microsoft/MMLU-CF
A Contamination-free Multi-task Language Understanding Benchmark [Official, ACL 2025]
This benchmark helps AI researchers and developers accurately measure the multi-task language understanding capabilities of large language models (LLMs). It provides a comprehensive set of multiple-choice questions across various academic disciplines. By using this benchmark, you can reliably compare how well different LLMs perform on complex reasoning tasks, free from data leakage issues that plague older benchmarks.
123 stars. No commits in the last 6 months.
Use this if you need a reliable and contamination-free benchmark to evaluate and compare the general knowledge and reasoning abilities of different large language models.
Not ideal if you are looking for a benchmark focused on very niche, highly specialized tasks or if you are not an AI developer or researcher working with LLM evaluation.
Stars
123
Forks
3
Language
—
License
MIT
Category
Last pushed
May 17, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/microsoft/MMLU-CF"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
MadryLab/context-cite
Attribute (or cite) statements generated by LLMs back to in-context information.
microsoft/augmented-interpretable-models
Interpretable and efficient predictors using pre-trained language models. Scikit-learn compatible.
Trustworthy-ML-Lab/CB-LLMs
[ICLR 25] A novel framework for building intrinsically interpretable LLMs with...
poloclub/LLM-Attributor
LLM Attributor: Attribute LLM's Generated Text to Training Data
THUDM/LongCite
LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA