Intermediate-Task Training Language Diversity and Inference Throughput Gains in Quantized Cross-Lingual NLP 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 correlation between intermediate-task training language diversity and inference throughput gains after applying 8-bit quantization on cross-lingual natural language understanding benchmarks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.8/10.