Diverse Intermediate Task Training for Zero-Shot Cross-Lingual Transfer in XTREME-R

Because it is not feasible to collect training data for every language, there is a growing interest in cross-lingual transfer learning. In this paper, we systematically explore zero-shot cross-lingual transfer learning on reading comprehension tasks with a language representation model pre-trained on multi-lingual corpus. The experimental results show that with pre-trained language representation zero-shot learning is feasible, and translating the source data into the target language is not necessary and even degrades the performance. We further explore what does the model learn in zero-shot s

Research goal: Does increasing the diversity of intermediate tasks (e.g., combining reasoning, commonsense, and reading comprehension) further improve zero-shot cross-lingual transfer performance on XTREME-R compared to single-task training, measured by accuracy on HellaSwag and RACE?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10.