# Empowering Cross-lingual Abilities of Instruction-tuned Large Language Models by Translation-following demonstrations

Source: [Empowering Cross-lingual Abilities of Instruction-tuned Large Language Models by Translation-following demonstrations](<https://www.sota2.com/research/paper/68fa756a1e209131cea67b53>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:41:59.710Z

Authors: Leonardo Ranaldi, Giulia Pucci, Andre Freitas

Published: 2023-08-27T19:22:12.000Z

## Abstract

The language ability of Large Language Models (LLMs) is often unbalanced towards English because of the imbalance in the distribution of the pre-training data. This disparity is demanded in further fine-tuning and affecting the cross-lingual abilities of LLMs. In this paper, we propose to empower Instructiontuned LLMs (It-LLMs) in languages other than English by building semantic alignment between them. Hence, we propose CrossAlpaca, an It-LLM with cross-lingual instruction-following and Translation-following demonstrations to improve semantic alignment between languages. We validate our approach on the multilingual Question Answering (QA) benchmarks XQUAD and MLQA and adapted versions of MMLU and BBH. Our models, tested over six different languages, outperform the It-LLMs tuned on monolingual data. The final results show that instruction tuning on non-English data is not enough and that semantic alignment can be further improved by Translation-following demonstrations.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2308.14186>)

[PDF](<https://arxiv.org/pdf/2308.14186.pdf>)
