← Search

Ningyu Xu

2 accepted papers

2023

Are Structural Concepts Universal in Transformer Language Models? Towards Interpretable Cross-Lingual Generalization

EMNLP 2023long findings

Large language models (LLMs) have exhibited considerable cross-lingual generalization abilities, whereby they implicitly transfer knowledge across languages. However, the transfer is not equally successful for all languages, especially for low-resource ones, which poses an ongoing challenge. It is u…

Cited by 0SourcecodeScholar
2022

Cross-Linguistic Syntactic Difference in Multilingual BERT: How Good is It and How Does It Affect Transfer?

EMNLP 2022main

Multilingual BERT (mBERT) has demonstrated considerable cross-lingual syntactic ability, whereby it enables effective zero-shot cross-lingual transfer of syntactic knowledge. The transfer is more successful between some languages, but it is not well understood what leads to this variation and whethe…