← Search

Ehsan Aghazadeh

2 accepted papers

2023

DecompX: Explaining Transformers Decisions by Propagating Token Decomposition

ACL 2023long

An emerging solution for explaining Transformer-based models is to use vector-based analysis on how the representations are formed. However, providing a faithful vector-based explanation for a multi-layer model could be challenging in three aspects: (1) Incorporating all components into the analysis…

2022

Metaphors in Pre-Trained Language Models: Probing and Generalization Across Datasets and Languages

ACL 2022long

Human languages are full of metaphorical expressions. Metaphors help people understand the world by connecting new concepts and domains to more familiar ones. Large pre-trained language models (PLMs) are therefore assumed to encode metaphorical knowledge useful for NLP systems. In this paper, we inv…