Mobile-Env: Building Qualified Evaluation Benchmarks for LLM-GUI Interaction
This addresses the need for better evaluation tools in AI-human interaction research, though it is incremental as it builds on existing benchmark challenges.
The authors tackled the lack of reliable benchmarks for evaluating LLM-based GUI agents by introducing Mobile-Env, a toolkit for creating Android GUI benchmarks, and found that even advanced models like GPT-4V struggle with simple human tasks, highlighting a significant performance gap.
The Graphical User Interface (GUI) is pivotal for human interaction with the digital world, enabling efficient device control and the completion of complex tasks. Recent progress in Large Language Models (LLMs) and Vision Language Models (VLMs) offers the chance to create advanced GUI agents. To ensure their effectiveness, there's a pressing need for qualified benchmarks that provide trustworthy and reproducible evaluations -- a challenge current benchmarks often fail to address. To tackle this issue, we introduce Mobile-Env, a comprehensive toolkit tailored for creating GUI benchmarks in the Android mobile environment. Mobile-Env offers an isolated and controllable setting for reliable evaluations, and accommodates intermediate instructions and rewards to reflect real-world usage more naturally. Utilizing Mobile-Env, we collect an open-world task set across various real-world apps and a fixed world set, WikiHow, which captures a significant amount of dynamic online contents for fully controllable and reproducible evaluation. We conduct comprehensive evaluations of LLM agents using these benchmarks. Our findings reveal that even advanced models (e.g., GPT-4V and LLaMA-3) struggle with tasks that are relatively simple for humans. This highlights a crucial gap in current models and underscores the importance of developing more capable foundation models and more effective GUI agent frameworks.