Culturally aware natural language inference

J Huang, D Yang - Findings of the Association for Computational …, 2023 - aclanthology.org
Humans produce and consume language in a particular cultural context, which includes
knowledge about specific norms and practices. A listener's awareness of the cultural context …

Latxa: An open language model and evaluation suite for Basque

J Etxaniz, O Sainz, N Miguel, I Aldabe… - Proceedings of the …, 2024 - aclanthology.org
We introduce Latxa, a family of large language models for Basque ranging from 7 to 70
billion parameters. Latxa is based on Llama 2, which we continue pretraining on a new …

MKQA: A linguistically diverse benchmark for multilingual open domain question answering

S Longpre, Y Lu, J Daiber - Transactions of the Association for …, 2021 - direct.mit.edu
Progress in cross-lingual modeling depends on challenging, realistic, and diverse
evaluation sets. We introduce Multilingual Knowledge Questions and Answers (MKQA), an …

Lost in translation: large language models in non-English content analysis

G Nicholas, A Bhatia - arxiv preprint arxiv:2306.07377, 2023 - arxiv.org
In recent years, large language models (eg, Open AI's GPT-4, Meta's LLaMa, Google's
PaLM) have become the dominant approach for building AI systems to analyze and …

Do multilingual language models think better in English?

J Etxaniz, G Azkune, A Soroa, OL de Lacalle… - arxiv preprint arxiv …, 2023 - arxiv.org
Translate-test is a popular technique to improve the performance of multilingual language
models. This approach works by translating the input into English using an external machine …

Detecting and mitigating hallucinations in multilingual summarisation

Y Qiu, Y Ziser, A Korhonen, EM Ponti… - arxiv preprint arxiv …, 2023 - arxiv.org
Hallucinations pose a significant challenge to the reliability of neural models for abstractive
summarisation. While automatically generated summaries may be fluent, they often lack …

Few-shot learning with multilingual language models

XV Lin, T Mihaylov, M Artetxe, T Wang, S Chen… - arxiv preprint arxiv …, 2021 - arxiv.org
Large-scale generative language models such as GPT-3 are competitive few-shot learners.
While these models are known to be able to jointly represent many different languages, their …