FLORES-200

The 'FLORES-200' dataset is a multilingual benchmark used to evaluate the performance of language models across 200 language directions, including their ability to handle low-resource languages like Sardinian.

Papers using FLORES-200 (5)

LLiMba: Sardinian on a Single GPU -- Adapting a 3B Language Model to a Vanishing Romance Language2026Dictionary Insertion Prompting for Multilingual Reasoning on Multilingual Large Language Models2024X-ALMA: Plug & Play Modules and Adaptive Rejection for Quality Translation at Scale2024Refining Translations with LLMs: A Constraint-Aware Iterative Prompting Approach2024Marco-LLM: Bridging Languages via Massive Multilingual Training for Cross-Lingual Enhancement2024