Dataset · KorQuAD
Korean text question-answering dataset (KorQuAD)
KorQuAD/squad_kor_v1
10k to 100k Korean question-answer pairs for training machine reading systems, built to the SQuAD 1.0 standard.
What's inside
A large Korean text corpus of 10k to 100k examples for question-answering tasks. Developed following the methodology of the famous SQuAD v1.0 benchmark to test models' ability to comprehend read text and find precise answers within it.
What you can do with it
- Training and evaluating Korean document question-answering (QA) systems
- Evaluating machine reading comprehension skills
- Fine-tuning Korean encoders and language models
- Tasks
- Question answering
- Languages
- Korean
- Size
- 10K–100K rows
- Downloads this month
- 2.1K
- Likes
- 34
Models trained on it
- HANTAEK/klue-roberta-large-korquad-v1-qa-finetunedQuestion answering 21
- oceann315/klue-roberta-large-korquad-v1Question answering 18
- NakJun/Llama-3.2-1B-Instruct-korQuAD-v1 15
- snumin44/biencoder-ko-bert-contextEmbeddings 10
- snumin44/biencoder-ko-bert-questionEmbeddings 9
- jiwonkaya/Llama-3.2-1B-Instruct-korQuAD-v1-Q4_K_M-GGUF 7
- rjslas7411/Llama-3.2-1B-Instruct-korQuAD-v1-Q4_K_M-GGUF 6
- seungwon-sw/q_lora_korqaText generation 0
How to load it
from datasets import load_dataset
ds = load_dataset("KorQuAD/squad_kor_v1")