In this work, we propose a novel framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries only ...

Oct 12, 2022 · Cross-lingual word embeddings (CLWE) provide a shared representation space for knowledge transfer between languages, yielding state-of-the-art ...

Mar 11, 2021 · In this work, we propose a novel framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries ...

In this work, we propose a novel framework to align contextual embeddings at the sense level by leverag- ing cross-lingual signal from bilingual dictionaries ...

This work proposes a novel framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries only ...

This is the source code of our method proposed in paper "Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings" accepted by COLING 2022.

In this work, we propose a novel framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries only ...

In this work, we propose a novel framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries only ...

Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings ... Multi-level Community-awareness Graph Neural Networks for Neural Machine Translation.

Contextual representation of the same token are clustered together. • The average distance between tokens is larger than within each token.