Zhejiang University develops AI for robots to 'think with their eyes'

Researchers from Zhejiang University, in partnership with Cornell University, the National University of Singapore, and Xidian University, have developed a novel visual reasoning system for robots. This system, named VisualThink-VLA, allows robots to process visual information directly, bypassing the need for language-based internal thought processes. The new system demonstrates a significant speed enhancement, being 22.8 times faster than traditional text-based reasoning methods, while also achieving superior accuracy. AI

IMPACT Enables robots to process visual information directly, potentially speeding up complex tasks and improving their interaction with the physical world.

RANK_REASON The cluster describes a novel research system developed by universities, not a commercial product release or frontier model. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Pandaily →

• Cornell University
• National University of Singapore
• VisualThink-VLA
• Xidian University
• Zhejiang University

• model release
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →