Meta-Learned Layer Selection Scaling for Zero-Shot Cross-Lingual Generalization

Multilingual pre-trained contextual embedding models (Devlin et al., 2019) have achieved impressive performance on zero-shot cross-lingual transfer tasks. Finding the most effective fine-tuning strategy to fine-tune these models on high-resource languages so that it transfers well to the zero-shot languages is a non-trivial task. In this paper, we propose a novel meta-optimizer to soft-select which layers of the pre-trained model to freeze during fine-tuning. We train the meta-optimizer by simulating the zero-shot transfer scenario. Results on cross-lingual natural language inference show that

Research goal: What is the impact of scaling the number of intermediate tasks (e.g., 5 vs. 10) on the generalization capability of meta-learned layer selection for zero-shot cross-lingual transfer, as evaluated by F1 score consistency across languages in the XTREME-R benchmark?

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