VMLU
Click to visit website
About
VMLU is a human-centric benchmark suite tailored to evaluate the performance of foundation models, with a specific emphasis on tasks involving the Vietnamese language. Comprising 10,880 multiple-choice questions across 58 subjects, the VMLU categorizes knowledge into four domains: STEM, Humanities, Social Sciences, and others. Each subject includes around 200 questions distributed across various difficulty levels, ranging from elementary to professional expertise. The dataset is sourced from educational examinations and is organized by the Ministry of Education and Training, covering a broad spectrum of areas like mathematics, history, literature, and law. Users have access to a dataset download, alongside extensive GitHub resources that present usage instructions, evaluation metrics, and replication codes. VMLU facilitates a comprehensive assessment of general knowledge and complex problem-solving capabilities of AI models.
Platform
Features
• 10,880 multiple-choice questions
• 58 distinct subjects
• four main domains: stem, humanities, social sciences, others
• questions span from elementary to advanced levels
• downloadable dataset and extensive github resources
FAQs
What is VMLU?
VMLU is a benchmark suite designed for assessing language models, specifically focusing on the Vietnamese language.
How can I download the dataset?
You can download the VMLU dataset directly from their website by clicking the provided download button.
What subjects does VMLU cover?
VMLU covers 58 subjects spanning STEM, Humanities, Social Sciences, and a broad category labeled 'Others'.
Who can I contact for collaboration?
For collaboration inquiries, you can contact developer@vmlu.ai.
Average Rating: 0.0
Average Rating: 0.0
5 Stars:
0 Ratings
4 Stars:
0 Ratings
3 Stars:
0 Ratings
2 Stars:
0 Ratings
1 Star:
0 Ratings
User Ratings
No ratings available.
Sign In to Rate this Tool
Featured Tools
Dezyn
Interactive architectural diagram tool with AI-powered features for flowcharts and cloud architectures.
View DetailsChoice AI
Personalized OTT entertainment platform using AI for tailored viewing experiences.
View Details