Feature Extraction
sentence-transformers
Safetensors
xlm-roberta
sentence-similarity
Generated from Trainer
dataset_size:1879136
loss:CachedGISTEmbedLoss
text-embeddings-inference
Instructions to use nlpai-lab/KURE-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use nlpai-lab/KURE-v1 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("nlpai-lab/KURE-v1") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Inference
- Notebooks
- Google Colab
- Kaggle
CachedGISTEmbedLoss 이면 가이드 모델은 무엇을 사용 하셨나요?
#1
by sigridjineth - opened
제곧내
bge-m3 사용했습니다.
yjoonjang changed discussion status to closed
KoE5에서는 CachedMNRLoss를 사용하시고 KURE-v1에서는 CachedGISTEmbedLoss를 사용하신 이유가 있으신가요?
안녕하세요, 같은 모델로, CachedGISTEmbedLoss를 사용한 것이 CachedMNRL과 비교하였을 때보다 벤치마크상 훨씬 높은 성능을 기록하였습니다. 아마 배치 내 false negatives를 잘 필터링해준 것으로 사료됩니다.