Can I read Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning on EtoBox?

Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning by Singh, Shivalika; Vargus, Freddie; Dsouza, Daniel; Karlsson, Börje F.; Mahendiran, Abinaya; Ko, Wei-Yin; Shandilya, Herumb; Patel, Jay; Mataciunas, Deividas; OMahony, Laura; Zhang, Mike; Hettiarachchi, Ramith; Wilson, Joseph; Machado, Marina; Moura, Luisa Souza; Krzemiński, Dominik; Fadaei, Hakimeh; Ergün, Irem; Okoh, Ifeoma; Alaagib, Aisha; Mudannayake, Oshan; Alyafeai, Zaid; Chien, Vu Minh; Ruder, Sebastian; Guthikonda, Surya; Alghamdi, Emad A.; Gehrmann, Sebastian; Muennighoff, Niklas; Bartolo, Max; Kreutzer, Julia; Üstün, Ahmet; Fadaee, Marzieh; Hooker, Sara is a scholarly article available to read on EtoBox.

What is Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning about?

Datasets are foundational to many breakthroughs in modern artificial intelligence. Many recent achievements in the space of natural language processing (NLP) can be attributed to the finetuning of pre-trained models on a diverse set of tasks that enables a large language model (LLM) to respond to instructions. Instruction fine-tuning (IFT) requires specifically constructed and annotated datasets. However, existing datasets are almost all in the English language. In this work, our primary goal is to bridge the language gap by building a human-curated instruction-following dataset spanning 65 languages. We worked with fluent speakers of languages from around the world to collect natural instances of instructions and completions. Furthermore, we create the most extensive multilingual collection to date, comprising 513 million instances through templating and translating existing datasets across 114 languages. In total, we contribute four key resources: we develop and open-source the Aya Annotation Platform, the Aya Dataset, the Aya Collection, and the Aya Evaluation Suite. The Aya initiative also serves as a valuable case study in participatory research, involving collaborators from 1

Author

Singh, Shivalika; Vargus, Freddie; Dsouza, Daniel; Karlsson, Börje F.; Mahendiran, Abinaya; Ko, Wei-Yin; Shandilya, Herumb; Patel, Jay; Mataciunas, Deividas; OMahony, Laura; Zhang, Mike; Hettiarachchi, Ramith; Wilson, Joseph; Machado, Marina; Moura, Luisa Souza; Krzemiński, Dominik; Fadaei, Hakimeh; Ergün, Irem; Okoh, Ifeoma; Alaagib, Aisha; Mudannayake, Oshan; Alyafeai, Zaid; Chien, Vu Minh; Ruder, Sebastian; Guthikonda, Surya; Alghamdi, Emad A.; Gehrmann, Sebastian; Muennighoff, Niklas; Bartolo, Max; Kreutzer, Julia; Üstün, Ahmet; Fadaee, Marzieh; Hooker, Sara

Published

2024

Language

EN