GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots

GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots

GUICrafter addresses GUI agent data challenges through a weakly-supervised approach using unannotated screenshots and a two-stage curriculum learning framework for visual grounding…

Hugging Face · Daily Papers ·Sunqi Fan, Lingshan Chen ·

Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.

Autores: Sunqi Fan, Lingshan Chen, Runqi Yin, Qingle Liu, Yongming Rao, Meng-Hao Guo

  • 0 upvotes da comunidade
  • Temas: GUI agents, weakly-supervised learning, curriculum learning, visual grounding, reinforcement learning, GUI interaction

Resumo

Resumo original (em inglês), extraído do paper:

GUICrafter addresses GUI agent data challenges through a weakly-supervised approach using unannotated screenshots and a two-stage curriculum learning framework for visual grounding and reinforcement learning calibration.

Onde ler

compartilhar: