InfiGUIAgent: a novel multimodal generalist GUI agent with native reasoning and reflection
The development of graphical user interface (GUI) agents faces two key challenges that hinder their effectiveness. First, existing agents lack ...
Tag: GUI
ScreenSpot-Pro: The First Benchmark Driving Multimodal LLMs Towards High-Resolution Professional GUI Agent and Computer Usage Environments
GUI agents face three critical challenges in professional environments: (1) the increased complexity of professional applications compared to general-purpose software, ...
Meet Android Agent Arena (A3): a complete, self-contained online evaluation system for GUI agents
The development of large language models (LLM) has significantly advanced artificial intelligence (ai) in several fields. Among these advances, mobile ...
OS-Genesis: a novel GUI data synthesis pipeline that reverses the conventional trajectory collection process
Designing GUI agents that perform human-like tasks in graphical user interfaces faces a critical hurdle: collecting high-quality trajectory data for ...
AutoDroid-V2: Leveraging Small Language Models for Automated Mobile GUI Control
Large Language Models (LLM) and Vision Language Models (VLM) have revolutionized the automation of mobile device control through natural language ...
Salesforce AI Research Introduces AGUVIS: A Unified Pure Vision Framework Transforming Autonomous GUI Interaction Across Platforms
Graphical user interfaces (GUIs) play a critical role in human-computer interaction, providing the means through which users perform tasks across ...
UGround: A Universal GUI Visual Connection Model Developed Using Large-Scale Web-Based Synthetic Data
Graphical user interface (GUI) agents are crucial for automating interactions within digital environments, similar to how humans operate software using ...
Researchers from Tsinghua University and Zhipu AI present CogAgent: a revolutionary visual language model for enhanced GUI interaction
The research has its roots in the field of visual language models (VLM), focusing particularly on its application in graphical ...
Recommended.
