1 paper · 1 filter
Aneerban Saha, Vijay Kumar, Sameer Madhav D. +2
Some questions about an image need commonsense or factual knowledge that its pixels do not contain. Large Language Models (LLMs) can supply this knowledge to Visual Question Answer…