Impact of Intermediate Task Diversity on Zero-Shot Cross-Lingual Transfer Performance in Multilingual Models

Engineering notes

Engineering notes will be added by the aipentium editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为大语言模型、生成式AI、ChatGPT相关技术、计算机视觉、深度学习等高价值论文补充中文说明。

Original abstract

Intermediate-task training---fine-tuning a pretrained model on an intermediate task before fine-tuning again on the target task---often improves model performance substantially on language understanding tasks in monolingual English settings. We investigate whether English intermediate-task training is still helpful on non-English target tasks. Using nine intermediate language-understanding tasks, we evaluate intermediate-task transfer in a zero-shot cross-lingual setting on the XTREME benchmark. We see large improvements from intermediate training on the BUCC and Tatoeba sentence retrieval tas Research goal: What is the impact of intermediate task diversity (e.g., monolingual vs. multilingual tasks) on the zero-shot cross-lingual transfer performance of models like mBERT or XLM-R on XTREME-R, measured by average accuracy and per-task F1 scores? Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.0/10.

Links and sources

Comments