One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding

One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding

InnerZoom addresses GUI grounding challenges by preserving target-region awareness across decoder layers through a single-forward pass that bridges cross-layer evidence, achieving…

Hugging Face · Daily Papers ·Chen Liu, Ling Chen · ·▲ 3 upvotes

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

Autores: Chen Liu, Ling Chen, Hanzhang Zhou, Liangyu Chen, Chenglin Cai, Xin Yu

  • 3 upvotes da comunidade
  • Temas: MLLM-based GUI grounding, autoregressive coordinate generation, spatial precision, decoder layers, cross-layer evidence bridging, ZoomIn-style methods

Resumo

Resumo original (em inglês), extraído do paper:

InnerZoom addresses GUI grounding challenges by preserving target-region awareness across decoder layers through a single-forward pass that bridges cross-layer evidence, achieving state-of-the-art performance with reduced computational cost.

Onde ler

compartilhar: