Exploring image–text combinations in visual humour through large language models (LLMs) | Publicación