Code
DocMSU: A Comprehensive Benchmark for Document-Level Multimodal Sarcasm Understanding
Why Lacuna links this resource
Repository names this paper
# DocMSU: A Comprehensive Benchmark for Document-level Multimodal Sarcasm Understanding [](https://arxiv.org/abs/2312.16023) This repo is the official dataset and Pytorch implementation of DocMSU: A Comprehensive Benchmark for Document-level Multimodal Sarcasm Understanding [AAAI2024]. ## DocMSU ### Introduction In document-level news, sarcasm clues are sparse or small and are often concealed in long text. Mor