arxiv:2501.04575

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

Published on Jan 8

· Submitted by

pengxiang on Jan 9

Upvote

Authors:

Yuhang Liu ,

Pengxiang Li ,

Zishu Wei ,

Congkai Xie ,

Xueyu Hu ,

Xinchen Xu ,

Xiaotian Han ,

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 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. 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 https://github.com/Reallm-Labs/InfiGUIAgent.

View arXiv page View PDF Add to collection

Community

pengxiang

Paper author Paper submitter 1 day ago

InfiGUIAgent is a multimodal LLM-based GUI agent that improves performance through a two-stage supervised fine-tuning approach:

Stage 1 enhances basic skills (GUI understanding and grounding)
Stage 2 integrates hierarchical and expectation-reflection reasoning.

This training method enables "native reasoning abilities," showing competitive performance on GUI benchmarks and improving automated GUI interactions.