1 paper
Qinhan Yu, Jun Guang, Chong Chen +1
In knowledge-intensive scenarios, providing reliable interleaved text-image responses requires Verified Image Grounding (VIG): the precise integration of retrieved authentic visual…