Automated Soap Opera Testing Directed by LLMs and Scenario Knowledge: Feasibility, Challenges, and Road Ahead

TL;DR

This paper investigates the potential of automating soap opera testing using large language models and scenario knowledge graphs, highlighting current capabilities, challenges, and future directions for large-scale, automated exploratory testing.

Contribution

It introduces a multi-agent system leveraging LLMs and a Scenario Knowledge Graph to automate soap opera testing, addressing manual testing limitations and outlining future research pathways.

Findings

Automated testing shows promise but lags behind manual testing in scenario coverage.

The system can identify potential bugs but struggles with complex scenario boundaries.

Future work should focus on neural-symbolic integration and human-AI collaboration.

Abstract

Exploratory testing (ET) harnesses tester's knowledge, creativity, and experience to create varying tests that uncover unexpected bugs from the end-user's perspective. Although ET has proven effective in system-level testing of interactive systems, the need for manual execution has hindered large-scale adoption. In this work, we explore the feasibility, challenges and road ahead of automated scenario-based ET (a.k.a soap opera testing). We conduct a formative study, identifying key insights for effective manual soap opera testing and challenges in automating the process. We then develop a multi-agent system leveraging LLMs and a Scenario Knowledge Graph (SKG) to automate soap opera testing. The system consists of three multi-modal agents, Planner, Player, and Detector that collaborate to execute tests and identify potential bugs. Experimental results demonstrate the potential of automated soap opera testing, but there remains a significant gap compared to manual execution, especially under-explored scenario boundaries and incorrectly identified bugs. Based on the observation, we envision road ahead for the future of automated soap opera testing, focusing on three key aspects: the synergy of neural and symbolic approaches, human-AI co-learning, and the integration of soap opera testing with broader software engineering practices. These insights aim to guide and inspire the future research.