VUA20 is (perhaps) the largest dataset of metaphor detection used in Figlang2020 workshop.
For the details of this dataset, we refer you to the release paper.
The annotation method of VUA20 is elabrated in the paper of MIP.
If you find this dataset helpful, please cite:
@inproceedings{Leong2020ARO… See the full description on the dataset page:
https://huggingface.co/datasets/CreativeLang/vua20_metaphor.
