NUSTM/FacialMMT

1---2license: gpl-3.03tags:4  - emotion5  - emotion-recognition6  - sentiment-analysis7  - roberta8language:9  - en10pipeline_tag: text-classification11---12 13## FacialMMT14 15This repo contains the data and pretrained models for FacialMMT, a framework that uses facial sequences of real speaker to help multimodal emotion recognition.16 17The model performance on MELD test set is:18 19| Release | W-F1(%) |20|:-------------:|:--------------:|21| 23-07-10 | 66.73 |22 23It is currently ranked third on [paperswithcode](https://paperswithcode.com/sota/emotion-recognition-in-conversation-on-meld?p=a-facial-expression-aware-multimodal-multi).24 25If you're interested, please check out this [repo](https://github.com/NUSTM/FacialMMT) for more in-detail explanation of how to use our model.26 27 28### Citation29 30Please consider citing the following if this repo is helpful to your research.31```32@inproceedings{zheng2023facial,33  title={A Facial Expression-Aware Multimodal Multi-task Learning Framework for Emotion Recognition in Multi-party Conversations},34  author={Zheng, Wenjie and Yu, Jianfei and Xia, Rui and Wang, Shijin},35  booktitle={Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},36  pages={15445--15459},37  year={2023}38}39```