PixelEyes: Decoupling Perception and Reasoning for Pinpoint Visual Evidence Seeking

PixelEyes: Decoupling Perception and Reasoning for Pinpoint Visual Evidence Seeking

Multi-turn visual reasoning agents suffer from entangled reasoning and perception that cause redundant trajectories; PixelEyes addresses this by decoupling these processes through…

Hugging Face · Daily Papers ·Dengxian Gong, Yuanzheng Wu · ·▲ 7 upvotes

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

Autores: Dengxian Gong, Yuanzheng Wu, Haobo Yuan, Zhengdong Hu, Tao Zhang, Yikang Zhou

  • 7 upvotes da comunidade
  • Temas: multi-turn visual reasoning, MLLMs, reasoning and perception entanglement, mask-guided visual search, semantic-region breadth-first search, referring segmentation model

Resumo

Resumo original (em inglês), extraído do paper:

Multi-turn visual reasoning agents suffer from entangled reasoning and perception that cause redundant trajectories; PixelEyes addresses this by decoupling these processes through mask-guided search and semantic-region breadth-first search, demonstrated on a new benchmark with expert-resynthesized data.

Onde ler

compartilhar: