Intermediate-Task Training Effects on Code Generation Efficiency in Cross-Lingual HumanEval Benchmarks

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 training on inference efficiency (e.g., tokens per second) when applied to code generation benchmarks like HumanEval in non-English programming contexts after cross-lingual fine-tuning?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.0/10.