Performance of CodeT5 in Zero-Shot Cross-Lingual Code Generation with Intermediate-Task Fine-Tuning Strategies

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: How does the performance of CodeT5 on zero-shot cross-lingual code generation tasks in the CodeXGLUE benchmark compare when using different intermediate-task fine-tuning strategies, such as task-specific vs. multi-task fine-tuning, measured by BLEU score and execution accuracy?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.3/10.