Typological Distance and Sequential Fine-Tuning Order for Cross-Lingual Transfer

Transfer learning has led to large gains in performance for nearly all NLP tasks while making downstream models easier and faster to train. This has also been extended to low-resourced languages, with some success. We investigate the properties of cross-lingual transfer learning between ten low-resourced languages, from the perspective of a named entity recognition task. We specifically investigate how much adaptive fine-tuning and the choice of transfer language affect zero-shot transfer performance. We find that models that perform well on a single language often do so at the expense of gene

Research goal: How does the typological distance between source and target languages influence the optimal order of sequential fine-tuning tasks for cross-lingual transfer on low-resource NLP benchmarks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.0/10.