Summary of Xuat-copilot: Multi-agent Collaborative System For Automated User Acceptance Testing with Large Language Model, by Zhitao Wang et al.
XUAT-Copilot: Multi-Agent Collaborative System for Automated User Acceptance Testing with Large Language Model
by Zhitao Wang, Wei Wang, Zirao Li, Long Wang, Can Yi, Xinjie Xu, Luyang Cao, Hanjing Su, Shouzhi Chen, Jun Zhou
First submitted to arxiv on: 5 Jan 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 In this paper, researchers aim to improve the automation level of the user acceptance testing (UAT) process for WeChat Pay, a prominent mobile payment application in China. The current system, XUAT, relies on human labor for test script generation. To address this limitation, the authors propose an LLM-powered multi-agent collaborative system, XUAT-Copilot, which leverages large language models (LLMs) to simulate human-like decision-making capabilities. The system consists of three agents responsible for action planning, state checking, and parameter selecting, as well as two additional modules for state sensing and case rewriting. Experimental studies show that the proposed system achieves comparable effectiveness to human testers, with a significant improvement in Pass@1 accuracy compared to single-agent architectures. Moreover, XUAT-Copilot has been successfully deployed in WeChat Pay’s formal testing environment, reducing the need for manual labor in daily development work. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making computers do more of the work when testing mobile apps like WeChat Pay. Right now, humans have to write test scripts, which takes a lot of time and effort. The researchers want to use special computer programs called large language models (LLMs) to help with this task. They created a system that uses multiple LLMs working together to make decisions and generate test commands. This system is like having human testers, but it’s faster and more efficient. In fact, the researchers tested their system and found it was just as good as human testers at finding problems in the app. Best of all, this system has already been used in real-life testing for WeChat Pay, which saves a lot of time and effort. |
