Global-MMLU
Emerging5papers using it
2025first seen
Global-MMLU is a benchmark dataset designed to evaluate the performance of large language models across multiple languages, focusing on their ability to handle diverse tasks and mitigate performance disparities.
Papers using Global-MMLU (5)
- SARA: Unlocking Multilingual Knowledge in Mixture-of-Experts via Semantically Anchored Routing AlignmentCOMPASS: COntinual Multilingual PEFT with Adaptive Semantic SamplingMultilingual Refusal Alignment for Safer Large Language ModelsThe Art of Asking: Multilingual Prompt Optimization for Synthetic DataEfficient Model Development through Fine-tuning Transfer