Dataset · shibing624
Chinese Semantic Similarity Datasets
shibing624/nli_zh
100k to 1M pairs of Chinese sentences with labels for training logical inference and deduplication models.
What's inside
Contains clean Chinese text data in triplet format (sentence1, sentence2, label). Combines 5 popular semantic matching sets: ATEC, BQ, LCQMC, PAWSX, and STS-B. Total collection size ranges from 100,000 to 1,000,000 examples.
What you can do with it
- Training embeddings and rerankers for Chinese search
- Paraphrase classification and semantic text similarity detection
- Testing models on natural language inference (NLI) tasks
- Tasks
- Text classification
- Languages
- Chinese
- Size
- 100K–1M rows
- Downloads this month
- 3.8K
- Likes
- 49
Models trained on it
- shibing624/text2vec-base-chineseSentence similarity 721.6K
- oldhu/text2vec-lert-base-5-tasksSentence similarity 103
- 1289dfaj/text2vec-base-chineseSentence similarity 53
- ChenZ95/text2vec-base-chineseSentence similarity 52
- dengdeng4/text2vec-base-chineseSentence similarity 51
- Baicbc666/text2vec-base-chineseSentence similarity 49
- wxm777/text2vec-base-chineseSentence similarity 44
- EricLee/text2vec-roberta-512Sentence similarity 38
How to load it
from datasets import load_dataset
ds = load_dataset("shibing624/nli_zh")