Publications

2022

BERT on a Data Diet: Finding Important Examples by Gradient-Based Pruning
Mohsen Fayyaz*, Ehsan Aghazadeh, Ali Modarressi, Mohammad Taher Pilehvar, Yadollah Yaghoobzadeh, Samira Ebrahimi Kahou
* Equal Contribution
ENLSP @ NeurIPS2022
[📝 paper]

GlobEnc: Quantifying Global Token Attribution by Incorporating the Whole Encoder Layer in Transformers
Ali Modarressi*, Mohsen Fayyaz*, Yadollah Yaghoobzadeh, Mohammad Taher Pilehvar
* Equal Contribution
NAACL 2022
[📝 paper] [🎥 video] [💻 code]

Metaphors in Pre-Trained Language Models: Probing and Generalization Across Datasets and Languages
Ehsan Aghazadeh*, Mohsen Fayyaz* and Yadollah Yaghoobzadeh
* Equal Contribution
ACL 2022
[📝 paper] [🎥 video] [🖼️ Poster] [💻 code]

2021

Not All Models Localize Linguistic Knowledge in the Same Place: A Layer-wise Probing on BERToids’ Representations
Mohsen Fayyaz*, Ehsan Aghazadeh*, Ali Modarressi, Hosein Mohebbi and Mohammad Taher Pilehvar
* Equal Contribution
BlackboxNLP @ EMNLP 2021
[📝 paper] [🌐 blog] [🖼️ Poster] [💻 code]