M2Lingual: Enhancing Multilingual, Multi-Turn Instruction Alignment in Large Language Models

Instruction finetuning (IFT) is critical for aligning Large Language Models (LLMs) to follow instructions. While many effective IFT datasets have been introduced recently, they predominantly focus on high-resource languages like English. To better align LLMs across a broad spectrum of languages and tasks, we propose a fully synthetic, novel taxonomy (Evol) guided Multilingual, Multi-turn instruction finetuning dataset, called M2Lingual. It is constructed by first selecting a diverse set of seed examples and then utilizing the proposed Evol taxonomy to convert these seeds into complex and challenging multi-turn instructions. We demonstrate the effectiveness of M2Lingual by training LLMs of varying sizes and showcasing the enhanced performance across a diverse set of languages. We contribute the 2 step Evol taxonomy with the guided generation code:this https URL, as well as the first fully synthetic, general and task-oriented, multi-turn, multilingual dataset built with Evol - M2Lingual:this https URLM2Lingual - containing 182K total IFT pairs, covering 70 languages and 17+ NLP tasks.

@article{maheshwary2025_2406.16783,
  title={ M2Lingual: Enhancing Multilingual, Multi-Turn Instruction Alignment in Large Language Models },
  author={ Rishabh Maheshwary and Vikas Yadav and Hoang Nguyen and Khyati Mahajan and Sathwik Tejaswi Madhusudhan },
  journal={arXiv preprint arXiv:2406.16783},
  year={ 2025 }
}