Have you ever wanted to test your mobile app or control iOS and Android devices with an LLM? You've probably encountered context problems due to the a

Search code, repositories, users, issues, pull requests...

submited by

Style Pass

2024-10-28 06:30:05

Have you ever wanted to test your mobile app or control iOS and Android devices with an LLM? You've probably encountered context problems due to the accessibility view being too long or just sending a screenshot to the LLM, which provides limited accuracy.

At Revyl, we use this approach to test mobile apps with LLMs. Our platform integrates resilient end-to-end tests using agentic LLMs with open telemetry tracing, offering proactive observability into your mobile app.

@inproceedings{shvoEtAl2021appbuddy, title={AppBuddy: Learning to Accomplish Tasks in Mobile Apps via Reinforcement Learning}, author={Maayan Shvo and Zhiming Hu and Rodrigo Toro Icarte and Iqbal Mohomed and Allan D. Jepson and Sheila A. McIlraith}, booktitle={Canadian Conference on Artificial Intelligence}, year={2021} }

@misc{google-research, title={Google Research}, author={Google}, year={2021}, howpublished={\url{https://github.com/Berrylcm/google-research}} }