Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

GlobEnc: Quantifying Global Token Attribution by Incorporating the Whole Encoder Layer in Transformers


May 06, 2022
Ali Modarressi, Mohsen Fayyaz, Yadollah Yaghoobzadeh, Mohammad Taher Pilehvar

* Accepted to NAACL 2022 (main conference) 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

On the Importance of Data Size in Probing Fine-tuned Models


Mar 17, 2022
Houman Mehrafarin, Sara Rajaee, Mohammad Taher Pilehvar

* To appear in ACL Findings 2022 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

AdapLeR: Speeding up Inference by Adaptive Length Reduction


Mar 16, 2022
Ali Modarressi, Hosein Mohebbi, Mohammad Taher Pilehvar

* Accepted to ACL 2022 (main conference) 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

An Isotropy Analysis in the Multilingual BERT Embedding Space


Oct 09, 2021
Sara Rajaee, Mohammad Taher Pilehvar


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Not All Models Localize Linguistic Knowledge in the Same Place: A Layer-wise Probing on BERToids' Representations


Sep 15, 2021
Mohsen Fayyaz, Ehsan Aghazadeh, Ali Modarressi, Hosein Mohebbi, Mohammad Taher Pilehvar

* Accepted to BlackboxNLP Workshop at EMNLP 2021 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

How Does Fine-tuning Affect the Geometry of Embedding Space: A Case Study on Isotropy


Sep 10, 2021
Sara Rajaee, Mohammad Taher Pilehvar

* To appear in Findings of EMNLP 2021 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Don't Discard All the Biased Instances: Investigating a Core Assumption in Dataset Bias Mitigation Techniques


Sep 01, 2021
Hossein Amirkhani, Mohammad Taher Pilehvar

* Findings of EMNLP, 2021 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

A Cluster-based Approach for Improving Isotropy in Contextual Embedding Space


Jun 02, 2021
Sara Rajaee, Mohammad Taher Pilehvar

* To appear in ACL 2021 main conference 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Exploring the Role of BERT Token Representations to Explain Sentence Probing Results


Apr 03, 2021
Hosein Mohebbi, Ali Modarressi, Mohammad Taher Pilehvar


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

XL-WiC: A Multilingual Benchmark for Evaluating Semantic Contextualization


Oct 13, 2020
Alessandro Raganato, Tommaso Pasini, Jose Camacho-Collados, Mohammad Taher Pilehvar

* EMNLP2020 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email
1
2
3
>>