Draw-and-Understand: Leveraging Visual Prompts to Enable MLLMs to Comprehend What You Want

Documenti analoghi

• LLM as Dataset Analyst: Subpopulation Structure Discovery with Large Language Model
  di: Luo, Yulin, et al.
  Pubblicazione: (2024)
• MoSA: Mixture of Sparse Adapters for Visual Efficient Tuning
  di: Zhang, Qizhe, et al.
  Pubblicazione: (2023)
• Agent Skills Should Go Beyond Text: The Case for Visual Skills
  di: Xu, Binxiao, et al.
  Pubblicazione: (2026)
• Rethinking VLM Representation for VLA Initialization
  di: Lin, Weifeng, et al.
  Pubblicazione: (2026)
• Vision Language Models See What You Want but not What You See
  di: Gao, Qingying, et al.
  Pubblicazione: (2024)