Cross-lingual model agreement accuracy with artificial code-switching training

Transferring information retrieval (IR) models from a high-resource language (typically English) to other languages in a zero-shot fashion has become a widely adopted approach. In this work, we show that the effectiveness of zero-shot rankers diminishes when queries and documents are present in different languages. Motivated by this, we propose to train ranking models on artificially code-switched data instead, which we generate by utilizing bilingual lexicons. To this end, we experiment with lexicons induced from (1) cross-lingual word embeddings and (2) parallel Wikipedia page titles. We use

Research goal: How does training on artificially code-switched data affect the grammatical agreement accuracy of cross-lingual language models on the BliMP benchmark compared to monolingual fine-tuning?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.6/10.