How Well Do Models Follow Visual Instructions? VIBE: A Systematic Benchmark for Visual Instruction-Driven Image Editing
Paper
•
2602.01851
•
Published
•
16
Dingshuo Chen
Dingshuo
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
4 months ago
Latent Sketchpad: Sketching Visual Thoughts to Elicit Multimodal
Reasoning in MLLMs
Organizations
None yet
upvoted
a
paper
11 days ago