Dataset · BAAI
BAAI industry instruction dataset across 12 domains
BAAI/IndustryInstruction
Millions of instruction and QA pairs in Chinese and English for domain-specific model fine-tuning.
What's inside
A large multimodal corpus ranging from 1 to 10 million records, containing text, tables, and images in Chinese and English. The data is designed for question-answering tasks and gathered from high-quality industry materials of the BAAI/IndustryCorpus2 corpus. It spans 12 fields, ranging from medicine, law, finance, and AI to transportation, education, and tourism.
What you can do with it
- SFT fine-tuning LLMs for specialized industries (medicine, finance, law, automotive)
- Building specialized corporate assistants in Chinese and English languages
- Evaluating the response quality of models in specific subject domains
- Tasks
- Question answering
- Languages
- Chinese, English
- Size
- 1M–10M rows
- Downloads this month
- 408
- Likes
- 34
- Access
- on request on Hugging Face
Models trained on it
- mradermacher/Artificial-llama3_1_8B_instruct-GGUF 856
- mradermacher/Hospitality-llama3_1_8B_instruct-GGUF 287
- Anoxiom/Artificial-llama3_1_8B_instruct-Q6_K-GGUF 15
- Anoxiom/Tranport-llama3_1_8B_instruct-Q6_K-GGUF 9
- Anoxiom/Finance-llama3_1_8B_instruct-Q6_K-GGUF 7
- XiaofengAlg/Aerospace-llama3_1_8B_instruct 5
- XiaofengAlg/Transport-llama3_1_8B_instruct 4
- XiaofengAlg/CareBot_Medical_multi-llama3-8b-base 2
How to load it
from datasets import load_dataset
ds = load_dataset("BAAI/IndustryInstruction")