I Came, I Saw, I Explained: Benchmarking Multimodal LLMs on Figurative Meaning in Memes
This work addresses the problem of understanding how multimodal models interpret complex online communication for researchers and developers, but it is incremental as it benchmarks existing models without proposing new methods.
The study evaluated eight state-of-the-art multimodal large language models on their ability to detect and explain figurative meaning in memes, finding that all models exhibited a strong bias toward associating memes with figurative meaning even when none was present, and correct predictions often lacked faithful explanations.
Internet memes represent a popular form of multimodal online communication and often use figurative elements to convey layered meaning through the combination of text and images. However, it remains largely unclear how multimodal large language models (MLLMs) combine and interpret visual and textual information to identify figurative meaning in memes. To address this gap, we evaluate eight state-of-the-art generative MLLMs across three datasets on their ability to detect and explain six types of figurative meaning. In addition, we conduct a human evaluation of the explanations generated by these MLLMs, assessing whether the provided reasoning supports the predicted label and whether it remains faithful to the original meme content. Our findings indicate that all models exhibit a strong bias to associate a meme with figurative meaning, even when no such meaning is present. Qualitative analysis further shows that correct predictions are not always accompanied by faithful explanations.