僅需少量多語性的調整即可進行多語言指導微調

摘要

隨著說明調整的大型語言模型（LLMs）在全球範圍內得到廣泛應用，其能夠遵循多種語言的指示的能力變得日益重要。一種有前途的方法是跨語言轉移，其中模型通過在另一種語言上微調來獲得在某種語言上的特定功能。在這項工作中，我們研究了多語言LLM在指示調整期間如何影響跨語言指示遵循。我們首先展示，許多語言從單語調整中甚至可以將一些指示遵循能力轉移到其他語言。此外，我們發現，在英文調整集中僅有40個多語言示例就顯著改善了跨語言指示遵循，無論是在調整期間看到的還是未看到的語言。一般而言，我們觀察到，在多語言混合調整的模型在多種語言上表現出與單語調整模型相比相當或更優秀的性能，儘管在這些語言中僅使用了10倍少的示例進行訓練。最後，我們發現，在指示調整集中將語言數量從1增加到2、3或4，可以增加跨語言泛化能力。我們的結果表明，建立大規模多語言指示調整模型只需一個非常小的多語言指示-回應集即可實現。

English

As instruction-tuned large language models (LLMs) gain global adoption, their ability to follow instructions in multiple languages becomes increasingly crucial. One promising approach is cross-lingual transfer, where a model acquires specific functionality on some language by finetuning on another language. In this work, we investigate how multilinguality during instruction tuning of a multilingual LLM affects instruction-following across languages. We first show that many languages transfer some instruction-following capabilities to other languages from even monolingual tuning. Furthermore, we find that only 40 multilingual examples in an English tuning set substantially improve multilingual instruction-following, both in seen and unseen languages during tuning. In general, we observe that models tuned on multilingual mixtures exhibit comparable or superior performance in several languages compared to monolingually tuned models, despite training on 10x fewer examples in those languages. Finally, we find that increasing the number of languages in the instruction tuning set from 1 to only 2, 3, or 4 increases cross-lingual generalization. Our results suggest that building massively multilingual instruction-tuned models can be done with only a very small set of multilingual instruction-responses.

僅需少量多語性的調整即可進行多語言指導微調

Multilingual Instruction Tuning With Just a Pinch of Multilinguality

摘要

Support