Cross-lingual Transfer Performance in XTREME: Domain and Resource-Level Variability

While natural language processing systems often focus on a single language, multilingual transfer learning has the potential to improve performance, especially for low-resource languages. We introduce XLDA, cross-lingual data augmentation, a method that replaces a segment of the input text with its translation in another language. XLDA enhances performance of all 14 tested languages of the cross-lingual natural language inference (XNLI) benchmark. With improvements of up to \$4.8\%\$, training with XLDA achieves state-of-the-art performance for Greek, Turkish, and Urdu. XLDA is in contrast to, a

Research goal: Does the choice of intermediate task domain (e.g., natural language inference vs. question answering) influence zero-shot cross-lingual transfer performance in XTREME, and how does this vary across languages with differing resource levels?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.