Visual editing of input images as a chain of thought improves multimodal LLM accuracy on structured image tasks by 3 to 12 points.
Visual chain- of-thought prompting for knowledge-based visual reasoning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
contest 1representative citing papers
citing papers explorer
-
ReFocus: Visual Editing as a Chain of Thought for Structured Image Understanding
Visual editing of input images as a chain of thought improves multimodal LLM accuracy on structured image tasks by 3 to 12 points.