Unveiling the Response of Large Vision-Language Models to Visually Absent Tokens.

Sohee Kim, Soohyun Ryu, Joonhyung Park, Eunho Yang

Browse the full EMNLP paper archive.