Cross-lingual Knowledge Transfer under Data Constraints via Lexical Interventions

Apple Machine Learning Research investigates cross-lingual knowledge transfer as a critical approach for building high-performing multilingual language models when target language data is limited. The research addresses the challenge of transferring knowledge required for downstream tasks involving scientific reasoning, commonsense inference, and world knowledge when target language data is scarce, with specific focus on lexical interventions as a solution framework.

Read the original source at Apple Machine Learning Research ↗