InfiGUIAgent: A Multimodal Generalist GUI Agent with Native Reasoning and Reflection

TL;DR

InfiGUIAgent is a multimodal GUI agent that leverages a two-stage training process to develop native reasoning and reflection skills, significantly improving multi-step GUI task automation.

Contribution

The paper introduces InfiGUIAgent, a novel MLLM-based GUI agent with a two-stage fine-tuning pipeline for native reasoning and reflection capabilities, surpassing existing models.

Findings

Achieves competitive performance on GUI benchmarks.

Enhances multi-step reasoning and grounding in GUI tasks.

Demonstrates the effectiveness of hierarchical reasoning skills.

Abstract

Graphical User Interface (GUI) Agents, powered by multimodal large language models (MLLMs), have shown great potential for task automation on computing devices such as computers and mobile phones. However, existing agents face challenges in multi-step reasoning and reliance on textual annotations, limiting their effectiveness. We introduce \textit{InfiGUIAgent}, an MLLM-based GUI Agent trained with a two-stage supervised fine-tuning pipeline. Stage 1 enhances fundamental skills such as GUI understanding and grounding, while Stage 2 integrates hierarchical reasoning and expectation-reflection reasoning skills using synthesized data to enable native reasoning abilities of the agents. \textit{InfiGUIAgent} achieves competitive performance on several GUI benchmarks, highlighting the impact of native reasoning skills in enhancing GUI interaction for automation tasks. Resources are available at \url{https://github.com/Reallm-Labs/InfiGUIAgent}.