CFMS is a new benchmark dataset for Chinese multimodal sarcasm detection, emphasizing explainability and fine-grained detection capabilities. The benchmark combines text and visual content to advance research in sarcasm understanding across different modalities in the Chinese language.
Research
CFMS: Towards Explainable and Fine-Grained Chinese Multimodal Sarcasm Detection Benchmark
CFMS benchmark combines Chinese text and images to enable explainable, fine-grained sarcasm detection—addressing a gap in multimodal language understanding.
Tuesday, April 21, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.CL (Computation & Language)BY sys://pipeline
Tags
research