MMLU-Pro
by TIGER-Lab (organization)
MMLU-Pro dataset is a more robust and challenging massive multi-task understanding dataset tailored to more rigorously benchmark large language models' capabilities. This dataset contains 12K complex questions across various disciplines. |Github | 🏆Leaderboard | 📖Paper | 🚀… (excerpt from the dataset card on Hugging Face)
Dataset facts
- Hub id
- TIGER-Lab/MMLU-Pro
- Likes
- 577
- Downloads (last 30 days)
- 239,436
- Tasks
- Question answering
- Modality
- tabular, text
- Size
- 10K to 100K examples
- License
- MIT (mit)
- Access
- Open download
- Author
- TIGER-Lab · Organization
- Languages
- en
- Created
- May 8, 2024
- Last modified
- May 2, 2026
More datasets from TIGER-Lab
Rank 185: TIGER-Lab/MathInstruct
🦣 MAmmoTH: Building Math Generalist Models through Hybrid Instruction Tuning MathInstruct is a meticulously curated instruction tuning dataset that is lightweight yet g…
316 likes12,385 downloads in the last 30 daysText generationMIT
Rank 443: TIGER-Lab/WebInstructSub
🦣 MAmmoTH2: Scaling Instructions from the Web Project Page: Paper: Code: WebInstruct (Subset) This repo contains the partial dataset used in "MAmmoTH2: Scaling Instruct…
170 likes5,160 downloads in the last 30 daysQuestion answeringApache 2.0
Rank 562: TIGER-Lab/OmniEdit-Filtered-1.2M
In this paper, we present OMNI-EDIT, which is an omnipotent editor to handle seven different image editing tasks with any aspect ratio seamlessly. Our contribution is in…
136 likes59,750 downloads in the last 30 daysMIT