Dataset for SemEval-2024 Task 3

The dataset for SemEval-2024 Task 3: The Competition of Multimodal Emotion Cause Analysis in Conversations is released here.

File Description

The text folder contains the files of training data for each subtask, where all instances are stacked into a list and each of them is stored in a dictionary.
The video_with_audio folder contains the video files required for Subtask 2. Note: The three compressed packages correspond to the training, validation, and testing sets of ECF respectively. All of them are used as the training set in this SemEval evaluation.

SemEval-2024_Task3
├── text
│   ├── Subtask_1_train.json
│   ├── Subtask_2_train.json
├── video_with_audio
│   ├── train_1.tar.gz
│   │   ├── dia1utt1.mp4
│   │   ├── dia1utt2.mp4
│   │   ├── ...
│   ├── train_2.tar.gz
│   ├── train_3.tar.gz
├── README.md

Dataset Statistics

# of Instances (Training Data)	# of Instances (Evaluation data)
1,374 conversations 13,619 utterances	To be updated.