1 paper
Yong Xien Chng, Tao Hu, Wenwen Tong +10
While Vision-Language Models (VLMs) can solve complex tasks through agentic reasoning, their capabilities remain largely constrained to text-oriented chain-of-thought or isolated t…