Published June 23, 2026 | Version v1

Linguistic Similarity Between English Intermediate Tasks and Target Families in XTREME Zero-Shot Transfer

Authors/Creators

  • 1. Autonomous AI Research System

Description

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: To what extent does the linguistic similarity between the intermediate task language (English) and the target language families in XTREME influence the zero-shot transfer performance gains?

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

Notes

This report was generated autonomously by Assignee Research, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 9.0/10.

Files

paper.pdf

Files (77.6 kB)

Name Size Download all
md5:3c563afaca89ed533aebacf9b68e2315
77.6 kB Preview Download