Researchers have developed UI-MOPD, a novel method for training GUI agents that can operate across multiple platforms. This approach addresses the scarcity of cross-platform data and the challenge of maintaining performance on existing platforms while adapting to new ones. UI-MOPD utilizes multi-teacher on-policy distillation, dynamically selecting a platform-specific teacher to transfer knowledge and prevent catastrophic forgetting. Experiments on OSWorld and MobileWorld demonstrated the method's effectiveness, achieving task success rates of 38.2% and 12.0% respectively. AI
IMPACT Enables more versatile and adaptable GUI agents, potentially streamlining cross-platform automation and user interaction.
RANK_REASON The cluster contains an academic paper detailing a new method and dataset for AI research.
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →
