for inferring component abstractions and generating scenario-based GUI tests are being introduced, enabling more effective and accurate testing. Noteworthy papers include:
- VISCA, which introduces a novel method for transforming webpages into hierarchical, semantically rich component abstractions, achieving an average feature coverage of 92%.
- LLM-Guided Scenario-based GUI Testing, which proposes a novel approach involving multi-agent collaboration to generate scenario-based GUI tests guided by LLMs.