Anupama Ray

Multimodal LLMs fine-tuned with new datasets might be the next step to advance sarcasm detection and understanding. However, current methods using multimodality and context are still better than single LLMs looking only at the textual inputs.