# CMMA: Benchmarking Multi-Affection Detection inChinese Multi-Modal Conversations
We hope that this dataset counld benchmark Chinese sentiment, emotion, sarcasm and humor analysis and benefit further studies.
# Our work
The visual and audio dataset will be released on https://drive.google.com/file/d/19NgqpYLPa3bLFm4YUjuCT2sbmQemziG-/view?usp=sharing
The link to our paper is https://proceedings.neurips.cc/paper_files/paper/2023/hash/3be60b4a739b95a07a944a1a2c41e05e-Abstract-Datasets_and_Benchmarks.html