Downloads · 30 days
114
17% of all-time downloads
klue-roberta-large-klueSTS-cross
leewaay/klue-roberta-large-klueSTS-cross
klue-roberta-large-klueSTS-cross is a sentence similarity model from leewaay. Use it when you need a score for how close two texts are. It is set up for sentence-transformers.
This is a sentence-transformers model: A cross encoder trained with the pretrained klue/roberta-large model on the KLUE STS dataset for sentence similarity tasks. It's specifically designed for direct evaluation of se…
All-time downloads
670
Public
Parameters
337M
2.7 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin1.3 GB · 50%
Parameter types
How the weights are stored.
F32337M · 100%
Model card
From the Hugging Face model README
leewaay/klue-roberta-large-klueSTS-cross
This is a sentence-transformers model: A cross encoder trained with the pretrained klue/roberta-large model on the KLUE STS dataset for sentence similarity tasks. It's specifically designed for direct evaluation of sentence pairs, making it highly effective for Re-Ranking and Augmented SBERT for data labeling tasks aimed at enhancing SBERT.
Usage (Sentence-Transformers)
Using this model becomes easy when you have sentence-transformers installed:
pip install -U sentence-transformers
Then you can use the model like this:
from sentence_transformers import CrossEncoder
pairs = [('Query', 'Paragraph1'), ('Query', 'Paragraph2') , ('Query', 'Paragraph3')]
model = CrossEncoder('leewaay/klue-roberta-large-klueSTS-cross')
scores = model.predict(pairs)
print(scores)
Citing & Authors
- Task
- Sentence Similarity
- Library
- sentence-transformers
- Type
- roberta
- Languages
- ko
- Created
- Mar 5, 2024
- Updated
- Mar 11, 2025