Comparison of SALT with mBERT and XLM-R in Zero-Shot Cross-Lingual Transfer on Low-Resource Languages

Existing zero-shot cross-lingual transfer methods rely on parallel corpora or bilingual dictionaries, which are expensive and impractical for low-resource languages. To disengage from these dependencies, researchers have explored training multilingual models on English-only resources and transferring them to low-resource languages. However, its effect is limited by the gap between embedding clusters of different languages. To address this issue, we propose Embedding-Push, Attention-Pull, and Robust targets to transfer English embeddings to virtual multilingual embeddings without semantic loss,

Research goal: How does the SALT method compare to state-of-the-art multilingual models like mBERT and XLM-R on XTREME-R benchmarks when evaluated for zero-shot cross-lingual transfer on low-resource languages?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.9/10.